# i

Lecture Notes on Optimization
Pravin Varaiya
ii
Contents
1 INTRODUCTION 1
2 OPTIMIZATION OVER AN OPEN SET 7
3 Optimization with equality constraints 15
4 Linear Programming 27
5 Nonlinear Programming 49
6 Discrete-time optimal control 75
7 Continuous-time linear optimal control 83
8 Coninuous-time optimal control 95
9 Dynamic programing 121
iii
iv CONTENTS
PREFACE to this edition
Notes on Optimization was published in 1971 as part of the Van Nostrand Reinhold Notes on Sys-
tem Sciences, edited by George L. Turin. Our aim was to publish short, accessible treatments of
graduate-level material in inexpensive books (the price of a book in the series was about ﬁve dol-
lars). The effort was successful for several years. Van Nostrand Reinhold was then purchased by a
conglomerate which cancelled Notes on System Sciences because it was not sufﬁciently proﬁtable.
Books have since become expensive. However, the World Wide Web has again made it possible to
publish cheaply.
Notes on Optimization has been out of print for 20 years. However, several people have been
using it as a text or as a reference in a course. They have urged me to re-publish it. The idea of
making it freely available over the Web was attractive because it reafﬁrmed the original aim. The
only obstacle was to retype the manuscript in LaTex. I thank Kate Klohe for doing just that.
I would appreciate knowing if you ﬁnd any mistakes in the book, or if you have suggestions for
(small) changes that would improve it.
Berkeley, California P.P. Varaiya
September, 1998
v
vi CONTENTS
PREFACE
These Notes were developed for a ten-week course I have taught for the past three years to ﬁrst-year
graduate students of the University of California at Berkeley. My objective has been to present,
in a compact and uniﬁed manner, the main concepts and techniques of mathematical programming
and optimal control to students having diverse technical backgrounds. A reasonable knowledge of
advanced calculus (up to the Implicit Function Theorem), linear algebra (linear independence, basis,
matrix inverse), and linear differential equations (transition matrix, adjoint solution) is sufﬁcient for
The treatment of the topics presented here is deep. Although the coverage is not encyclopedic,
an understanding of this material should enable the reader to follow much of the recent technical
literature on nonlinear programming, (deterministic) optimal control, and mathematical economics.
The examples and exercises given in the text form an integral part of the Notes and most readers will
need to attend to them before continuing further. To facilitate the use of these Notes as a textbook,
I have incurred the cost of some repetition in order to make almost all chapters self-contained.
However, Chapter V must be read before Chapter VI, and Chapter VII before Chapter VIII.
The selection of topics, as well as their presentation, has been inﬂuenced by many of my students
and colleagues, who have read and criticized earlier drafts. I would especially like to acknowledge
the help of Professors M. Athans, A. Cohen, C.A. Desoer, J-P. Jacob, E. Polak, and Mr. M. Ripper. I
also want to thank Mrs. Billie Vrtiak for her marvelous typing in spite of starting from a not terribly
legible handwritten manuscript. Finally, I want to thank Professor G.L. Turin for his encouraging
and patient editorship.
Berkeley, California P.P. Varaiya
November, 1971
vii
viii CONTENTS
Chapter 1
INTRODUCTION
In this chapter, we present our model of the optimal decision-making problem, illustrate decision-
making situations by a few examples, and brieﬂy introduce two more general models which we
cannot discuss further in these Notes.
1.1 The Optimal Decision Problem
These Notes show how to arrive at an optimal decision assuming that complete information is given.
The phrase complete information is given means that the following requirements are met:
1. The set of all permissible decisions is known, and
2. The cost of each decision is known.
When these conditions are satisﬁed, the decisions can be ranked according to whether they incur
greater or lesser cost. An optimal decision is then any decision which incurs the least cost among
the set of permissible decisions.
In order to model a decision-making situation in mathematical terms, certain further requirements
must be satisﬁed, namely,
1. The set of all decisions can be adequately represented as a subset of a vector space with each
vector representing a decision, and
2. The cost corresponding to these decisions is given by a real-valued function.
Some illustrations will help.
Example 1: The Pot Company (Potco) manufacturers a smoking blend called Acapulco Gold.
The blend is made up of tobacco and mary-john leaves. For legal reasons the fraction α of mary-
john in the mixture must satisfy 0 < α <
1
2
. From extensive market research Potco has determined
their expected volume of sales as a function of α and the selling price p. Furthermore, tobacco can
be purchased at a ﬁxed price, whereas the cost of mary-john is a function of the amount purchased.
If Potco wants to maximize its proﬁts, how much mary-john and tobacco should it purchase, and
what price p should it set?
Example 2: Tough University provides “quality” education to undergraduate and graduate stu-
dents. In an agreement signed with Tough’s undergraduates and graduates (TUGs), “quality” is
1
2 CHAPTER 1. INTRODUCTION
deﬁned as follows: every year, each u (undergraduate) must take eight courses, one of which is a
seminar and the rest of which are lecture courses, whereas each g (graduate) must take two seminars
and ﬁve lecture courses. A seminar cannot have more than 20 students and a lecture course cannot
have more than 40 students. The University has a faculty of 1000. The Weary Old Radicals (WORs)
have a contract with the University which stipulates that every junior faculty member (there are 750
of these) shall be required to teach six lecture courses and two seminars each year, whereas every
senior faculty member (there are 250 of these) shall teach three lecture courses and three seminars
each year. The Regents of Touch rate Tough’s President at α points per u and β points per g “pro-
cessed” by the University. Subject to the agreements with the TUGs and WORs how many u’s and
g’s should the President admit to maximize his rating?
Example 3: (See Figure 1.1.) An engineer is asked to construct a road (broken line) connection
point a to point b. The current proﬁle of the ground is given by the solid line. The only requirement
is that the ﬁnal road should not have a slope exceeding 0.001. If it costs \$c per cubic foot to excavate
or ﬁll the ground, how should he design the road to meet the speciﬁcations at minimum cost?
Example 4: Mr. Shell is the manager of an economy which produces one output, wine. There
are two factors of production, capital and labor. If K(t) and L(t) respectively are the capital stock
used and the labor employed at time t, then the rate of output of wine W(t) at time is given by the
production function
W(t) = F(K(t), L(t))
As Manager, Mr. Shell allocates some of the output rate W(t) to the consumption rate C(t), and
the remainder I(t) to investment in capital goods. (Obviously, W, C, I, and K are being measured
in a common currency.) Thus, W(t) = C(t) + I(t) = (1 − s(t))W(t) where s(t) = I(t)/W(t)

.
.
a
b
Figure 1.1: Admissable set of example.
∈ [0, 1] is the fraction of output which is saved and invested. Suppose that the capital stock decays
exponentially with time at a rate δ > 0, so that the net rate of growth of capital is given by the
following equation:
˙
K(t) =
d
dt
K(t) (1.1)
= −δK(t) +s(t)W(t)
= −δK(t) +s(t)F(K(t), L(t)).
The labor force is growing at a constant birth rate of β > 0. Hence,
1.1. THE OPTIMAL DECISION PROBLEM 3
˙
L(t) = βL(t).
(1.2)
Suppose that the production function F exhibits constant returns to scale, i.e., F(λK, λL) =
λF(K, L) for all λ > 0. If we deﬁne the relevant variable in terms of per capita of labor, w =
W/L, c = C/L, k = K/l, and if we let f(k) = F(k, l), then we see that F(K, L)−LF(K/L, 1) =
Lf(k), whence the consumption per capita of labor becomes c(t) = (l −s(t))f(k(t)). Using these
deﬁnitions and equations (1.1) and (1.2) it is easy to see that K(t) satisﬁes the differential equation
(1.3).
˙
k(t) = s(t)f(k(t)) −µk(t)
(1.3)
where µ = (δ +β). The ﬁrst term of the right-hand side in (3) is the increase in the capital-to-labor
ratio due to investment whereas the second terms is the decrease due to depreciation and increase in
the labor force.
Suppose there is a planning horizon time T, and at time 0 Mr. Shell starts with capital-to-labor
ratio k
o
. If “welfare” over the planning period [0, T] is identiﬁed with total consumption

T
0
c(t)dt,
what should Mr. Shell’s savings policy s(t), 0 ≤ t ≤ T, be so as to maximize welfare? What
savings policy maximizes welfare subject to the additional restriction that the capital-to-labor ratio
at time T should be at least k
T
? If future consumption is discounted at rate α > 0 and if time horizon
is ∞, the welfare function becomes

0
e

αt c(t)dt. What is the optimum policy corresponding to
this criterion?
These examples illustrate the kinds of decision-making problems which can be formulated math-
ematically so as to be amenable to solutions by the theory presented in these Notes. We must always
remember that a mathematical formulation is inevitably an abstraction and the gain in precision may
have occurred at a great loss of realism. For instance, Example 2 is caricature (see also a faintly re-
lated but more more elaborate formulation in Bruno [1970]), whereas Example 4 is light-years away
from reality. In the latter case, the value of the mathematical exercise is greater the more insensitive
are the optimum savings policies with respect to the simplifying assumptions of the mathematical
model. (In connection with this example and related models see the critique by Koopmans [1967].)
In the examples above, the set of permissible decisions is represented by the set of all points
in some vector space which satisfy certain constraints. Thus, in the ﬁrst example, a permissible
decision is any two-dimensional vector (α, p) satisfying the constraints 0 < α <
1
2
and 0 <
p. In the second example, any vector (u, g) with u ≥ 0, g ≥ 0, constrained by the number
of faculty and the agreements with the TUGs and WORs is a permissible decision. In the last
example, a permissible decision is any real-valued function s(t), 0 ≤ t ≤ T, constrained by
0 ≤ s(t) ≤ 1. (It is of mathematical but not conceptual interest to note that in this case a decision
is represented by a vector in a function space which is inﬁnite-dimensional.) More concisely then,
these Notes are concerned with optimizing (i.e. maximizing or minimizing) a real-valued function
over a vector space subject to constraints. The constraints themselves are presented in terms of
functional inequalities or equalities.
4 CHAPTER 1. INTRODUCTION
At this point, it is important to realize that the distinction between the function which is to be
optimized and the functions which describe the constraints, although convenient for presenting the
mathematical theory, may be quite artiﬁcial in practice. For instance, suppose we have to choose
the durations of various trafﬁc lights in a section of a city so as to achieve optimum trafﬁc ﬂow.
Let us suppose that we know the transportation needs of all the people in this section. Before we
can begin to suggest a design, we need a criterion to determine what is meant by “optimum trafﬁc
ﬂow.” More abstractly, we need a criterion by which we can compare different decisions, which in
this case are different patterns of trafﬁc-light durations. One way of doing this is to assign as cost to
each decision the total amount of time taken to make all the trips within this section. An alternative
and equally plausible goal may be to minimize the maximum waiting time (that is the total time
spent at stop lights) in each trip. Now it may happen that these two objective functions may be
inconsistent in the sense that they may give rise to different orderings of the permissible decisions.
Indeed, it may be the case that the optimum decision according to the ﬁrst criterion may be lead to
very long waiting times for a few trips, so that this decision is far from optimum according to the
second criterion. We can then redeﬁne the problem as minimizing the ﬁrst cost function (total time
for trips) subject to the constraint that the waiting time for any trip is less than some reasonable
bound (say one minute). In this way, the second goal (minimum waiting time) has been modiﬁed
and reintroduced as a constraint. This interchangeability of goal and constraints also appears at a
deeper level in much of the mathematical theory. We will see that in most of the results the objective
function and the functions describing the constraints are treated in the same manner.
1.2 Some Other Models of Decision Problems
Our model of a single decision-maker with complete information can be generalized along two
very important directions. In the ﬁrst place, the hypothesis of complete information can be relaxed
by allowing that decision-making occurs in an uncertain environment. In the second place, we
can replace the single decision-maker by a group of two or more agents whose collective decision
determines the outcome. Since we cannot study these more general models in these Notes, we
merely point out here some situations where such models arise naturally and give some references.
1.2.1 Optimization under uncertainty.
A person wants to invest \$1,000 in the stock market. He wants to maximize his capital gains, and
at the same time minimize the risk of losing his money. The two objectives are incompatible, since
the stock which is likely to have higher gains is also likely to involve greater risk. The situation
is different from our previous examples in that the outcome (future stock prices) is uncertain. It is
customary to model this uncertainty stochastically. Thus, the investor may assign probability 0.5 to
the event that the price of shares in Glamor company increases by \$100, probability 0.25 that the
price is unchanged, and probability 0.25 that it drops by \$100. A similar model is made for all the
other stocks that the investor is willing to consider, and a decision problem can be formulated as
follows. How should \$1,000 be invested so as to maximize the expected value of the capital gains
subject to the constraint that the probability of losing more than \$100 is less than 0.1?
As another example, consider the design of a controller for a chemical process where the decision
variable are temperature, input rates of various chemicals, etc. Usually there are impurities in the
chemicals and disturbances in the heating process which may be regarded as additional inputs of a
1.2. SOME OTHER MODELS OF DECISION PROBLEMS 5
random nature and modeled as stochastic processes. After this, just as in the case of the portfolio-
selection problem, we can formulate a decision problem in such a way as to take into account these
random disturbances.
If the uncertainties are modelled stochastically as in the example above, then in many cases
the techniques presented in these Notes can be usefully applied to the resulting optimal decision
problem. To do justice to these decision-making situations, however, it is necessary to give great
attention to the various ways in which the uncertainties can be modelled mathematically. We also
need to worry about ﬁnding equivalent but simpler formulations. For instance, it is of great signif-
icance to know that, given appropriate conditions, an optimal decision problem under uncertainty
is equivalent to another optimal decision problem under complete information. (This result, known
as the Certainty-Equivalence principle in economics has been extended and baptized the Separation
Theorem in the control literature. See Wonham [1968].) Unfortunately, to be able to deal with
these models, we need a good background in Statistics and Probability Theory besides the material
presented in these Notes. We can only refer the reader to the extensive literature on Statistical De-
cision Theory (Savage [1954], Blackwell and Girshick [1954]) and on Stochastic Optimal Control
(Meditch [1969], Kushner [1971]).
1.2.2 The case of more than one decision-maker.
Agent Alpha is chasing agent Beta. The place is a large circular ﬁeld. Alpha is driving a fast, heavy
car which does not maneuver easily, whereas Beta is riding a motor scooter, slow but with good
maneuverability. What should Alpha do to get as close to Beta as possible? What should Beta
do to stay out of Alpha’s reach? This situation is fundamentally different from those discussed so
far. Here there are two decision-makers with opposing objectives. Each agent does not know what
the other is planning to do, yet the effectiveness of his decision depends crucially upon the other’s
decision, so that optimality cannot be deﬁned as we did earlier. We need a new concept of rational
(optimal) decision-making. Situations such as these have been studied extensively and an elaborate
structure, known as the Theory of Games, exists which describes and prescribes behavior in these
situations. Although the practical impact of this theory is not great, it has proved to be among the
most fruitful sources of unifying analytical concepts in the social sciences, notably economics and
political science. The best single source for Game Theory is still Luce and Raiffa [1957], whereas
the mathematical content of the theory is concisely displayed in Owen [1968]. The control theorist
will probably be most interested in Isaacs [1965], and Blaquiere, et al., [1969].
The difﬁculty caused by the lack of knowledge of the actions of the other decision-making agents
arises even if all the agents have the same objective, since a particular decision taken by our agent
may be better or worse than another decision depending upon the (unknown) decisions taken by the
other agents. It is of crucial importance to invent schemes to coordinate the actions of the individual
decision-makers in a consistent manner. Although problems involving many decision-makers are
present in any system of large size, the number of results available is pitifully small. (See Mesarovic,
et al., [1970] and Marschak and Radner [1971].) In the author’s opinion, these problems represent
one of the most important and challenging areas of research in decision theory.
6 CHAPTER 1. INTRODUCTION
Chapter 2
OPTIMIZATION OVER AN OPEN
SET
In this chapter we study in detail the ﬁrst example of Chapter 1. We ﬁrst establish some notation
which will be in force throughout these Notes. Then we study our example. This will generalize
to a canonical problem, the properties of whose solution are stated as a theorem. Some additional
properties are mentioned in the last section.
2.1 Notation
2.1.1
All vectors are column vectors, with two consistent exceptions mentioned in 2.1.3 and 2.1.5 below
and some other minor and convenient exceptions in the text. Prime denotes transpose so that if
x ∈ R
n
then x

is the row vector x

= (x
1
, . . . , x
n
), and x = (x
1
, . . . , x
n
)

. Vectors are normally
denoted by lower case letters, the ith component of a vector x ∈ R
n
is denoted x
i
, and different
vectors denoted by the same symbol are distinguished by superscripts as in x
j
and x
k
. 0 denotes
both the zero vector and the real number zero, but no confusion will result.
Thus if x = (x
1
, . . . , x
n
)

and y = (y
1
, . . . , y
n
)

then x

y = x
1
y
1
+ . . . + x
n
y
n
as in ordinary
matrix multiplication. If x ∈ R
n
we deﬁne [x[ = +

x

x.
2.1.2
If x = (x
1
, . . . , x
n
)

and y = (y
1
, . . . , y
n
)

then x ≥ y means x
i
≥ y
i
, i = 1, . . . , n. In particular if
x ∈ R
n
, then x ≥ 0, if x
i
≥ 0, i = 1, . . . , n.
2.1.3
Matrices are normally denoted by capital letters. If A is an m n matrix, then A
j
denotes the jth
column of A, and A
i
denotes the ith row of A. Note that A
i
is a row vector. A
j
i
denotes the entry
of A in the ith row and jth column; this entry is sometimes also denoted by the lower case letter
a
ij
, and then we also write A = ¦a
ij
¦. I denotes the identity matrix; its size will be clear from the
context. If confusion is likely, we write I
n
to denote the n n identity matrix.
7
8 CHAPTER 2. OPTIMIZATION OVER AN OPEN SET
2.1.4
If f : R
n
→ R
m
is a function, its ith component is written f
i
, i = 1, . . . , m. Note that f
i
: R
n
→ R.
Sometimes we describe a function by specifying a rule to calculate f(x) for every x. In this case
we write f : x → f(x). For example, if A is an mn matrix, we can write F : x → Ax to denote
the function f : R
n
→ R
m
whose value at a point x ∈ R
n
is Ax.
2.1.5
If f : R
n
→ Ris a differentiable function, the derivative of f at ˆ x is the row vector ((∂f/∂x
1
)(ˆ x), . . . , (∂f/∂x
n
)(ˆ x)).
This derivative is denoted by (∂f/∂x)(ˆ x) or f
x
(ˆ x) or ∂f/∂x[
x=ˆ x
or f
x
[
x=ˆ x
, and if the argument ˆ x
is clear from the context it may be dropped. The column vector (f
x
(ˆ x))

is also denoted ∇
x
f(ˆ x),
and is called the gradient of f at ˆ x. If f : (x, y) → f(x, y) is a differentiable function from
R
n
R
m
into R, the partial derivative of f with respect to x at the point (ˆ x, ˆ y) is the n-dimensional
row vector f
x
(ˆ x, ˆ y) = (∂f/∂x)(ˆ x, ˆ y) = ((∂f/∂x
1
)(ˆ x, ˆ y), . . . , (∂f/∂x
n
)(ˆ x, ˆ y)), and similarly
f
y
(ˆ x, ˆ y) = (∂f/∂y)(ˆ x, ˆ y) = ((∂f/∂y
1
)(ˆ x, ˆ y), . . . , (∂f/∂y
m
)(ˆ x, ˆ y)). Finally, if f : R
n
→ R
m
is
a differentiable function with components f
1
, . . . , f
m
, then its derivative at ˆ x is the mn matrix
∂f
∂x
(ˆ x) = f
x
ˆ x =

f
1x
(ˆ x)
.
.
.
f
mx
(ˆ x)
¸
¸
¸
=

∂f
1
∂x
1
(ˆ x)
.
.
.
∂fm
∂x
1
(ˆ x)
. . .
. . .
∂f
1
∂xn
(ˆ x)
.
.
.
∂fm
∂xn
(ˆ x)
¸
¸
¸
¸
2.1.6
If f : R
n
→ Ris twice differentiable, its second derivative at ˆ x is the nn matrix (∂
2
f/∂x∂x)(ˆ x) =
f
xx
(ˆ x) where (f
xx
(ˆ x))
j
i
= (∂
2
f/∂x
j
∂x
i
)(ˆ x). Thus, in terms of the notation in Section 2.1.5 above,
f
xx
(ˆ x) = (∂/∂x)(f
x
)

(ˆ x).
2.2 Example
We consider in detail the ﬁrst example of Chapter 1. Deﬁne the following variables and functions:
α = fraction of mary-john in proposed mixture,
p = sale price per pound of mixture,
v = total amount of mixture produced,
f(α, p) = expected sales volume (as determined by market research) of mixture as a function of(α, p).
2.2. EXAMPLE 9
Since it is not proﬁtable to produce more than can be sold we must have:
v = f(α, p),
m = amount (in pounds) of mary-john purchased, and
t = amount (in pounds) of tobacco purchased.
Evidently,
m = αv, and
t = (l −α)v.
Let
P
1
(m) = purchase price of m pounds of mary-john, and
P
2
= purchase price per pound of tobacco.
Then the total cost as a function of α, p is
C(α, p) = P
1
(αf(α, p)) +P
2
(1 −α)f(α, p).
The revenue is
R(α, p) = pf(α, p),
so that the net proﬁt is
N(α, p) = R(α, p) −C(α, p).
The set of admissible decisions is Ω, where Ω = ¦(α, p)[0 < α <
1
2
, 0 < p < ∞¦. Formally, we
have the the following decision problem:
Maximize
subject to
N(α, p),
(α, p) ∈ Ω.
Suppose that (α

, p∗) is an optimal decision, i.e.,

, p

) ∈ Ω
N(α

, p

) ≥ N(α, p)
and
for all (α, p) ∈ Ω.
(2.1)
We are going to establish some properties of (a

, p

). First of all we note that Ω is an open subset
of R
2
. Hence there exits ε > 0 such that
(α, p) ∈ Ω whenever [(α, p) −(α

, p

)[ < ε (2.2)
In turn (2.2) implies that for every vector h = (h
1
, h
2
)

in R
2
there exists η > 0 (η of course
depends on h) such that
((α

, p

) +δ(h
1
, h
2
)) ∈ Ω for 0 ≤ δ ≤ η (2.3)
10 CHAPTER 2. OPTIMIZATION OVER AN OPEN SET
|
.

, p

) +δ(h
1
, h
2
)

α
1
2

δh
h
p
(a

, p

)
Figure 2.1: Admissable set of example.
Combining (2.3) with (2.1) we obtain (2.4):
N(α

, p

) ≥ N(α

+δh
1
, p

+δh
2
) for 0 ≤ δ ≤ η (2.4)
Now we assume that the function N is differentiable so that by Taylor’s theorem
N(α

+δh
1
, p

+δh
2
) =
N(α

, p

)
+δ[
∂N
∂α

, p

)h
1
+
∂N
∂p

, p

)h
2
]
+o(δ),
(2.5)
where

δ
→ 0 as δ → 0. (2.6)
Substitution of (2.5) into (2.4) yields
0 ≥ δ[
∂N
∂α

, p

)h
1
+
∂N
∂p

, p

)h
2
] +o(δ).
Dividing by δ > 0 gives
0 ≥ [
∂N
∂α

, p

)h
1
+
∂N
∂p

, p

)h
2
] +
o(δ)
δ
. (2.7)
Letting δ approach zero in (2.7), and using (2.6) we get
0 ≥ [
∂N
∂α

, p

)h
1
+
∂N
∂p

, p

)h
2
]. (2.8)
Thus, using the facts that N is differentiable, (α

, p

) is optimal, and δ is open, we have concluded
that the inequality (2.9) holds for every vector h ∈ R
2
. Clearly this is possible only if
∂N
∂α

, p

) = 0,
∂N
∂p

, p

) = 0. (2.9)
Before evaluating the usefulness of property (2.8), let us prove a direct generalization.
2.3. THE MAIN RESULT AND ITS CONSEQUENCES 11
2.3 The Main Result and its Consequences
2.3.1 Theorem
.
Let Ω be an open subset of R
n
. Let f: R
n
→ R be a differentiable function. Let x

be an optimal
solution of the following decision-making problem:
Maximize
subject to
f(x)
x ∈ Ω.
(2.10)
Then
∂f
∂x
(x

) = 0. (2.11)
Proof: Since x

∈ Ω and Ω is open, there exists ε > 0 such that
x ∈ Ω whenever [x −x

[ < ε. (2.12)
In turn, (2.12) implies that for every vector h ∈ R
n
there exits η > 0 (η depending on h) such that
(x

+δh) ∈ Ω whenever 0 ≤ δ ≤ η. (2.13)
Since x

is optimal, we must then have
f(x

) ≥ f(x

+δh) whenever 0 ≤ δ ≤ η. (2.14)
Since f is differentiable, by Taylor’s theorem we have
f(x

+δh) = f(x

) +
∂f
∂x
(x

)δh +o(δ), (2.15)
where
o(δ)
δ
→ 0 as δ → 0 (2.16)
Substitution of (2.15) into (2.14) yields
0 ≥ δ
∂f
∂x
(x

)h +o(δ)
and dividing by δ > 0 gives
0 ≥
∂f
∂x
(x

)h +
o(δ)
δ
(2.17)
Letting δ approach zero in (2.17) and taking (2.16) into account, we see that
0 ≥
∂f
∂x
(x

)h, (2.18)
Since the inequality (2.18) must hold for every h ∈ R
n
, we must have
0 =
∂f
∂x
(x

),
and the theorem is proved. ♦
12 CHAPTER 2. OPTIMIZATION OVER AN OPEN SET
Table 2.1
Does there exist At how many points
an optimal deci- in Ω is 2.2.2 Further
Case sion for 2.2.1? satisﬁed? Consequences
1 Yes Exactly one point, x

is the
say x

unique optimal
2 Yes More than one point
3 No None
4 No Exactly one point
5 No More than one point
2.3.2 Consequences.
Let us evaluate the usefulness of (2.11) and its special case (2.18). Equation (2.11) gives us n
equations which must be satisﬁed at any optimal decision x

= (x

1
, . . . , x

n
)

.
These are
∂f
∂x
1
(x

) = 0,
∂f
∂x
2
(x

) = 0, . . . ,
∂f
∂xn
(x

) = 0 (2.19)
Thus, every optimal decision must be a solution of these n simultaneous equations of n variables, so
that the search for an optimal decision from Ω is reduced to searching among the solutions of (2.19).
In practice this may be a very difﬁcult problem since these may be nonlinear equations and it may
be necessary to use a digital computer. However, in these Notes we shall not be overly concerned
with numerical solution techniques (but see 2.4.6 below).
The theorem may also have conceptual signiﬁcance. We return to the example and recall the
N = R − C. Suppose that R and C are differentiable, in which case (2.18) implies that at every
optimal decision (α

, p

)
∂R
∂α

, p

) =
∂C
∂α

, p

),
∂R
∂p

, p

) =
∂C
∂p

, p

),
or, in the language of economic analysis, marginal revenue = marginal cost. We have obtained an
important economic insight.
2.4 Remarks and Extensions
2.4.1 A warning.
Equation (2.11) is only a necessary condition for x

to be optimal. There may exist decisions ˜ x ∈ Ω
such that f
x
(˜ x) = 0 but ˜ x is not optimal. More generally, any one of the ﬁve cases in Table 2.1 may
occur. The diagrams in Figure 2.1 illustrate these cases. In each case Ω = (−1, 1).
Note that in the last three ﬁgures there is no optimal decision since the limit points -1 and +1 are
not in the set of permissible decisions Ω = (−1, 1). In summary, the theorem does not give us any
clues concerning the existence of an optimal decision, and it does not give us sufﬁcient conditions
either.
2.4. REMARKS AND EXTENSIONS 13
Case 1 Case 2 Case 3
Case 5 Case 4
-1
1 -1 1
-1 1
1 -1 -1 1
Figure 2.2: Illustration of 4.1.
2.4.2 Existence.
If the set of permissible decisions Ω is a closed and bounded subset of R
n
, and if f is continuous,
then it follows by the Weierstrass Theorem that there exists an optimal decision. But if Ω is closed
we cannot assert that the derivative of f vanishes at the optimum. Indeed, in the third ﬁgure above,
if Ω = [−1, 1], then +1 is the optimal decision but the derivative is positive at that point.
2.4.3 Local optimum.
We say that x

∈ Ω is a locally optimal decision if there exists ε > 0 such that f(x

) ≥ f(x)
whenever x ∈ Ω and [x

− x[ ≤ ε. It is easy to see that the theorem holds (i.e., 2.11) for local
optima also.
2.4.4 Second-order conditions.
Suppose f is twice-differentiable and let x

∈ Ω be optimal or even locally optimal. Then f
x
(x

) =
0, and by Taylor’s theorem
f(x

+δh) = f(x

) +
1
2
δ
2
h

f
xx
(x

)h +o(δ
2
), (2.20)
where
o(δ
2
)
δ
2
→ 0 as δ → 0. Now for δ > 0 sufﬁciently small f(x

+δh) ≤ f(x

), so that dividing
by δ
2
> 0 yields
0 ≥
1
2
h

f
xx
(x

)h +
o(δ
2
)
δ
2
and letting δ approach zero we conclude that h

f
xx
(x

)h ≤ 0 for all h ∈ R
n
. This means that
f
xx
(x

) is a negative semi-deﬁnite matrix. Thus, if we have a twice differentiable objective function,
we get an additional necessary condition.
2.4.5 Sufﬁciency for local optimal.
Suppose at x

∈ Ω, f
x
(x

) = 0 and f
xx
is strictly negative deﬁnite. But then from the expansion
(2.20) we can conclude that x

is a local optimum.
14 CHAPTER 2. OPTIMIZATION OVER AN OPEN SET
2.4.6 A numerical procedure.
At any point ˜ x ∈ Ω the gradient
x
f(˜ x) is a direction along which f(x) increases, i.e., f(˜ x+ε
x
f(˜ x)) > f(˜ x) for all ε > 0 sufﬁciently small. This observation suggests the following scheme for
ﬁnding a point x

∈ Ω which satisﬁes 2.11. We can formalize the scheme as an algorithm.
Step 1. Pick x
0
∈ Ω. Set i = 0. Go to Step 2.
Step 2. Calculate
x
f(x
i
). If
x
f(x
i
) = 0, stop.
Otherwise let x
i+1
= x
i
+d
i

x
f(x
i
) and go
to Step 3.
Step 3. Set i = i + 1 and return to Step 2.
The step size d
i
can be selected in many ways. For instance, one choice is to take d
i
to be an
optimal decision for the following problem:
Max¦f(x
i
+d
x
f(x
i
))[d > 0, (x
i
+d
x
f(x
i
)) ∈ Ω¦.
This requires a one-dimensional search. Another choice is to let d
i
= d
i−1
if f(x
i
+ d
i−1

x
f(x
i
)) > f(x
i
); otherwise let d
i
= 1/k d
i−1
where k is the smallest positive integer such that
f(x
i
+ 1/k d
i−1

x
f(x
i
)) > f(x
i
). To start the process we let d
−1
> 0 be arbitrary.
Exercise: Let f be continuous differentiable. Let ¦d
i
¦ be produced by either of these choices and
let
¦x
i
¦ be the resulting sequence. Then
1. f(x
i+1
) > f(x
i
) if x
i+1
= x
i
, i
2. if x

∈ Ω is a limit point of the sequence ¦x
i
¦, f
x
(x

) = 0.
For other numerical procedures the reader is referred to Zangwill [1969] or Polak [1971].
Chapter 3
OPTIMIZATION OVER SETS
DEFINED BY EQUALITY
CONSTRAINTS
We ﬁrst study a simple example and examine the properties of an optimal decision. This will
generalize to a canonical problem, and the properties of its optimal decisions are stated in the form
of a theorem. Additional properties are summarized in Section 3 and a numerical scheme is applied
to determine the optimal design of resistive networks.
3.1 Example
We want to ﬁnd the rectangle of maximum area inscribed in an ellipse deﬁned by
f
1
(x, y) =
x
2
a
2
+
y
2
b
2
= α.
(3.1)
The problem can be formalized as follows (see Figure 3.1):
Maximize
subject to
f
0
(x, y)
(x, y) ∈ Ω
= 4xy
= ¦(x, y)[f
1
(x, y) = α¦.
(3.2)
The main difference between problem (3.2) and the decisions studied in the last chapter is that
the set of permissible decisions Ω is not an open set. Hence, if (x

, y

) is an optimal decision we
cannot assert that f
0
(x

, y

) ≥ f
0
(x, y) for all (x, y) in an open set containing (x

, y

). Returning
to problem (3.2), suppose (x

, y

) is an optimal decision. Clearly then either x

= 0 or y

= 0. Let
us suppose y

= 0. Then from ﬁgure 3.1 it is evident that there exist (i)ε > 0, (ii) an open set V
containing (x

, y

), and (iii) a differentiable function g : (x

−ε, x

+ε) → V such that
f
1
(x, y) = α and (x, y) ∈ V iff fy = g(x).
1
(3.3)
In particular this implies that y

= g(x

), and that f
1
(x, g(x)) = α whenever [x − x

[ < ε. Since
1
Note that y

= 0 implies f1y(x

, Y

) = 0, so that this assertion follows from the Implicit Function Theorem. The
assertion is false if y

= 0. In the present case let 0 < ε ≤ a −x

and g(x) = +b[α −(x/a)
2
]
1/2
.
15
16 CHAPTER 3. OPTIMIZATION WITH EQUALITY CONSTRAINTS
) ( |
-
y

g(x)
Tangent plane to
Ω at (x

, y

)
(f
1x
, f
1y
)
V

x

x

Figure 3.1: Illustration of example.
(x

, y

) = (x

, g(x

)) is optimum for (3.2), it follows that x

is an optimal solution for (3.4):
Maximize
subject to
ˆ
f
0
(x) = f
0
(x, g(x))
[x −x

[ < ε.
(3.4)
But the constraint set in (3.4) is an open set (in R
1
) and the objective function
ˆ
f
0
is differentiable,
so that by Theorem 2.3.1,
ˆ
f
0x
(x

) = 0, which we can also express as
f
0x
(x

, y

) +f
0y
(x

, y

)g
x
(x

) = 0 (3.5)
Using the fact that f
1
(x, g(x)) ≡ α for [x −x

[ < ε, we see that
f
1x
(x

, y

) +f
1y
(x

, y

)g
x
(x

) = 0,
and since f
1y
(x

, y

) = 0 we can evaluate g
x
(x

),
g
x
(x

) = −f
−1
1y
f
1x
(x

, y

),
and substitute in (3.5) to obtain the condition (3.6):
f
0x
−f
0y
f
−1
1y
f
1x
= 0 at (x

, y

). (3.6)
Thus an optimal decision (x

, y

) must satisfy the two equations f
1
(x

, y

) = α and (3.6). Solving
these yields
x

=
+

(α/2)
1/2
a , y

=
+

(α/2)
1/2
b.
3.2. GENERAL CASE 17
Evidently there are two optimal decisions, (x

, y

) =
+

(α/2)
1/2
(a, b), and the maximum area is
m(α) = 2αab. (3.7)
The condition (3.6) can be interpreted differently. Deﬁne
λ

= f
0y
f
−1
1y
(x

, y

). (3.8)
Then (3.6) and (3.8) can be rewritten as (3.9):
(f
0x
, f
0y
) = λ

(f
1x
, f
1y
) at (x

, y

) (3.9)
In terms of the gradients of f
0
, f
1
, (3.9) is equivalent to
f
0
(x

, y

) = [f
1
(x

, y

)]λ

, (3.10)
which means that at an optimal decision the gradient of the objective function f
0
is normal to the
plane tangent to the constraint set Ω.
Finally we note that
λ

=
∂m
∂α
. (3.11)
where m(α) = maximum area.
3.2 General Case
3.2.1 Theorem.
Let f
i
: R
n
→ R, i = 0, 1, . . . , m (m < n), be continuously differentiable functions and let x

be
an optimal decision of problem (3.12):
Maximize
subject to
f
0
(x)
f
i
(x) = α
i
, i = 1, . . . , m.
(3.12)
Suppose that at x

the derivatives f
ix
(x

), i = 1, . . . , m, are linearly independent. Then there exists
a vector λ

= (λ

1
, . . . , λ

m
)

such that
f
0x
(x

) = λ

1
f
1x
(x

) +. . . +λ

m
f
mx
(x

) (3.13)
Furthermore, let m(α
1
, . . . , α
m
) be the maximum value of (3.12) as a function of α = (α
1
, . . . , α
m
)

.
Let x

(α) be an optimal decision for (3.12). If x

(α) is a differentiable function of α then m(α) is
a differentiable function of α, and

)

=
∂m
∂α
(3.14)
Proof. Since f
ix
(x

), i = 1, . . . , m, are linearly independent, then by re-labeling the coordinates of
x if necessary, we can assume that the mmmatrix [(∂f
i
/∂x
j
)(x

)], 1 ≤ i, j ≤ m, is nonsingular.
By the Implicit Function Theorem (see Fleming [1965]) it follows that there exist (i) ε > 0, (ii) an
18 CHAPTER 3. OPTIMIZATION WITH EQUALITY CONSTRAINTS
open set V in R
n
containing x

, and (iii) a differentiable function g : U → R
m
, where U =
[(x
m+1
, . . . , x
n
)][ [x
m+
−x

m+
[ < ε, = 1, . . . , n −m], such that
f
i
(x
1
, . . . , x
n
) = α
i
, 1 ≤ i ≤ m, and (x
1
, . . . , x
n
) ∈ V
iff
x
j
= g
j
(x
m+1
, . . . , x
n
), 1 ≤ j ≤ m, and (x
m+1
, . . . , x
n
) ∈ U (3.15)
(see Figure 3.2).
In particular this implies that x

j
= g
j
(x

m+1
, . . . , x

n
), 1 ≤ j ≤ m, and
f
i
(g(x
m+1
, . . . , x
n
), x
m+1
, . . . , x
n
) = α
i
, i = 1, . . . , m. (3.16)
For convenience, let us deﬁne w = (x
1
, . . . , x
m
)

, u = (x
m+1
, . . . , x
n
)

and f = (f
1
, . . . , f
m
)

.
Then, since x

= (w

, u

) = (g(u

), u

) is optimal for (3.12), it follows that u

is an optimal
decision for (3.17):
Maximize
subject to
ˆ
f
0
(u) = f
0
(g(u), u)
u ∈ U.
(3.17)
But U is an open subset of R
n−m
and
ˆ
f
0
is a differentiable function on U (since f
0
and g are
differentiable), so that by Theorem 2.3.1 ,
ˆ
f
0u
(u

) = 0, which we can also express using the chain
rule for derivatives as
ˆ
f
0u
(u

) = f
0w
(x

)g
u
(u

) +f
0u
(x

) = 0. (3.18)
Differentiating (3.16) with respect to u = (x
m+1
, . . . , x
n
)

, we see that
f
w
(x

)g
u
(u

) +f
u
(x

) = 0,
and since the mm matrix f
w
(x

) is nonsingular we can evaluate g
u
(u

),
g
u
(u

) = −[f
w
(x∗)]
−1
f
u
(x

),
and substitute in (3.18) to obtain the condition
−f
0w
f
−1
w
f
u
+f
0u
= 0 at x

= (w

, u

). (3.19)
Next, deﬁne the m-dimensional column vector λ

by

)

= f
0w
f
−1
w
[x

. (3.20)
Then (3.19) and (3.20) can be written as (3.21):
(f
0w
(x

), f
0u
(x

)) = (λ

)

(f
w
(x

), f
u
(x

)). (3.21)
Since x = (w, u), this is the same as
f
0x
(x

) = (λ

)

f
x
(x

) = λ

1
f
1x
(x

) +. . . +λ

m
f
mx
(x

),
3.2. GENERAL CASE 19
.
.
.
.
x
1
, . . . , x
m
x

V
x
m+1
(x
m+1
, . . . , x
n
)
(x

m+1
, . . . , x

n
)
2
U
x
n
Ω =
¦x[f
i
(x) = α
i
¦
i = 1, . . . , m
(x

1
, . . . , x

m
)
g(x
m+1
, . . . , x
n
)
Figure 3.2: Illustration of theorem.
which is equation (3.13).
To prove (3.14), we vary α in a neighborhood of a ﬁxed value, say α. We deﬁne w

(α) =
(x

1
(α), . . . , x

m
(α))

and u

(α) = (x

m+1
(α), . . . , x

(
α))

. By hypothesis, f
w
is nonsingular at
x

(α). Since f(x) and x

(α) are continuously differentiable by hypothesis, it follows that f
w
is
nonsingular at x

(α) in a neighborhood of α, say N. We have the equation
f(w

(α), u

(α)) = α, (3.22)
−f
0w
f
−1
w
f
u
+f
0u
= 0 at (w

(α), u

(α)), (3.23)
for α ∈ N. Also, m(α) = f
0
(x

(α)), so that
m
α
= f
0w
w

α
+f
0u
u

α
(3.24)
Differentiating (3.22) with respect to α gives
f
w
w

α
+f
u
u

α
= I,
so that
w

α
+f
−1
w
f
u
u

α
= f
−1
w
,
20 CHAPTER 3. OPTIMIZATION WITH EQUALITY CONSTRAINTS
and multiplying on the left by f
0w
gives
f
0w
w

α
+f
0w
f
−1
w
f
u
u

α
= f
0w
f
−1
w
.
Using (3.23), this equation can be rewritten as
f
0w
w

α
+f
0u
u

α
= f
0w
f
−1
w
. (3.25)
In (3.25), if we substitute from (3.20) and (3.24), we obtain (3.14) and the theorem is proved. ♦
3.2.2 Geometric interpretation.
The equality constraints of the problem in 3.12 deﬁne a n −m dimensional surface
Ω = ¦x[f
i
(x) = α
i
, i = 1, . . . , m¦.
The hypothesis of linear independence of ¦f
ix
(x

)[1 ≤ i ≤ m¦ guarantees that the tangent plane
through Ω at x

is described by
¦h[f
ix
(x

)h = 0 , i = 1, . . . , m¦, (3.26)
so that the set of (column vectors orthogonal to this tangent surface is
¦λ
1

x
f
1
(x

) +. . . +λ
m

x
f
m
(x

)[λ
i
∈ R, i = 1, . . . , m¦.
Condition (3.13) is therefore equivalent to saying that at an optimal decision x

objective function
x
f
0
(x

) is normal to the tangent surface (3.12).
3.2.3 Algebraic interpretation.
Let us again deﬁne w = (x
1
, . . . , x
m
)

and u = (x
m+1
, . . . , x
n
)

. Suppose that f
w
(˜ x) is nonsin-
gular at some point ˜ x = ( ˜ w, ˜ u) in Ω which is not necessarily optimal. Then the Implicit Function
Theorem enables us to solve, in a neighborhood of ˜ x, the mequations f(w, u) = α. u can then vary
arbitrarily in a neighborhood of ˜ u. As u varies, w must change according to w = g(u) (in order to
maintain f(w, u) = α), and the objective function changes according to
ˆ
f
0
(u) = f
0
(g(u), u). The
derivative of
ˆ
f
0
at ˜ u is
ˆ
f
0u
(˜ u) = f
0w
g
u
+f
0u˜ x
= −
˜
λ

f
u
(˜ x) +f
0u
(˜ x),
where
˜
λ

= f
0w
f
−1
w˜ x
, (3.27)
Therefore, the direction of steepest increase of
ˆ
f
0
at ˜ u is

u
ˆ
f
0
(˜ u) = −f

u
(˜ x)
˜
λ +f

Ou
(˜ x) , (3.28)
and if ˜ u is optimal,
u
ˆ
f
0
(˜ u) = 0 which, together with (3.27) is equation (3.13). We shall use (3.27)
and (3.28) in the last section.
3.3. REMARKS AND EXTENSIONS 21
3.3 Remarks and Extensions
3.3.1 The condition of linear independence.
The necessary condition (3.13) need not hold if the derivatives f
ix
(x

), 1 ≤ i ≤ m, are not linearly
independent. This can be checked in the following example
Minimize
subject to sin(x
2
1
+x
2
2
)
π
2
(x
2
1
+x
2
2
) = 1.
(3.29)
3.3.2 An alternative condition.
Keeping the notation of Theorem 3.2.1, deﬁne the Lagrangian function L : R
n+m
→ R by L :
(x, λ) → f
0
(x) −
¸
m
i=1
λ
i
f
i
(x). The following is a reformulation of 3.12, and its proof is left as
an exercise.
Let x

be optimal for (3.12), and suppose that f
ix
(x

), 1 ≤ i ≤ m, are linearly independent.
Then there exists λ

∈ R
m
such that (x

, λ

) is a stationary point of L, i.e., L
x
(x

, λ

) = 0 and
L
λ
(x

, λ

) = 0.
3.3.3 Second-order conditions.
Since we can convert the problem (3.12) into a problem of maximizing
ˆ
f
0
over an open set, all
the comments of Section 2.4 will apply to the function
ˆ
f
0
. However, it is useful to translate these
remarks in terms of the original function f
0
and f. This is possible because the function g is
uniquely speciﬁed by (3.16) in a neighborhood of x

. Furthermore, if f is twice differentiable, so
is g (see Fleming [1965]). It follows that if the functions f
i
, 0 ≤ i ≤ m, are twice continuously
differentiable, then so is
ˆ
f
0
, and a necessary condition for x

to be optimal for (3.12) and (3.13) and
the condition that the (n − m) (n − m) matrix
ˆ
f
0uu
(u

) is negative semi-deﬁnite. Furthermore,
if this matrix is negative deﬁnite then x

is a local optimum. the following exercise expresses
f
ˆ
f
0uu
(u

) in terms of derivatives of the functions f
i
.
Exercise: Show that
ˆ
f
0uu
(u

) = [g

u
.
.
.I]
¸
L
ww
L
uw
L
wu
L
uu

g
u
. . .
I
¸
¸

(w

, u

)
where
g
u
(u

) = −[f
w
(x

)]
−1
f
u
(x

), L(x) = f
0
(x) −
m
¸
i=1
λ

i
f
i
(x).
22 CHAPTER 3. OPTIMIZATION WITH EQUALITY CONSTRAINTS
3.3.4 A numerical procedure.
We assume that the derivatives f
ix
(x), 1 ≤ i ≤ m, are linearly independent for all x. Then the
following algorithm is a straightforward adaptation of the procedure in Section 2.4.6.
Step 1. Find x
0
arbitrary so that f
i
(x
0
) = α
i
, 1 ≤ i ≤ m. Set k = 0 and go to Step 2.
Step 2. Find a partition x = (w, u)
2
of the variables such that f
w
(x
k
) is nonsingular. Calculate λ
k
by (λ
k
)

= f
0w
f
−1
w(xk)
, and
ˆ
f
k
0
(u
k
) = −f

u
(x
k

k
+f

0u
(x
k
). If
ˆ
f
k
0
(u
k
) = 0, stop. Otherwise
go to Step 3.
Step 3. Set ˜ u
k
= u
k
+d
k

ˆ
f
k
0
(u
k
). Find ˜ w
k
such that f
i
( ˜ w
k
, ˜ u
k
) = 0, 1 ≤ i ≤ m. Set
x
k+1
= ( ˜ w
k
, ˜ u
k
), set k = k + 1, and return to Step 2.
Remarks. As before, the step sizes d
k
> 0 can be selected various ways. The practical applicability
of the algorithm depends upon two crucial factors: the ease with which we can ﬁnd a partition
x = (w, u) so that f
w
(x
k
) is nonsingular, thus enabling us to calculate λ
k
; and the ease with which
we can ﬁnd ˜ w
k
so that f( ˜ w
k
, ˜ u
k
) = α. In the next section we apply this algorithm to a practical
problem where these two steps can be carried out without too much difﬁculty.
3.3.5 Design of resistive networks.
Consider a network N with n + 1 nodes and b branches. We choose one of the nodes as datum
and denote by e = (e
1
, . . . , e
n
)

the vector of node-to-datum voltages. Orient the network graph
and let v = (v
1
, . . . , v
b
)

and j = (j
1
, . . . , j
b
)

respectively, denote the vectors of branch voltages
and branch currents. Let A be the n b reduced incidence matrix of the network graph. Then the
Kirchhoff current and voltage laws respectively yield the equations
Aj = 0 and A

e = v (3.30)
Next we suppose that each branch k contains a (possibly nonlinear)resistive element with the form
shown in Figure 3.3, so that
j
k
−j
sk
= g
k
(v
rk
) = g
k
(v
k
−v
sk
), 1 ≤ k ≤ b, (3.31)
where v
rk
is the voltage across the resistor. Here j
sk
, v
sk
are the source current and voltage in the
kth branch, and g
k
is the characteristic of the resistor. Using the obvious vector notation j
s
∈ R
b
,
v
s
∈ R
b
for the sources, v
r
∈ R
b
for the resistor voltages, and g = (g
1
, . . . , g
b
)

, we can rewrite
(3.30) as (3.31):
j −j
s
= g(v −v
s
) = g(v
r
). (3.32)
Although (3.30) implies that the current (j
k
−j
s
k) through the kth resistor depends only on the
voltage v
rk
= (v
k
−v
sk
) across itself, no essential simpliﬁcation is achieved. Hence, in (3.31) we
shall assume that g
k
is a function of v
r
. This allows us to include coupled resistors and voltage-
controlled current sources. Furthermore, let us suppose that there are design parameters p =
(p
1
, . . . , p

)

which are under our control, so that (3.31) is replaced by (3.32):
j −j
x
= g(v
r
, p) = g(v−v
s
, p). (3.33)
2
This is just a notational convenience. The w variable may consist of any m components of x.
3.3. REMARKS AND EXTENSIONS 23
o
-
+ -
+ -
+
o
j
sk
v
rk
v
sk
j
k
−j
sk
j
k
v
k
Figure 3.3: The kth branch.
If we combine (3.29) and (3.32) we obtain (3.33):
Ag(A

e −v
s
, p) = i
s
, (3.34)
where we have deﬁned i
s
= A
js
. The network design problem can then be stated as ﬁnding p, v
s
, i
s
so as to minimize some speciﬁed function f
0
(e, p, v
s
, i
s
). Formally, we have the optimization prob-
lem (3.34):
Minimize
subject to
f
0
(e, p, v
s
, i
s
)
Ag(A

e −v
s
, p) −i
s
= 0.
(3.35)
We shall apply the algorithm 3.3.4 to this problem. To do this we make the following assumption.
Assumption: (a) f
0
is differentiable. (b) g is differentiable and the nn matrix A(∂g/∂v)(v, p)A

is nonsingular for all v ∈ R
b
, p ∈ R

. (c) The network N described by (3.33) is determinate i.e.,
for every value of (p, v
s
, i
s
) there is a unique e = E(p, v
s
, i
s
) satisfying (3.33).
In terms of the notation of 3.3.4, if we let x = (e, p, v
s
, i
s
), then assumption (b) allows us to
identify w = e, and u = (p, v
s
, i
s
). Also let f(x) = f(e, p, v
s
, i
s
) = Ag(A

e−v
s
, p) −i
s
. Now the
crucial part in the algorithm is to obtain λ
k
at some point x
k
. To this end let ˜ x = (˜ e, ˜ p, ˜ v
s
,
˜
i
s
) be a
ﬁxed point. Then the corresponding λ =
˜
λ is given by (see (3.27))
˜
λ

= f
0w
(˜ x)f
−1
w
(˜ x) = f
0e
(˜ x)f
−1
e
(˜ x). (3.36)
From the deﬁnition of f we have
f
e
(˜ x) = AG(˜ v
r
, ˜ p)A

,
where ˜ v
r
= A

˜ e − ˜ v
s
, and G(˜ v
r
, ˜ p) = (∂g/∂v
r
)(˜ v
r
, ˜ p). Therefore,
˜
λ is the solution (unique by
assumption (b)) of the following linear equation:
AG

(˜ v
r
, ˜ p)A

˜
λ = f

0e
(˜ x). (3.37)
Now (3.36) has the following extremely interesting physical interpretation. If we compare (3.33)
with (3.36) we see immediately that
˜
λ is the node-to-datum response voltages of a linear network
N(˜ v
r
, ˜ p) driven by the current sources f

0e
(˜ x). Furthermore, this network has the same graph as
the original network (since they have the same incidence matrix); moreover, its branch admittance
matrix, G

(˜ v
r
, ˜ p), is the transpose of the incremental branch admittance matrix (evaluated at (˜ v
r
, ˜ p))
of the original network N. For this reason, N(˜ v
r
, ˜ p) is called the adjoint network (of N) at (˜ v
r
, ˜ p).
24 CHAPTER 3. OPTIMIZATION WITH EQUALITY CONSTRAINTS
Once we have obtained
˜
λ we can obtain
u
ˆ
f
0
(˜ u) using (3.28). Elementary calculations yield
(3.37):

u
ˆ
f
0
(˜ u) =

ˆ
f

0p
(˜ u)
ˆ
f

0vs
(˜ u)
ˆ
f

0is
(˜ u)
¸
¸
¸ =

[
∂g
∂p
(˜ v
r
, ˜ p)]

A

G

(˜ v
r
, ˜ p)A

−I
¸
¸ ˜
λ +

f

0p
(˜ x)
f

0vs
(˜ x)
f

0is
(˜ x)
¸
¸
(3.38)
We can now state the algorithm.
Step 1. Select u
0
= (p
0
, v
0
s
, i
0
s
) arbitrary. Solve (3.33) to obtain e
0
= E(p
0
, v
0
s
, i
0
s
). Let k = 0 and
go to Step 2.
Step 2. Calculate v
k
r
= A

e
k
−v
k
s
. calculate f

0e
(x
k
). Calculate the node-to-datum response λ
k
of
k
r
, p
k
) driven by the current source f

0e
(x
k
). Calculate
u
ˆ
f
0
(u
k
) from
(3.37). If this gradient is zero, stop. Otherwise go to Step 3.
Step 3. Let u
k+1
= (p
k+1
, v
k+1
s
, i
k+1
s
) = u
k
−d
k

u
ˆ
f
0
(u
k
), where d
k
> 0 is a predetermined
step size.
3
Solve (3.33) to obtain e
k+1
= (Ep
k+1
, v
k+1
s
, i
k+1
s
Remark 1. Each iteration from u
k
to u
k+1
requires one linear network analysis step (the
computation of λ
k
in Step 2), and one nonlinear network analysis step (the computation of e
k+1
in
step 3). This latter step may be very complex.
Remark 2. In practice we can control only some of the components of v
s
and i
s
, the rest being
ﬁxed. The only change this requires in the algorithm is that in Step 3 we set
p
k+1
= p
k
−d
k
ˆ
f

0p
(u
k
) just as before, where as v
k+1
sj
= v
k
sj
−d
k
(∂
ˆ
f
0
/∂v
sj
)(u
k
) and
i
k+1
sm
= i
k
sm
−d
k
(∂
ˆ
f
0
/∂i
sm
)(u
k
) with j and m ranging only over the controllable components and
the rest of the components equal to their speciﬁed values.
Remark 3. The interpretation of λ as the response of the adjoint network has been exploited for
particular function f
0
in a series of papers (director and Rohrer [1969a], [1969b], [1969c]). Their
derivation of the adjoint network does not appear as transparent as the one given here. Although
we have used the incidence matrix A to obtain our network equation (3.33), one can use a more
general cutset matrix. Similarly, more general representations of the resistive elements may be
employed. In every case the “adjoint” network arises from a network interpretation of (3.27),
[f
w
(˜ x)]

˜
λ = f
0w
(˜ x),
with the transpose of the matrix giving rise to the adjective “adjoint.”
Exercise: [DC biasing of transistor circuits (see Dowell and Rohrer [1971]).] Let N be a transistor
circuit, and let (3.33) model the dc behavior of this circuit. Suppose that i
s
is ﬁxed, v
sj
for j ∈ J
are variable, and v
sj
for j / ∈ J are ﬁxed. For each choice of v
sj
, j ∈ J, we obtain the vector e and
hence the branch voltage vector v = A

e. Some of the components v
t
, t ∈ T, will correspond to
bias voltages for the transistors in the network, and we wish to choose v
sj
, j ∈ J, so that v
t
is as
close as possible to a desired bias voltage v
d
t
, t ∈ T. If we choose nonnegative numbers α
t
, with
relative magnitudes reﬂecting the importance of the different transistors then we can formulate the
criterion
3
k
−d
k
u
ˆ
f0(u
k
). Remember we are minimizing f0, which is equivalent to
maximizing (−f0).
3.3. REMARKS AND EXTENSIONS 25
f
0
(e) =
¸
t∈T
α
t
[v
t
−v
d
t
[
2
.
(i) Specialize the algorithm above for this particular case.
(ii) How do the formulas change if the network equations are written using an arbitrary cutset matrix
26 CHAPTER 3. OPTIMIZATION WITH EQUALITY CONSTRAINTS
Chapter 4
OPTIMIZATION OVER SETS
DEFINED BY INEQUALITY
CONSTRAINTS: LINEAR
PROGRAMMING
In the ﬁrst section we study in detail Example 2 of Chapter I, and then we deﬁne the general linear
programming problem. In the second section we present the duality theory for linear program-
ming and use it to obtain some sensitivity results. In Section 3 we present the Simplex algorithm
which is the main procedure used to solve linear programming problems. In section 4 we apply
the results of Sections 2 and 3 to study the linear programming theory of competitive economy.
cessible treatment of the material presented in this chapter see the companion volume in this Series
(Sakarovitch [1971]).
4.1 The Linear Programming Problem
4.1.1 Example.
Recall Example 2 of Chapter I. Let g and u respectively be the number of graduate and undergradu-
ate students admitted. Then the number of seminars demanded per year is
2g+u
20
, and the number of
lecture courses demanded per year is
5g+7u
40
. On the supply side of our accounting, the faculty can
offer 2(750) + 3(250) = 2250 seminars and 6(750) + 3(250) = 5250 lecture courses. Because of
his contractual agreements, the President must satisfy
2g+u
20
≤ 2250 or 2g +u ≤ 45, 000
and
5g+7u
40
≤ 5250 or 5g + 7u ≤ 210, 000 .
27
28 CHAPTER 4. LINEAR PROGRAMMING
Since negative g or u is meaningless, there are also the constraints g ≥ 0, u ≥ 0. Formally then the
President faces the following decision problem:
Maximize αg +βu
subject to 2g +u ≤ 45, 000
5g + 7u ≤ 210, 000
g ≥ 0, u ≥ 0 .
(4.1)
It is convenient to use a more general notation. So let x = (g, u)

, c = (α, β)

, b = (45000, 210000, 0, 0)

and let A be the 42 matrix
A =

2
5
−1
0
1
7
0
−1
¸
¸
¸
¸
.
Then (4.1) can be rewritten as (4.2)
1
Maximize c

x
subject to Ax ≤ b .
(4.2)
Let A
i
, 1 ≤ i ≤ 4, denote the rows of A. Then the set Ω of all vectors x which satisfy the constraints
in (4.2) is given by Ω = ¦x[A
i
x ≤ b
i
, 1 ≤ i ≤ 4¦ and is the polygon OPQR in Figure 4.1.
For each choice x, the President receives the payoff c

x. Therefore, the surface of constant payoff
k say, is the hyperplane π(k) = ¦x[c

x = k¦. These hyperplanes for different values of k are
parallel to one another since they have the same normal c. Furthermore, as k increases π(k) moves
in the direction c. (Obviously we are assuming in this discussion that c = 0.) Evidently an optimal
decision is any point x

∈ Ω which lies on a hyperplane π(k) which is farthest along the direction
c. We can rephrase this by saying that x∗ ∈ Ω is an optimal decision if and only if the plane π

through x

does not intersect the interior of Ω, and futhermore at x

the direction c points away
from Ω. From this condition we can immediately draw two very important conclusions: (i) at least
one of the vertices of Ω is an optimal decision, and (ii) x

yields a higher payoff than all points
in the cone K

consisting of all rays starting at x

and passing through Ω, since K

lies “below”
π

. The ﬁrst conclusion is the foundation of the powerful Simplex algorithm which we present in
Section 3. Here we pursue consequences of the second conclusion. For the situation depicted in
Figure 4.1 we can see that x

= Q is an optimal decision and the cone K

is shown in Figure 4.2.
Now x

satisﬁes A
x
x

= b
1
, A
2
x

= b
2
, and A
3
x

< b
3
, A
4
x

< b
4
, so that K

is given by
K

= ¦x

+h[A
1
h ≤ 0 , A
2
h ≤ 0¦ .
Since c

x

≥ c

y for all y ∈ K

we conclude that
c

h ≤ 0 for all h such that A
1
h ≤ 0, A
2
h ≤ 0 . (4.3)
We pause to formulate the generalization of (4.3) as an exercise.
1
Recall the notation introduced in 1.1.2, so that x ≤ y means xi ≤ yi for all i.
4.1. THE LINEAR PROGRAMMING PROBLEM 29
,
-
-
-
-
-
-
-
-
-
-
x
2
π(k) = ¦x[c

x = k¦
π

Q = x

direction of
increasing
payoff k
¦x[A
2
x = b
2
¦
x
1
¦x[A
1
x = b
1
¦
R
A
4
O A
3
A
1
⊥ QR
c ⊥ π

A
2
⊥ PQ
P
Figure 4.1: Ω = OPQR.
Exercise 1: Let A
i
, 1 ≤ i ≤ k, be n-dimensional row vectors. Let c ∈ R
n
, and let b
i
, 1 ≤ i ≤ k,
be real numbers. Consider the problem
Maximize c

x
subject to A
i
x ≤ b
i
, 1 ≤ i ≤ k .
For any x satisfying the constraints, let I(x) ⊂ ¦1, . . . , n¦ be such that A
i
(x) = b
i
, i ∈ I(x), A
i
x <
b
i
, i / ∈ I(x). Suppose x

satisﬁes the constraints. Show that x

is optimal if an only if
c

h ≤ 0 for all h such that A
i
h ≤ 0 , i ∈ I(x

).
Returning to our problem, it is clear that (4.3) is satisﬁed as long as c lies between A
1
and A
2
.
Mathematically this means that (4.3) is satisﬁed if and only if there exist λ

1
≥ 0, λ

2
≥ 0 such that
2
c

= λ

1
, A
1

2
A
2
. (4.4)
As c varies, the optimal decision will change. We can see from our analysis that the situation is as
follows (see Figure 4.1):
2
Although this statement is intuitively obvious, its generalization to n dimensions is a deep theorem known as Farkas’
lemma (see Section 2).
30 CHAPTER 4. LINEAR PROGRAMMING
P
x

= Q
K

π

R
A
4
O
A
3
A
2
c
A
1
Figure 4.2: K

is the cone generated by Ω at x

.
1. x

= Qis optimal iff c lies between A
1
and A
2
iff c

= λ

1
A
1

2
A
2
for some λ

1
≥ 0, λ

2

0,
2. x

∈ QP is optimal iff c lies along A
2
iff c

= λ

2
A
2
for some λ

2
≥ 0,
3. x

= P is optimal iff c lies between A
3
and A
2
iff c

= λ

2
A
2

3
A
3
for some λ

2
≥ 0, λ

3

0, etc.
These statements can be made in a more elegant way as follows:
x

∈ Ω is optimal iff there exists λ

i
≥ 0 , 1 ≤ i ≤ 4, such that
(a) c

=
4
¸
i=1
λ

i
a
i
, (b) if A
i
x

< b
i
then λ

i
= 0 . (4.5)
For purposes of application it is useful to separate those constraints which are of the form x
i
≥ 0,
from the rest, and to reformulate (4.5) accordingly We leave this as an exercise.
Exercise 2: Show that (4.5) is equivalent to (4.6), below. (Here A
i
= (a
i1
, a
i2
).) x

∈ Ω is optimal
iff there exist λ

1
≥ 0 , λ

2
≥ 0 such that
(a) c
i
≤ λ

1
a
1i

2
a
2i
, i = 1, 2,
(b) if a
j1
x

1
+a
j2
x

2
< b
j
then x

j
= 0, j = 1, 2.
(c) if c
i
< λ

1i

2
a
2i
then x

i
= 0, i = 1, 2.
(4.6)
4.1. THE LINEAR PROGRAMMING PROBLEM 31
4.1.2 Problem formulation.
A linear programming problem (or LP in brief) is any decision problem of the form 4.7.
Maximize c
1
x
1
+c
2
x
2
+. . . +c
n
x
n
subject to
a
il
x
1
+a
i2
x
2
+. . . +a
in
x
n
≤ b
i
, l ≤ i ≤ k ,
a
il
x
1
+. . . . . . . . . +a
in
x
n
≥ b
i
, k + 1 ≤ i ≤ ,
a
il
x
1
+. . . . . . . . . +a
in
x
n
= b
i
, + 1 ≤ i ≤ m ,
and
x
j
≥ 0 , 1 ≤ j ≤ p ,
x
j
≥ 0 , p + 1 ≤ j ≤ q;
x
j
arbitary , q + 1 ≤ j ≤ n ,
(4.7)
where the c
j
, a
ij
, b
i
are ﬁxed real numbers.
There are two important special cases:
Case I: (4.7) is of the form (4.8):
Maximize
n
¸
j=1
c
j
x
j
subject to
n
¸
j=1
a
ij
x
j
≤ b
i
,
x
j
≥ 0 ,
1 ≤ i ≤ m ,
1 ≤ j ≤ n
(4.8)
Case II: (4.7) is of the form (4.9):
Maximize
n
¸
j=1
c
j
x
j
subject to
n
¸
j=1
a
ij
x
j
= b
i
,
x
j
≥ 0 ,
1 ≤ i ≤ m ,
1 ≤ j ≤ n .
(4.9)
Although (4.7) appears to be more general than (4.8) and (4.9), such is not the case.
Proposition: Every LP of the form (4.7) can be transformed into an equivalent LP of the form (4.8).
Proof.
Step 1: Replace each inequality constraint
¸
a
ij
x
j
≥ b
i
by
¸
(−a
ij
)x
j
≤ (−b
i
).
Step 2: Replace each equality constraint
¸
a
ij
x
j
= b
i
by two inequality constraints:
¸
a
ij
x
j
≤ b
i
,
¸
(−a
ij
)x
j
≤ (−b
i
).
Step 3: Replace each variable x
j
which is constrained x
j
≤ 0 by a variable y
j
= −x
j
constrained
y
j
≥ 0 and then replace a
ij
x
j
by (−a
ij
)y
j
for every i and c
j
x
j
by (−c
j
)y
j
.
32 CHAPTER 4. LINEAR PROGRAMMING
Step 4: Replace each variable x
j
which is not constrained in sign by a pair of variables
y
j
−z
j
= x
j
constrained y
j
≥ 0, z
j
≥ 0 and then replace a
ij
x
j
by a
ij
y
j
+ (−a
ij
)z
j
for every i and
c
j
x
j
by c
j
y
j
+ (−c
j
)z
j
. Evidently the resulting LP has the form (4.8) and is equivalent to the
original one. ♦
Proposition: Every LP of the form (4.7) can be transformed into an equivalent LP of the from (4.9)
Proof.
Step 1: Replace each inequality constraint
¸
a
ij
x
j
≤ b
i
by the equality constraint
¸
a
ij
x
j
+y
i
= b
i
where y
i
is an additional variable constrained y
i
≥ 0.
Step 2: Replace each inequality constraint
¸
a
ij
x
j
≥ b
i
by the equality constraint
¸
a
ij
x
j
−y
i
= b
i
where y
i
is an additional variable constrained by y
i
≥ 0. (The new variables
added in these steps are called slack variables.)
Step 3, Step 4: Repeat these steps from the previous proposition. Evidently the new LP has the
form (4.9) and is equivalent to the original one. ♦
4.2 Qualitative Theory of Linear Programming
4.2.1 Main results.
We begin by quoting a fundamental result. For a proof the reader is referred to (Mangasarian
[1969]).
Farkas’ Lemma. Let A
i
, 1 ≤ i ≤ k, be n-dimensional row vectors. Let c ∈ R
n
be a column vector.
The following statements are equivalent:
(i) for all x ∈ R
n
, A
i
x ≤ 0 for 1 ≤ i ≤ k implies c

x ≤ 0,
(ii) there exists λ
1
≥ 0, . . . , λ
k
≥ 0 such that c

=
k
¸
i=1
λ
i
A
i
.
An algebraic version of this result is sometimes more convenient.
Farkas’ Lemma (algebraic version). Let Abe a kn matrix. Let c ∈ R
n
. The following statements
are equivalent.
(i) for all x ∈ R
n
, Ax ≤ 0 implies c

x ≤ 0,
(ii) there exists λ ≥ 0, λ ∈ R
k
, such that A

λ = c.
Using this result it is possible to derive the main results following the intuitive reasoning of (4.1).
We leave this development as two exercises and follow a more elegant but less intuitive approach.
Exercise 1: With the same hypothesis and notation of Exercise 1 in 4.1, use the ﬁrst version of
Farkas

lemma to show that there exist λ

i
≥ 0 for i ∈ I(x

) such that
¸
i∈I(x

)
λ

i
A
i
= c

.
Exercise 2: Let x

satisfy the constraints for problem (4.17). Use the previous exercise to show
that x

is optimal iff there exist λ

1
≥ 0, . . . , λ

m
≥ 0 such that
(a) c
j

m
¸
i=1
λ

i
a
ij
, 1 ≤ j ≤ n
(b) if
n
¸
j=1
a
ij
x

j
< b
i
then λ

i
= 0 , 1 ≤ i ≤ m (c) if
m
¸
i=1
λ

i
a
ij
> c
j
then x

j
= 0 , 1 ≤ j ≤ m.
In the remaining discussion, c ∈ R
n
, b ∈
n
are ﬁxed vectors, and A = ¦a
ij
¦ is a ﬁxed m n
matrix, whereas x ∈ R
n
and λ ∈ R
m
will be variable. Consider the pair of LPs (4.10) and (4.11)
4.2. QUALITATIVE THEORY OF LINEAR PROGRAMMING 33
below. (4.10) is called the primal problem and (4.11) is called the dual problem.
Maximize
subject to
c
1
x
1
+. . . +c
n
x
n
a
i1
x
1
+. . . +a
in
x
n
≤ b
i
,
x
j
≥ 0 ,
1 ≤ i ≤ m
1 ≤ j ≤ n .
(4.10)
Maximize
subject to
λ
1
b
1
+. . . +λ
m
b
m
λ
1
a
1j
+. . . +λ
m
a
mj
≥ c
j
,
λ
i
≥ 0 ,
1 ≤ j ≤ n
1 ≤ i ≤ m .
(4.11)
Deﬁnition: Let Ω
p
= ¦x ∈ R
n
[Ax ≤ b, x ≥ 0¦ be the set of all points satisfying the constraints
of the primal problem. Similarly let Ω
d
= ¦λ ∈ R
m

A ≥ c

, λ ≥ 0¦. A point x ∈ Ω
p
(λ ∈ Ω
d
) is
said to be a feasible solution or feasible decision for the primal (dual).
The next result is trivial.
Lemma 1: (Weak duality) Let x ∈ Ω
p
, λ ∈ Ω
d
. Then
c

x ≤ λ

Ax ≤ λ

b. (4.12)
Proof: x ≥ 0 and λ

A − c

≥ 0 implies (λ

A−c

)x ≥ 0 giving the ﬁrst inequality. b−Ax ≥ 0 and
λ

≥ 0 implies λ

(b−Ax) ≥ 0 giving the second inequality. ♦
Corollary 1: If x

∈ Ω and λ

∈ Ω
d
such that c

x

= (λ

)

b, then x

is optimal for (4.10) and λ

is
optimal for (4.11).
Theorem 1: (Strong duality) Suppose Ω
p
= φ and Ω
d
= φ. Then there exists x

which is optimum
for (4.10) and λ

which is optimum for (4.11). Furthermore, c

x

= (λ

)

b.
Proof: Because of the Corollary 1 it is enough to prove the last statement, i.e., we must show that
there exist x ≥ 0, λ ≥ 0, such that Ax ≤ b, A

λ ≥ c and b

λ−c

x ≤ 0. By introducing slack
variables y ∈ R
m
, µ ∈ R
m
, r ∈ R, this is equivalent to the existence of x ≥ 0, y ≥ 0, λ ≥ 0, µ ≤
0, r ≤ 0 such that

A I
m
A

−I
n
−c

b

1
¸
¸

x
y
λ
µ
r
¸
¸
¸
¸
¸
¸
=

b
c
0
¸
¸
By the algebraic version of Farkas’ Lemma, this is possible only if
A

ξ −cθ ≤ 0 , ξ ≤ 0 ,
Aw = bθ ≤ 0 , −w ≤ 0 ,
θ ≤ 0
(4.13)
implies
b

ξ +c

w ≤ 0. (4.14)
34 CHAPTER 4. LINEAR PROGRAMMING
Case (i): Suppose (w, ξ, θ) satisﬁes (4.13) and θ < 0. Then (ξ/θ) ∈ Ω
d
, (w/−θ) ∈ Ω
p
, so that by
Lemma 1 c

w/(−θ) ≤ b

ξ/θ, which is equivalent to (4.14) since θ < 0.
Case (ii): Suppose (w, ξ, θ) satisﬁes (4.13) and θ = 0, so that −A

ξ ≥ 0, −ξ ≥ 0, Aw ≤ 0, w ≥ 0.
By hypothesis, there exist x ∈ Ω
p
, λ ∈ Ω
d
. Hence, −b

ξ = b

(−ξ) ≥ (Ax)

(−ξ) = x

(−A

ξ) ≥ 0,
and c

w ≤ (A

λ)

w = λ

(Aw) ≤ 0. So that b

ξ +c

w ≤ 0. ♦
The existence part of the above result can be strengthened.
Theorem 2: (i) Suppose Ω
p
= φ. Then there exists an optimum decision for the primal LP iff

d
= φ.
(ii) Suppose Ω
d
= φ. Then there exists an optimum decision for the dual LP iff Ω
p
= φ.
Proof Because of the symmetry of the primal and dual it is enough to prove only (i). The
sufﬁciency part of (i) follows from Theorem 1, so that only the necessity remains. Suppose, in
d
= φ. We will show that sup ¦c

x[x ∈ Ω
p
¦ = +∞. Now, Ω
d
= φ means
there does not exist λ ≥ 0 such that A

λ ≥ c. Equivalently, there does not exist λ ≥ 0, µ ≤ 0 such
that
¸
A

[
[
−I
n

λ
−−−
µ
¸
¸
=

c

By Farkas’ Lemma there exists w ∈ R
n
such that Aw ≤ 0, −w ≤ 0, and c

w > 0. By hypothesis,

p
= φ, so there exists x ≥ 0 such that Ax ≤ b. but then for any θ > 0, A(x + θw) ≤ b,
(x + θw) ≥ 0, so that (x + θw) ∈ Ω
p
. Also, c

(x + θw) = c

x + θc

w. Evidently then, sup
¦c

x[x ∈ Ω
p
¦ = +∞so that there is no optimal decision for the primal. ♦
Remark: In Theorem 2(i), the hypothesis that Ω
p
= φ is essential. Consider the following exercise.
Exercise 3: Exhibit a pair of primal and dual problems such that neither has a feasible solution.
Theorem 3: (Optimality condition) x

∈ Ω
p
is optimal if and only if there exists λ

∈ Ω
d
such that
m
¸
j=1
a
ij
x

j
< b
i
implies λ

i
= 0 ,
and
m
¸
i=1
λ

i
a
ij
< c
j
implies x

j
= 0 .
(4.15)
((4.15) is known as the condition of complementary slackness.)
Proof: First of all we note that for x

∈ Ω
p
, λ

∈ Ω
d
, (4.15) is equivalent to (4.16):

)

(Ax

−b) = 0, and (A

λ

−c)

x

= 0 . (4.16)
Necessity. Suppose x

∈ Ω
p
is optimal. Then from Theorem 2, Ω
d
= φ, so that by Theorem 1
there exists λ

∈ Ω
d
such that c

x

= (λ

)

b. By Lemma 1 we always have
c

x

≤ (λ

)

Ax

≤ (λ

)

b so that we must have c

x

= (λ

)

Ax

= (λ

)

b. But (4.16) is just an
equivalent rearrangement of these two equalities.
Sufﬁciency. Suppose (4.16) holds for some x

∈ Ω
p
, λ

∈ Ω
d
. The ﬁrst equality in (4.16) yields

)

b = (λ

)

Ax

= (A

λ

)

x

, while the second yields (A

λ

)

x

= c

x

, so that c

x

= (λ

)

b.
By Corollary 1, x

is optimal. ♦
4.2. QUALITATIVE THEORY OF LINEAR PROGRAMMING 35
The conditions x

∈ Ω
p
, x

∈ Ω
d
in Theorem 3 can be replaced by the weaker x

≥ 0, λ

≥ 0
provided we strengthen (4.15) as in the following result, whose proof is left as an exercise.

≥ 0 is optimal for the primal if and only if there exists λ

≥ 0 such
that
L(x, λ

) ≤ L(x

, λ

) ≤ L(x

, λ) for all x ≥ 0, and allλ ≥ 0, (4.17)
where L: R
n
xR
m
→ R is deﬁned by
L(x, λ) = c

x −λ

(Ax −b) (4.18)
Exercise 4: Prove Theorem 4.
Remark. The function L is called the Lagrangian. A pair (x

, λ

) satisfying (4.17) is said to form
a saddle-point of L over the set ¦x[x ∈ R
n
, x ≥ 0¦ ¦λ[λ ∈ R
m
, λ ≥ 0¦.
4.2.2 Results for problem (4.9).
It is possible to derive analogous results for LPs of the form (4.9). We state these results as exercises,
indicating how to use the results already obtained. We begin with a pair of LPs:
Maximize
subject to
c
1
x
1
+. . . +c
n
x
n
a
il
x
1
+. . . +a
in
x
n
= b
i
,
x
j
≥ 0 ,
1 ≤ i ≤ m ,
1 ≤ j ≤ n .
(4.19)
Minimize
subject to
λ
1
b
1
+. . . +λ
m
b
m
λ
1
a
1j
+. . . +λ
m
a
mj
≥ c
j
, 1 ≤ j ≤ n .
(4.20)
Note that in (4.20) the λ
i
are unrestricted in sign. Again (4.19) is called the primal and (4.20) the
dual. We let Ω
p
, Ω
d
denote the set of all x, λ satisfying the constraints of (4.19), (4.20) respectively.
Exercise 5: Prove Theorems 1 and 2 with Ω
p
and Ω
d
interpreted as above. (Hint. Replace (4.19)
by the equivalent LP: maximize c

x, subject to Ax ≤ b, (−A)x ≤ (−b), x ≥ 0. This is now of the
form (4.10). Apply Theorems 1 and 2.)
Exercise 6: Show that x

∈ Ω
p
is optimal iff there exists λ

∈ Ω
d
such that
x

j
> 0 implies
m
¸
i=1
λ

i
a
ij
= c
j
.
Exercise 7: x

≥ 0 is optimal iff there exists λ

∈ R
m
such that
L(x, λ

) ≤ L(x

, λ

) ≤ L(x

, λ) for all x ≥ 0, λ ∈ R
m
.
where L is deﬁned in (4.18). (Note that, unlike (4.17), λ is not restricted in sign.)
Exercise 8: Formulate a dual for (4.7), and obtain the result analogous to Exercise 5.
36 CHAPTER 4. LINEAR PROGRAMMING
4.2.3 Sensitivity analysis.
We investigate how the maximum value of (4.10) or (4.19) changes as the vectors b and c change.
The matrix Awill remain ﬁxed. Let Ω
p
and Ω
d
be the sets of feasible solutions for the pair (4.10) and
(4.11) or for the pair (4.19) and (4.20). We write Ω
p
(b) and Ω
d
(c) to denote the explicit dependence
on b and c respectively. Let B = ¦b ∈ R
m
[Ω
p
(b) = φ¦ and C = ¦c ∈ R
n
[Ω
d
(c) = φ¦, and for
(b, c) ∈ B C deﬁne
M(b, c) = max ¦c

x[x ∈ Ω
p
(b)¦ = min ¦λ

b[λ ∈ Ω
d
(c)¦ . (4.21)
For 1 ≤ i ≤ m, ε ∈ R, b ∈ R
m
denote
b(i, ε) = (b
1
, b
2
, . . . , b
i−1
, b
i
+ε, b
i+1
, . . . , b
m
)

,
and for 1 ≤ j ≤ n, ε ∈ R, c ∈ R
n
denote
c(j, ε) = (c
1
, c
2
, . . . , c
j−1
, c
j
+ε, c
j+1
, . . . , c
n
)

.
We deﬁne in the usual way the right and left hand partial derivatives of M at a point (
ˆ
b, ˆ c) ∈ BC
as follows:
∂M
+
∂b
i
(
ˆ
b, ˆ c) = lim
ε → 0
ε > 0
1
ε
¦M(
ˆ
b(i, ε), ˆ c) −M(
ˆ
b, ˆ c)¦ ,
∂M

∂b
i
(
ˆ
b, ˆ c) = lim
ε → 0
ε > 0
1
ε
¦M(
ˆ
b, ˆ c) −M(
ˆ
b(i, −ε), ˆ c)¦ ,
∂M
+
∂c
j
(
ˆ
b, ˆ c) = lim
ε → 0
ε > 0
1
ε
¦M(
ˆ
b, ˆ c(j, ε)) −M(
ˆ
b, ˆ c¦ ,
∂M

∂c
j
(
ˆ
b, ˆ c) = lim
ε → 0
ε > 0
1
ε
¦M(
ˆ
b, ˆ c −M(
ˆ
b, ˆ c(j, −ε))¦ ,
Let

B,

C denote the interiors of B, C respectively.
Theorem 5: At each (
ˆ
b, ˆ c) ∈

B

C, the partial derivatives given above exist. Furthermore, if
ˆ x ∈ Ω
p
(
ˆ
b),
ˆ
λ ∈ Ω
d
(ˆ c) are optimal, then
∂M
+
∂b
i
(
ˆ
b, ˆ c) ≤
ˆ
λ
i

∂M

∂b
i
(
ˆ
b, ˆ c) , 1 ≤ i ≤ m , (4.22)
4.3. THE SIMPLEX ALGORITHM 37
∂M
+
∂c
j
(
ˆ
b, ˆ c) ≥ ˆ x
j

∂M

∂c
j
(
ˆ
b, ˆ c) , 1 ≤ j ≤ n , (4.23)
Proof: We ﬁrst show (4.22), (4.23) assuming that the partial derivatives exist. By strong duality
M(
ˆ
b, ˆ c) =
ˆ
λ

ˆ
b, and by weak duality M(
ˆ
b(i, ε), ˆ c) ≤
ˆ
λ

ˆ
b(i, ε), so that
1
ε
¦M(
ˆ
b(i, ε), ˆ c) −M(
ˆ
b, ˆ c)¦ ≤
1
ε
ˆ
λ

¦
ˆ
b(i, ε) −
ˆ

ˆ
λ
i
, for ε > 0,
1
ε
¦M(
ˆ
b, ˆ c) −M(
ˆ
b(i, −ε), ˆ c)¦ ≥
1
ε
ˆ
λ

¦
ˆ
b −
ˆ
b(i, −ε)¦ =
ˆ
λ
i
, for ε > 0.
Taking limits as ε → 0, ε > 0, gives (4.22).
On the other hand, M(
ˆ
b, ˆ c) = ˆ c

ˆ x, and M(
ˆ
b, ˆ c(j, ε)) ≥ (ˆ c(j, ε))

ˆ x, so that
1
ε
¦M(
ˆ
b, ˆ c(j, ε)) −M(
ˆ
b, ˆ c)¦ ≥
1
ε
¦ˆ c(j, ε)

− ˆ c¦

ˆ x = ˆ x
j
, for ε > 0,
1
ε
¦M(
ˆ
b, ˆ c) −M(
ˆ
b, ˆ c(j, −ε))¦ ≤
1
ε
¦ˆ c − ˆ c(j, −ε)¦

ˆ x = ˆ x
j
, for ε > 0,
which give (4.23) as ε → 0, ε > 0.
Finally, the existence of the right and left partial derivatives follows from Exercises 8, 9 below. ♦
We recall some fundamental deﬁnitions from convex analysis.
Deﬁnition: X ⊂ R
n
is said to be convex if x, y ∈ X and 0 ≤ θ ≤ 1 implies (θx+(1−θ)y) ∈ X.
Deﬁnition: Let X ⊂ R
n
and f : X → R. (i) f is said to be convex if X is convex, and x, y ∈ X,
0 ≤ θ ≤ 1 implies f(θx + (1 − θ)y) ≤ θf(x) + (1 − θ)f(y). (ii) f is said to be concave if −f is
convex, i.e., x, y ∈ X, 0 ≤ θ ≤ 1 implies f(θx + (1 −θ)y) ≥ θf(x) + (1 −θ)f(y).
Exercise 8: (a) Show that Ω
p
, Ω
d
, and the sets B ⊂ R
m
, C ⊂ R
n
deﬁned above are convex sets.
(b) Show that for ﬁxed c ∈ C, M(, c) : B → R is concave and for ﬁxed b ∈ B, M(b, ) : C → R
is convex.
Exercise 9: Let X ⊂ R
n
, and f : X → R be convex. Show that at each point ˆ x in the interior of
X, the left and right hand partial derivatives of f exist. (Hint: First show that for
ε
2
> ε
1
> 0 > δ
1
> δ
2
,(1/ε
2
)¦f(ˆ x(i, ε
2
)) −f(ˆ x)¦ ≥ (1/ε
1
)¦f(ˆ x(i, ε
1
)) −f(ˆ x))¦ ≥
(1/δ
1
)¦f(ˆ x(i, δ
1
)) −f(ˆ x))¦ ≥ (1/δ
2
)¦f(ˆ x(i, δ
2
)) −f(ˆ x)¦. Then the result follows immediately.)
Remark 1: Clearly if (∂M/∂b
i
)(
ˆ
b) exists, then we have equality in (4.22), and then this result
compares with 3.14).
Remark 2: We can also show without difﬁculty that M(, c) and M(b, ) are piecewise linear (more
accurately, linear plus constant) functions on B and C respectively. This is useful in some
computational problems.
Remark 3: The variables of the dual problem are called Lagrange variables or dual variables or
shadow-prices. The reason behind the last name will be clear in Section 4.
4.3 The Simplex Algorithm
4.3.1 Preliminaries
We now present the celebrated Simplex algorithm for ﬁnding an optimum solution to any LP of the
form (4.24):
Maximize
subject to
c
1
x
1
+. . . +c
n
x
n
a
il
x
1
+. . . +a
in
x
n
= b
i
,
x
j
≥ 0 ,
1 ≤ i ≤ m
1 ≤ j ≤ n .
(4.24)
38 CHAPTER 4. LINEAR PROGRAMMING
As mentioned in 4.1 the algorithm rests upon the observations that if an optimal exists, then at least
one vertex of the feasible set Ω
p
is an optimal solution. Since Ω
p
has only ﬁnitely many vertices (see
Corollary 1 below), we only have to investigate a ﬁnite set. The practicability of this investigation
depends on the ease with which we can characterize the vertices of Ω
p
. This is done in Lemma 1.
In the following we let A
j
denote the jth column of A, i.e., A
j
= (a
1j
, . . . , a
mj
)

. We begin with
a precise deﬁnition of a vertex.
Deﬁnition: x ∈ Ω
p
is said to be a vertex of Ω
p
if x = λy + (1 −λ)z, with y, z in Ω
p
and
0 < λ < 1, implies x = y = z.
Deﬁnition: For x ∈ Ω
p
, let I(x) = ¦j[x
j
> 0¦.
Lemma 1: Let x ∈ Ω
p
. Then x is a vertex of Ω
p
iff ¦A
j
[j ∈ I(x)¦ is a linearly independent set.
Exercise 1: Prove Lemma 1.
Corollary 1: Ω
p
has at most
m
¸
j=1
n!
(n −j)!
vertices.
Lemma 2: Let x

be an optimal decision of (4.24). Then there is a vertex z

of Ω
p
which is optimal.
Proof: If ¦A
j
[j ∈ I(x

)¦ is linearly independent, let z

= x

and we are done. Hence suppose
¦A
j
[j ∈ I(x

)¦ is linearly dependent so that there exist γ
j
, not all zero, such that
¸
j∈I(x

)
γ
j
A
j
= 0 .
For θ ∈ R deﬁne z(θ) ∈ R
n
by
z
j
(θ) =

x

j
= θγ
j
,
x

j
= 0 ,
j ∈ I(x

)
j ∈ I(x

) .
Az(θ) =
¸
j∈I(x

)
z
j
(θ)A
j
=
¸
j∈I(x

)
x

j
A
j

¸
j∈I(x

)
γ
j
A
j
= b +θ 0 = b .
Since x

j
> 0 for j ∈ I(x

), it follows that z(θ) ≥ 0 when
[θ[ ≤ min

x

j

j
|

j ∈ I(x

)
¸
= θ

say .
Hence z(θ) ∈ Ω
p
whenever [θ[ ≤ θ

. Since x

is optimal we must have
c

x

≥ c

z(θ) = c

x

¸
j∈I(x

)
c
j
y
j
for −

θ ≤ θ ≤ θ

.
Since θ can take on positive and negative values, the inequality above can hold on if
¸
J∈I(x

)
c
j
γ
j
=
0, and then c

x

= c

z(θ), so that z(θ) is also an optimal solution for [θ[ ≤ θ

. But from the
deﬁnition of z(θ) it is easy to see that we can pick θ
0
with [θ
0
[ = θ

such that z
j

0
) = x

j

0
γ
j
= 0
for at least one j = j
0
in I(x

). Then,
I(z(θ
0
)) ⊂ I(x

) −¦j
0
¦ .
4.3. THE SIMPLEX ALGORITHM 39
Again, if ¦A
j
[j ∈ I(z(θ
0
))¦ is linearly independent, then we let z

= z(θ
0
) and we are done.
Otherwise we repeat the procedure above with z(θ
0
). Clearly, in a ﬁnite number of steps we will
ﬁnd an optimal decision z

which is also vertex. ♦
At this point we abandon the geometric term “vertex” and how to established LP terminology.
Deﬁnition: (i) z is said to be a basic feasible solution if z ∈ Ω
p
, and ¦A
j
[j ∈ I(z)¦ is linearly
independent. The set I(z) is then called the basis at z, and x
j
, j ∈ I(z), are called the basic
variables at z. x
j
, j ∈ I(z) are called the non-basic variables at z.
Deﬁnition: A basic feasible solution z is said to be non-degenerate if I(z) has m elements.
Notation: Let z be a non-degenerate basic feasible solution, and let j
1
< j
2
< . . . < j
m
constitute I(z). Let D(z) denote the m m non-singular matrix D(z) = [A
j
1
.
.
.A
j
2
.
.
. . . .
.
.
.A
jm
], let
c(z) denote the m-dimensional column vector c(z) = (c
j
1
, . . . , c
jm
)

and deﬁne λ(z) by λ

(z) =
c

(z)[D(z)]
−1
. We call λ(z) the shadow-price vector at z.
Lemma 3: Let z be a non-degenerate basic feasible solution. Then z is optimal if and only if
λ

(z)A ≥ c
j
, for all , j ∈ I(z) . (4.25)
Proof: By Exercise 6 of Section 2.2, z is optimal iff there exists λ such that
λ

A
j
= c
j
, for , j ∈ I(z) , (4.26)
λ

A
j
≥ c
j
, for , j ∈ I(z) , (4.27)
But since z is non-degenerate, (4.26) holds iff λ = λ(z) and then (4.27) is the same as (4.25). ♦
4.3.2 The Simplex Algorithm.
The algorithm is divided into two parts: In Phase I we determine if Ω
p
is empty or not, and if not,
we obtain a basic feasible solution. Phase II starts with a basic feasible solution and determines if
it is optimal or not, and if not obtains another basic feasible solution with a higher value. Iterating
on this procedure, in a ﬁnite number of steps, either we obtain an optimum solution or we discover
that no optimum exists, i.e., sup ¦c

x[x ∈ Ω
p
¦ = +∞. We shall discuss Phase II ﬁrst.
We make the following simplifying assumption. We will comment on it later.
Assumption of non-degeneracy. Every basic feasible solution is non-degenerate.
Phase II:
Step 1. Let z
0
be a basic feasible solution obtained from Phase I or by any other means. Set k = 0
and go to Step 2.
Step 2. Calculate [D(z
k
)]
−1
,c(z
k
), and the shadow-price vector λ

(z
k
) = c

(z
k
)[D(z
k
)]
−1
. For
each j ∈ I(z
k
) calculate c
j
−λ

(z
k
)A
j
. If all these numbers are ≤ 0, stop, because z
k
is optimal
by Lemma 3. Otherwise pick any
ˆ
j ∈ I(z
k
) such that c
ˆ
j
−λ

(z
k
)A
ˆ
j
> 0 and go to Step 3.
Step 3. Let I(z
k
) consist of j
1
< j
2
< . . . < j
m
. Compute the vector
γ
k
= (γ
k
j
1
, . . . γ
k
jm
)

= [D(z
k
)]
−1
A
ˆ
j
. If γ
k
≤ 0, stop, because by Lemma 4 below, there is no
ﬁnite optimum. Otherwise go to Step 4.
Step 4. Compute θ = min ¦(z
k
j
γ
k
j
)[j ∈ i(z), γ
k
j
> 0¦. Evidently 0 < θ < ∞. Deﬁne z
k+1
by
40 CHAPTER 4. LINEAR PROGRAMMING
z
k+1
j
=

z
k
j
−θγ
k
j
θ
z
k
j
= 0
,
,
,
j ∈ I(z)
j =
ˆ
j
j =
ˆ
j and j ∈ I(z) .
(4.28)
By Lemma 5 below, z
k+1
is a basic feasible solution with c

z
k+1
> c

z
k
. Set k = k + 1 and return
to Step 2.
Lemma 4: If γ
k
≤ 0, sup ¦c

x[x ∈ Ω
p
¦ = ∞.
Proof: Deﬁne z(θ) by
z
j
(θ) =

z
j
−θγ
k
j
θ
z
j
= 0
,
,
,
j ∈ I(z)
j =
ˆ
j
j ∈ I(z) and j =
ˆ
j .
(4.29)
First of all, since γ
k
≤ 0 it follows that z(θ) ≥ 0 for θ ≥ 0. Next, Az(θ) = Az − θ
¸
j∈I(z)
γ
k
j
A
j
+
θA
ˆ
j
= Az by deﬁnition of γ
k
. Hence, z(θ) ∈ Ω
p
for θ ≥ 0. Finally,
c

z(θ)
= c

z −θc

(z
k

k
+θc
ˆ
j
= c

z +θ¦c
ˆ
j
−c

(z
k
)[D(z
k
)]
−1
A
ˆ
j
¦
= c

z +θ¦c
ˆ
j
−λ

(z
k
)A
ˆ
j
¦i .
(4.30)
But from step 2 ¦c
ˆ
j −λ

(z
k
)A
ˆ
j
¦ > 0, so that c

z(θ) → ∞as θ → ∞. ♦
Lemma 5: z
k+1
is a basic feasible solution and c

z
k+1
> c

z
k
.
Proof: Let
˜
j ∈ I(z
k
) be such that γ
k
˜
j
> 0 and z
k
˜
j
= θγ
k
˜
j
. Then from (4.28) we see that z
k+1
˜
j
= 0,
hence
I(z
k+1
) ⊂ (I(z) −¦
˜
j¦)
¸
¦
ˆ
j¦ , (4.31)
so that it is enough to prove that A
˜
j
is independent of ¦A
j
[j ∈ I(z), j =
˜
j¦. But if this is not the
case, we must have γ
k
˜
j
= 0, giving a contradiction. Finally if we compare (4.28) and (4.29), we see
from (4.30) that
c

z
k+1
−c

z
k
= θ¦c
ˆ
j
−γ

(z
k
)A
ˆ
j
¦ ,
which is positive from Step 2. ♦
Corollary 2: In a ﬁnite number of steps Phase II will obtain an optimal solution or will determine
that sup¦c

x[x ∈ Ω
p
¦ = ∞.
Corollary 3: Suppose Phase II terminates at an optimal basic feasible solution z

. Then γ(z

) is an
optimal solution of the dual of (4.24).
Exercise 2: Prove Corollaries 2 and 3.
Remark 1: By the non-degeneracy assumption, I(z
k+1
) has m elements, so that in (4.31) we must
have equality. We see then that D(z
k+1
) is obtained from D(z
k
) by replacing the column A
j
by
4.3. THE SIMPLEX ALGORITHM 41
the column A
ˆ
j
. More precisely if D(z
k
) = [A
j
1
.
.
. . . .
.
.
.A
j
i−1
.
.
.A
˜
j
.
.
.A
j
i+1
.
.
. . . .
.
.
.A
jm
] and if
j
k
<
ˆ
j < j
k+1
then D(z
k+1
) = [A
j
1
.
.
. . . .
.
.
.A
j
i−1
.
.
.A
j
i+1
.
.
. . . .
.
.
.A
j
k
.
.
.A
ˆ
j
.
.
.A
j
k+1
.
.
. . . .
.
.
.A
jm
]. Let E be the
matrix E = [A
j
1
.
.
. . . .
.
.
.A
j
i−1
.
.
.A
ˆ
j
.
.
.A
j
i+1
.
.
. . . .
.
.
.A
jm
]. Then [D(z
k+1
)]
−1
= P E
−1
where the matrix P
permutes the columns of D(z
k+1
) such that E = D(z
k+1
)P. Next, if A
ˆ
j
=
m
¸
=1
γ
j

A
j

, it is easy
to check that E
−1
= M[D(z
k
)]
−1
where
M =

1
1
.
.
.
1
−γ
j
1
γ
˜
j
1
γ
˜
j
−γ
jm
γ
˜
j
1
.
.
.
1
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸

ith column
Then [D(z
k+1
)]
−1
= PM[D(z
k
)]
−1
, so that these inverses can be easily computed.
Remark 2: The similarity between Step 2 of Phase II and Step 2 of the algorithm in 3.3.4 is
striking. The basic variables at z
k
correspond to the variables w
k
and non-basic variables
correspond to u
k
. For each j ∈ I(z
k
) we can interpret the number c
j
−λ

(z
k
)A
j
to be the net
increase in the objective value per unit increase in the jth component of z
k
. This net increase is due
to the direct increase c
j
minus the indirect decrease λ

(z
k
)A
j
due to the compensating changes in
the basic variables necessary to maintain feasibility. The analogous quantity in 3.3.4 is
(∂f
0
/∂u
j
)(x
k
) −(λ
k
)

(∂f/∂u
j
)(x
k
).
Remark 3: By eliminating any dependent equations in (4.24) we can guarantee that the matrix A
has rank n. Hence at any degenerate basic feasible solution z
k
we can always ﬁnd
¯
I(z
k
) ⊃ I(z
k
)
such that
¯
I(z
k
) has m elements and ¦A
j
[j ∈
¯
I(z
k
)¦ is a linearly independent set. We can apply
Phase II using
¯
I(z
k
k
). But then in Step 4 it may turn out that θ = 0 so that
z
k+1
= z
k
. The reason for this is that
¯
I(z
k
) is not unique, so that we have to try various
alternatives for
¯
I(z
k
) until we ﬁnd one for which θ > 0. In this way the non-degeneracy
assumption can be eliminated. For details see (Canon, et al., [1970]).
We now describe how to obtain an initial basic feasible solution.
Phase I:
Step I. by multiplying some of the equality constraints in (4.24) by −1 if necessary, we can assume
that b ≥ 0. Replace the LP (4.24) by the LP (4.32) involving the variables x and y:
Maximize −
m
¸
i=1
y
i
subject to a
il
x
1
+. . . +a
in
x
n
+y
i
= b
i
, 1 ≤ i ≤ m ,
x
j
≥ 0 , y
i
≥ 0 , 1 ≤ j ≤ n , 1 ≤ i ≤ m .
(4.32)
42 CHAPTER 4. LINEAR PROGRAMMING
Go to step 2.
Step 2. Note that (x
0
, y
0
) = (0, b) is a basic feasible solution of (4.32). Apply phase II to (4.32)
starting with this solution. Phase II must terminate in an optimum based feasible solution (x

, y

)
since the value of the objective function in (4.32) lies between −
m
¸
i=1
b
i
and 0. Go to Step 3.
Step 3. If y

= 0, x

is a basic feasible solution for (4.24). If y

= 0, by Exercise 3 below, (4.24)
has no feasible solution.
Exercise 3: Show that (4.24) has a feasible solution iff y

= 0.
4.4 LP Theory of a Firm in a Competitive Economy
4.4.1 Activity analysis of the ﬁrm.
We think of a ﬁrm as a system which transforms input into outputs. There are m kinds of inputs
and k kinds of outputs. Inputs are usually classiﬁed into raw materials such as iron ore, crude oil,
or raw cotton; intermediate products such as steel, chemicals, or textiles; capital goods
3
such as
machines of various kinds, or factory buildings, ofﬁce equipment, or computers; ﬁnally various
kinds of labor services. The ﬁrm’s outputs themselves may be raw materials (if it is a mining
company) or intermediate products (if it is a steel mill) or capital goods (if it manufactures lathes)
or ﬁnished goods (if it makes shirts or bakes cookies) which go directly to the consumer. Labor is
not usually considered an output since slavery is not practiced; however, it may be considered an
output in a “closed,” dynamic Malthusian framework where the increase in labor is a function of the
output. (See the von Neumann model in (Nikaido [1968]), p. 141.)
Within the ﬁrm, this transformation can be conducted in different ways, i.e., different combina-
tions of inputs can be used to produce the same combination of outputs, since human labor can
do the same job as some machines and machines can replace other kinds of machines, etc. This
substitutability among inputs is a fundamental concept in economics. We formalize it by specifying
which transformation possibilities are available to the ﬁrm.
By an input vector we mean any m-dimensional vector r = (r
1
, . . . , r
m
)

with r ≥ 0, and by an
output vector we mean any k-dimensional vector y = (y
1
, . . . , y
k
)

with y ≥ 0. We now make three
(i) The transformation of inputs into outputs is organized into a ﬁnite number, say n, of processes
or activities.
(ii) Each activity combines the k inputs in ﬁxed proportions into the m outputs in ﬁxed propor-
tions. Furthermore, each activity can be conducted at any non-negative intensity or level. Pre-
cisely, the jth activity is characterized completely by two vectors A
j
= (a
1j
, a
2j
, . . . , a
mj
)

and
B
j
= (b
ij
, . . . , b
kj
)

so that if it is conducted at a level x
j
≥ 0, then it combines (transforms) the
input vector (a
1j
x
j
, . . . , a
mj
x
j
)

= x
j
A
j
into the output vector (b
1j
x
j
, . . . , b
kj
x
j
)

= x
j
B
j
. Let
A be the mn matrix [A
1
.
.
. . . .
.
.
.A
n
] and B be the k n matrix B = [B
1
.
.
. . . .
.
.
.B
n
].
3
It is more accurate to think of the services of capital goods rather than these goods themselves as inputs. It is these
services which are consumed in the transformation into outputs.
4.4. LP THEORY OF A FIRM IN A COMPETITIVE ECONOMY 43
(iii) If the ﬁrmconducts all the activities simultaneously with the jth activity at level x
j
≥ 0, 1 ≤ j ≤
n, then it transforms the input vector x
1
A
1
+. . . +x
n
A
n
into the output vector x
1
B
1
+. . . +x
n
B
n
.
With these assumptions we know all the transformations technically possible as soon as we spec-
ify the matrices A and B. Which of these possible transformations will actually take place depends
upon their relative proﬁtability and availability of inputs. We study this next.
4.4.2 Short-term behavior.
In the short-term, the ﬁrm cannot change the amount available to it of some of the inputs such as
capital equipment, certain kinds of labor, and perhaps some raw materials. Let us suppose that these
inputs are 1, 2, . . . , and they are available in the amounts r

1
, . . . , r

, whereas the supply of the
remaining inputs can be varied. We assume that the ﬁrm is operating in a competitive economy
which means that the unit prices p = (p
1
, . . . , p
k
)

of the outputs, and q = (q
1
, . . . , q
m
)

of the
inputs is ﬁxed. Then the manager of the ﬁrm, if he is maximizing the ﬁrm’s proﬁts, faces the
following decision problem:
Maximize p

y −
m
¸
j=+1
q
j
r
j
subject to y = Bx,
a
i1
x
1
+. . . +a
in
x
n
≤ r

i
, 1 ≤ i ≤ ,
a
i1
x
1
+. . . +a
in
x
n
≤ r
i
, + 1 ≤ i ≤ m ,
x
j
≥ 0, 1 ≤ j ≤ n; r
i
≥ 0 , + 1 ≤ i ≤ m .
(4.33)
The decision variables are the activity levels x
1
, . . . , x
n
, and the short-term input supplies r
+1
, . . . , r
m
.
The coefﬁcients of B and Aare the ﬁxed technical coefﬁcients of the ﬁrm, the r

i
are the ﬁxed short-
term supplies, whereas the p
i
, q
j
are prices determined by the whole economy, which the ﬁrm ac-
cepts as given. Under realistic conditions (4.33) has an optimal solution, say, x

1
, . . . , x

n
, r

+1
, . . . , r

m
.
4.4.3 Long-term equilibrium behavior.
In the long run the supplies of the ﬁrst inputs are also variable and the ﬁrm can change these
supplies from r

1
, . . . , r

by buying or selling these inputs at the market price q
1
, . . . , q

. Whether
the ﬁrm will actually change these inputs will depend upon whether it is proﬁtable to do so, and in
turn this depends upon the prices p, q. We say that the prices (p

, q

) and a set of input supplies
r

= (r

1
, . . . , r

m
) are in (long-term) equilibrium if the ﬁrm has no proﬁt incentive to change r

under the prices (p

, q

).
Theorem 1: p

, q

, r

are in equilibrium if and only if q

is an optimal solution of (4.34):
Minimize (r

)

q
subject to A

q ≥ B

p

q ≥ 0 .
(4.34)
Proof: Let c = B

p

. By deﬁnition, p

, q

, r

are in equilibrium iff for all ﬁxed ∆ ∈ R
m
,
M(∆) ≤ M(0) where M(∆) is the maximum value of the LP (4.35):
Maximize c

x −(q

)

subject to Ax ≤ r

+ ∆ ,
x ≥ 0 .
(4.35)
44 CHAPTER 4. LINEAR PROGRAMMING
For ∆ = 0, (4.34) becomes the dual of (4.35) so that by the strong duality theorem, M(0) = (r

)

q

.
Hence p

, q

, r

are in equilibrium iff
c

x −(q

)

∆ ≤ M(0) = (r

)

q

, (4.36)
whenever x is feasible for (4.35). By weak duality if x is feasible for (4.35) and q is feasible for
(4.34),
c

x −(q

)

∆ ≤ q

(r

= ∆) −(q

)

∆ , (4.37)
and, in particular, for q = q

,
c

x −(q

)

∆ ≤ (q

)

(r

+ ∆) −(q

)

∆ = (q

)

r

Remark 1: We have shown that (p

, q

, r

are in long-term equilibrium iff q

is an optimum
solution to the dual (namely (4.34)) of (4.38):
Maximize c

x
subject to Ax ≤ r

x ≥ 0 .
(4.38)
This relation between p

, q

, r

has a very nice economic interpretation. Recall that c = B

p

, i.e.,
c
j
= p

1
b
1j
+p

2
b
2j
+. . . +p

k
b
kj
. Now b
ij
is the amount of the ith output produced by operating the
jth activity at a unit level x
j
= 1. Hence, c
j
is the revenue per unit level operation of the jth activity
so that c

x is the revenue when the n activities are operated at levels x. On the other hand if the jth
activity is operated at level x
j
= 1, it uses an amount a
ij
of the ith input. If the ith input is valued at
a

i
, then the input cost of operating at x
j
= 1, is
m
¸
i=1
q
i
a
ij
, so that the input cost of operating the n
activities at levels x is (A

q

)

= (q

)

Ax. Thus, if x

is the optimum activity levels for (4.38) then
the output revenue is c

x

and the input cost is (q

)

Ax

. But from (4.16), (q

)

(Ax

−r

) = 0 so
that
c

x

= (q

)

r

, (4.39)
i.e., at the optimum activity levels, in equilibrium, total revenues = total cost of input supplies. In
fact, we can say even more. From (4.15) we see that if x
=
ast
j
> 0 then
c
j
=
m
¸
i=1
q

i
a
ij
,
i.e., at the optimum, the revenue of an activity operated at a positive level = input cost of that activity.
Also if
c
j
<
m
¸
i=1
q

i
a
ij
,
then x

j
= 0, i.e., if the revenue of an activity is less than its input cost, then at the optimum it is
operated at zero level. Finally, again from (4.15), if an equilibrium the optimum ith input supply r

i
is greater than the optimum demand for the ith input,
4.4. LP THEORY OF A FIRM IN A COMPETITIVE ECONOMY 45
r

i
>
n
¸
j=1
a
ij
x

j
,
then q

i
= 0, i.e., the equilibrium price of an input which is in excess supply must be zero, in other
words it must be a free good.
Remark 2: Returning to the short-term decision problem (4.33), suppose that

1
, . . . , λ

, λ

+1
, . . . , λ

m
) is an optimum solution of the dual of (4.33). Suppose that the market
prices of inputs 1, . . . , are q
1
, . . . , q

. Let us denote by M(∆
1
, . . . , ∆

) the optimum value of
(4.33) when the amounts of the inputs in ﬁxed supply are r

1
+ ∆
1
, . . . , r

+ ∆

. Then if
(∂M/∂∆
i
)[
∆=0
exists, we can see from (4.22) that it is always proﬁtable to increase the ith input
i
if λ

i
> q
i
, and conversely it is proﬁtable to sell some
of the ith input at price q
i
if λ

i
< q
i
. Thus λ

i
can be interpreted as the ﬁrm’s internal valuation of
the ith input or the ﬁrm’s imputed or shadow price of the ith input. This interpretation has wide
applicability, which we mention brieﬂy. Often engineering design problems can be formulated as
LPs of the form (4.10) or (4.19), where some of the coefﬁcients b
i
are design parameters. The
design procedure is to ﬁx these parameters at some nominal value b

i
, and carry out the
optimization problem. Suppose the resulting optimal dual variables are λ

i
. then we see (assuming
differentiability) that it is worth increasing b

i
if the unit cost of increasing this parameter is less
than λ

i
, and it is worth decreasing this parameter if the reduction in total cost per unit decrease is
greater than λ

i
.
4.4.4 Long-term equilibrium of a competitive, capitalist economy.
The proﬁt-maximizing behavior of the ﬁrm presented above is one of the two fundamental building
blocks in the equilibrium theory of a competitive, capitalist economy. Unfortunately we cannot
present the details here. We shall limit ourselves to a rough sketch. We think of the economy as
a feedback process involving ﬁrms and consumers. Let us suppose that there are a total of h com-
modities in the economy including raw materials, intermediate and capital goods, labor, and ﬁnished
products. By adding zero rows to the matrices (A, B) characterizing a ﬁrm we can suppose that all
the h commodities are possible inputs and all the h commodities are possible outputs. Of course,
for an individual ﬁrm most of the inputs and most of the outputs will be zero. the sole purpose for
making this change is that we no longer need to distinguish between prices of inputs and prices of
outputs. We observe the economy starting at time T. At this time there exists within the economy
an inventory of the various commodities which we can represent by a vector ω = (ω
1
, . . . , ω
h
) ≥ 0.
ω is that portion of the outputs produced prior to T which have not been consumed up to T. We are
assuming that this is a capitalist economy, which means that the ownership of ω is divided among
the various consumers j = 1, . . . , J. More precisely, the jth consumer owns the vector of commodi-
ties ω(j) ≥ 0, and
J
¸
j=1
ω(j) = ω. We are including in ω(j) the amount of his labor services which
consumer j is willing to sell. Now suppose that at time T the prevailing prices of the h commodities
are λ = (λ
1
, . . . , λ
h
)

≥ 0. Next, suppose that the managers of the various ﬁrms assume that the
prices λ are not going to change for a long period of time. Then, from our previous analysis we
know that the manager of the ith ﬁrm will plan to buy input supplies r(i) ≥ 0, r(i) ∈ R
h
, such
46 CHAPTER 4. LINEAR PROGRAMMING
that (λ, r(i)) is in long term equilibrium, and he will plan to produce an optimum amount, say y(i).
Here i = 1, 2, . . . , I, where I is the total number of ﬁrms. We know that r(i) and y(i) depend on
λ, so that we explicitly write r(i, λ), y(i, λ). We also recall that (see (4.38))
λ

r(i, λ) = λ

y(i, λ) , 1 ≤ i ≤ I . (4.40)
Now the ith manager can buy r(i) from only two sources: outputs from other ﬁrms, and the con-
sumers who collectively own ω. Similarly, the ith manager can sell his planned output y(i) either as
input supplies to other ﬁrms or to the consumers. Thus, the net supply offered for sale to consumers
is S(λ), where
S(λ) =
J
¸
j=1
ω(j) +
I
¸
i=1
y(i, λ) −
i
¸
i=1
r(i, λ) . (4.41)
We note two important facts. First of all, from (4.40), (4.41) we immediately conclude that
λ

S(λ) =
J
¸
j=1
λ

ω(j) , (4.42)
that is the value of the supply offered to consumers is equal to the value of the commodities (and
labor) which they own. The second point is that there is no reason to expect that S(λ) ≥ 0.
Now we come to the second building block of equilibrium theory. The value of the jth consumer’s
possessions is λ

ω(j). The theory assumes that he will plan to buy a set of commodities d(j) =
(d
1
(j), . . . , d
h
(j)) ≥ 0 so as to maximize his satisfaction subject to the constraint λ

d(j) = λ

ω(j).
Here also d(j) will depend on λ, so we write d(j, λ). If we add up the buying plans of all the
consumers we obtain the total demand
D(λ) =
J
¸
j=1
d(j, λ) ≥ 0 , (4.43)
which also satisﬁes
λ

D(λ) =
J
¸
j=1
λ

ω(j) . (4.44)
The most basic question of equilibrium theory is to determine conditions under which there exists a
price vector λ
E
such that the economy is in equilibrium, i.e., S(λ
E
) = D(λ
E
), because if such an
equilibrium price λ
E
exists, then at that price the production plans of all the ﬁrms and the buying
plan of all the consumers can be realized. Unfortunately we must stop at this point since we cannot
proceed further without introducing some more convex analysis and the ﬁxed point theorem. For
a simple treatment the reader is referred to (Dorfman, Samuelson, and Solow [1958], Chapter 13).
For a much more general mathematical treatment see (Nikaido [1968], Chapter V).
4.5.1 Some mathematical tricks.
It is often the case in practical decision problems that the objective is not well-deﬁned. There may
be a number of plausible objective functions. In our LP framework this situation can be formulated
as follows. The constraints are given as usual by Ax ≤ b, x ≥ 0. However, there are, say, k
objective functions (c
1
)

x, . . . , (c
k
)

x. It is reasonable then to deﬁne a single objective function
f
0
(x) by f
0
(x) = minimum ¦(c
1
)

x, (c
2
)

x, . . . , (c
k
)

x¦, so that we have the decision problem,
Maximize f
0
(x)
subject to Ax ≤ b, x ≥ 0 .
(4.45)
This is not a LP since f
0
is not linear. However, the following exercise shows how to transform
(4.45) into an equivalent LP.
Exercise 1: Show that (4.45) is equivalent to (4.46) below, in the sense that x

is optimal for (4.45)
iff (x

, y

) = (x

, f
0
(x

)) is optimal for (4.46).
Maximize y
subject to Ax ≤ b, x ≤ 0
y ≤ (c
i
)

x , 1 ≤ i ≤ k .
(4.46)
Exercise 1 will also indicate how to do Exercise 2.
Exercise 2: Obtain an equivalent LP for (4.47):
Maximize
n
¸
j=1
c
i
(x
i
)
subject to Ax ≤ b, x ≤ 0 ,
(4.47)
where c
i
: R → R are concave, piecewise-linear functions of the kind shown in Figure 4.3.
The above-given assumption of the concavity of the c
i
is crucial. In the next exercise, the inter-
pretation of “equivalent” is purposely left ambiguous.
Exercise 3: Construct an example of the kind (4.47), where the c
i
are piecewise linear (but not
concave), and such that there is no equivalent LP.
It turns out however, that even if the c
i
are not concave, an elementary modiﬁcation of the Simplex
algorithm can be given to obtain a “local” optimal decision. See (Miller [1963]).
4.5.2 Scope of linear programming.
LP is today the single most important optimization technique. This is because many decision prob-
lems can be adequately formulated as LPs, and, given the capabilities of modern computers, the
Simplex method (together with its variants) is an extremely powerful technique for solving LPs in-
volving thousands of variables. To obtain a feeling for the scope of LP we refer the reader to the
book by one of the originators of LP (Dantzig [1963]).
48 CHAPTER 4. LINEAR PROGRAMMING
.
.
.
c
i
(x
i
)
x
i
Figure 4.3: A function of the form used in Exercise 2.
Chapter 5
OPTIMIZATION OVER SETS
DEFINED BY INEQUALITY
CONSTRAINTS: NONLINEAR
PROGRAMMING
In many decision-making situations the assumption of linearity of the constraint inequalities in LP
is quite restrictive. The linearity of the objective function is not restrictive as shown in the ﬁrst
exercise below. In Section 1 we present the general nonlinear programming problem (NP) and
prove the Kuhn-Tucker theorem. Section 2 deals with Duality theory for the case where appropriate
convexity conditions are satisﬁed. Two applications are given. Section 3 is devoted to the important
special case of quadratic programming. The last section is devoted to computational considerations.
5.1 Qualitative Theory of Nonlinear Programming
5.1.1 The problem and elementary results.
The general NP is a decision problem of the form:
Maximize f
0
(x)
subject to (x) ≤ 0 , i = 1, . . . , m,
(5.1)
where x ∈ R
n
, f
i
: R
n
→ R, i = 0, 1, . . . , m, are differentiable functions. As in Chapter 4,
x ∈ R
n
is said to be a feasible solution if it satisﬁes the constraints of (5.1), and Ω ⊂ R
n
is the
subset of all feasible solutions; x

∈ Ω is said to be an optimal decision or optimal solution if
f
0
(x

) ≥ f
0
(x) for x ∈ Ω. From the discussion in 4.1.2 it is clear that equality constraints and sign
constraints on some of the components of x can all be transformed into the form (5.1). The next
exercise shows that we could restrict ourselves to objective functions which are linear; however, we
will not do this.
Exercise 1: Show that (5.2), with variables y ∈ R, x ∈ R
n
, is equivalent to (5.1):
49
50 CHAPTER 5. NONLINEAR PROGRAMMING
Maximize y
subject to f
i
(x) ≤ 0, 1 ≤ i ≤ m, and y −f
0
(x) ≤ 0 .
(5.2)
Returning to problem (5.1), we are interested in obtaining conditions which any optimal decision
must satisfy. The argument parallels very closely that developed in Exercise 1 of 4.1 and Exercise 1
of 4.2. The basic idea is to linearize the functions f
i
in a neighborhood of an optimal decision x

.
Deﬁnition: Let x be a feasible solution, and let I(x) ⊂ ¦1, . . . , m¦ be such that f
i
(x) = 0 for
ı ∈ I(x), f
i
(x) < 0 for i ∈ I(x). (The set I(x) is called the set of active constraints at x.)
Deﬁnition: (i) Let x ∈ Ω. A vector h ∈ R
n
is said to be an admissible direction for Ω at x if there
exists a sequence x
k
, k = 1, 2, . . . , in Ω and a sequence of numbers ε
k
, k = 1, . . . , with ε
k
> 0
for all k such that
lim
k→∞
x
k
= x ,
lim
k→∞
1
ε
k
(x
k
−x) = h .
(ii) Let C(Ω, x) = ¦h[h is an admissible direction for Ω at x¦. C(Ω, x) is called the tangent cone
of Ω at x. Let K(Ω, x) = ¦x + h[h ∈ C(Ω, x)¦. (See Figures 5.1 and 5.2 and compare them with
Figures 4.1 and 4.2.)
If we take x
k
= x and ε
k
= 1 for all k, we see that 0 ∈ C(Ω, x) so that the tangent cone is always
nonempty. Two more properties are stated below.
Exercise 2: (i) Show that C(Ω, x) is a cone, i.e., if h ∈ C(Ω, x) and θ ≥ 0, then θh ∈ C(Ω, x).
(ii) Show that C(Ω, x) is a closed subset of R
n
. (Hint for (ii): For m = 1, 2, . . . , let h
m
and
¦x
mk
, ε
mk
> 0¦

k=1
be such that x
mk
→ x and (1/ε
mk
)(x
mk
−x) → h
m
as k → ∞. Suppose
that h
m
→ h as m → ∞. Show that there exist subsequences ¦x
mkm
, ε
mkm
¦

m=1
such that
x
mkm
→ x and (1/ε
mkm
)(x
mkm
−x) → h as m → ∞.)
In the deﬁnition of C(Ω, x) we made no use of the particular functional description of Ω. The
following elementary result is more interesting in this light and should be compared with (2.18) in
Chapter 2 and Exercise 1 of 4.1.
Lemma 1: Suppose x

∈ Ω is an optimum decision for (5.1).
Then
f
0x
(x

)h ≤ 0 for all h ∈ C(Ω, x

) . (5.3)
Proof: Let x
k
∈ Ω, ε
k
> 0, k = 1, 2, 3, . . . , be such that
5.1. QUALITATIVE THEORY OF NONLINEAR PROGRAMMING 51
,
P
¦x[f
3
(x) = 0¦
Q
x

direction of
increasing
payoff
π(k) =
¦x[f
0
(x) = k¦
¦x[f
1
(x) = 0¦

R
¦x[f
2
(x) = 0¦
Figure 5.1: Ω = PQR
lim
k→∞
x
k
= x

,
lim
k→∞
1
ε
k
(x
k
−x

) = h . (5.4)
Note that in particular (5.4) implies
lim
k→∞
1
ε
k
[x
k
−x

[ = [h[ . (5.5)
Since f
0
is differentiable, by Taylor’s theorem we have
f
0
(x
k
) = f
0
(x

+ (x
k
−x

)) = f
0
(x

) +f
0x
(x

)(x
k
−x

) +o([x
k
−x

[) . (5.6)
Since x
k
∈ Ω, and x

is optimal, we have f
0
(x
k
) ≤ f
0
(x

), so that
0 ≥ f
0x
(x

)
(x
k
−x

)
ε
k
+
o(|x
k
−x

|)
ε
k
.
Taking limits as k → ∞, using (5.4) and (5.5), we can see that
0 ≥
=
lim
k→∞
f
0x
(x

)h. ♦
f
0x
(x

)
(x
k
−x

)
ε
k
+
lim
k→∞
o(|x
k
−x

|)
|x
k
−x

| lim
k→∞
|x
k
−x

|
ε
k
52 CHAPTER 5. NONLINEAR PROGRAMMING
-
-
-
-
-
-
-
-
x

K(Ω, x

)
0
C(Ω, x

)
Figure 5.2: C(Ω, x

) is the tangent cone of Ω at x

.
The basic problem that remains is to characterize the set C(Ω, x

) in terms of the derivatives of the
functions f
i
. Then we can apply Farkas’ Lemma just as in Exercise 1 of 4.2.
Lemma 2: Let x

∈ Ω. Then
C(Ω, x

) ⊂ ¦h[f
ix
(x

)h ≤ 0 for all i ∈ I(x

)¦ . (5.7)
Proof: Let h ∈ R
n
and x
k
∈ Ω, ε
k
> 0, k = 1, 2, . . . , satisfy (5.4). Since f
i
is differentiable, by
Taylor’s theorem we have
f
i
(x
k
) = f
i
(x

) +f
ix
(x

)(x
k
−x

) +o([x
k
−x

[) .
Since x
k
∈ Ω, f
i
(x
k
) ≤ 0, and if i ∈ I(x

), f
i
(x

) = 0, so that f
i
(x
k
) ≤ f
i
(x

). Following the
proof of Lemma 1 we can conclude that 0 ≥ f
ix
(x

)h. ♦
Lemma 2 gives us a partial characterization of C(Ω, x

). Unfortunately, in general the inclusion
sign in (5.7) cannot be reversed. The main reason for this is that the set ¦f
ix
(x

)[i ∈ I(x

)¦ is not
in general linearly independent.
Exercise 3: Let x ∈ R
2
, f
1
(x
1
, x
2
) = (x
1
−1)
3
+x
2
, and f
2
(x
1
, x
2
) = −x
2
. Let
(x

1
, x

2
) = (1, 0). Then I(x

) = ¦1, 2¦. Show that
5.1. QUALITATIVE THEORY OF NONLINEAR PROGRAMMING 53
C(Ω, x

) = ¦h[f
ix
(x

)h ≤ 0 , i = 1, 2, ¦.
(Note that ¦f
1x
(x

), f
2x
(x

)¦ is not a linearly independent set; see Lemma 4 below.)
5.1.2 Kuhn-Tucker Theorem.
Deﬁnition: Let x

∈ Ω. We say that the constraint qualiﬁcation (CQ) is satisﬁed at x

if
C(Ω, x) = ¦h[f
ix
(x

)h ≤ 0 for all i ∈ I(x

)¦,
and we say that CQ is satisﬁed if CQ is satisﬁed at all x ∈ Ω. (Note that by Lemma 2 C(Ω, x) is
always a subset of the right-hand side.)
Compare the next result with Exercise 2 of 4.2.
Theorem 1: (Kuhn and Tucker [1951]) Let x

be an optimum solution of (5.1), and suppose that
CQ is satisﬁed at x

. Then there exist λ

i
≥ 0, for i ∈ I(x

), such that
f
0x
(x

) =
¸
i∈I(x

)
λ

i
f
ix
(x

)
(5.8)
Proof: By Lemma 1 and the deﬁnition of CQ it follows that f
0x
(x

)h ≤ 0 whenever f
ix
(x

)h ≤ 0
for all i ∈ I(x

). By the Farkas’ Lemma of 4.2.1 it follows that there exist λ

i
≥ 0 for i ∈ I(x

)
such that (5.8) holds. ♦
In the original formulation of the decision problem we often have equality constraints of the form
r
j
(x) = 0, which get replaced by r
j
(x) ≤ 0, −r
j
(x) ≤ 0 to give the form (5.1). It is convenient in
application to separate the equality constraints from the rest. Theorem 1 can then be expressed as
Theorem 2.
Theorem 2: Consider the problem (5.9).
Maximize f
0
(x)
subject to f
i
(x) ≤ 0 , i = 1, . . . , m,
r
j
(x) = 0 , j = 1, . . . , k .
(5.9)
Let x

be an optimum decision and suppose that CQ is satisﬁed at x

. Then there exist λ

i
≥ 0, i =
1, . . . , m, and µ

j
, j = 1, . . . , k such that
f
0x
(x

) =
m
¸
i=1
λ

i
f
ix
(x

) +
k
¸
j=1
µ

j
r
jx
(x

) , (5.10)
and
λ

i
= 0 whenever f
i
(x

) < 0 . (5.11)
Exercise 4: Prove Theorem 2.
54 CHAPTER 5. NONLINEAR PROGRAMMING
An alternative form of Theorem 1 will prove useful for computational purposes (see Section 4).
Theorem 3: Consider (5.9), and suppose that CQ is satisﬁed at an optimal solution x

. Deﬁne
ψ : R
n
→ R by
ψ(h) = max ¦−f
0x
(x

)h, f
1
(x

) +f
1x
(x

)h, . . . , f
m
(x

) +f
mx
(x

)h¦ ,
and consider the decision problem
Minimize ψ(h)
subject to −ψ(h) −f
0x
(x

)h ≤ 0,
−ψ(h) +f
i
(x

) +f
ix
(x

)h ≤ 0 , 1 ≤ i ≤ m
−1 ≤ h
i
≤ 1 , i = 1, . . . , n .
(5.12)
Then h = 0 is an optimal solution of (5.12).
Exercise 5: Prove Theorem 3. (Note that by Exercise 1 of 4.5, (5.12) can be transformed into a
LP.)
Remark: For problem (5.9) deﬁne the Lagrangian function L:
(x
1
, . . . , x
n
; λ
1
, . . . , λ
m
; µ
1
, . . . , µ
k
) → f
0
(x) −
m
¸
i=1
λ
i
f
i
(x) −
k
¸
j=1
µ
j
r
j
(x).
Then Theorem 2 is equivalent to the following statement: if CQ is satisﬁed and x

is optimal, then
there exist λ

≥ 0 and µ

such that L
x
(x

, λ

, µ

) = 0 and L(x

, λ

, µ

) ≤ L(x

, λ, µ) for all
λ ≥ 0, µ.
There is a very important special case when the necessary conditions of Theorem 1 are also
sufﬁcient. But ﬁrst we need some elementary properties of convex functions which are stated as an
exercise. Some additional properties which we will use later are also collected here.
Recall the deﬁnition of convex and concave functions in 4.2.3.
Exercise 6: Let X ⊂ R
n
be convex. Let h : X → R be a differentiable function. Then
(i) h is convex iff h(y) ≥ h(x) +h
x
(x)(y −x) for all x, y, in X,
(ii) h is concave iff h(y) ≤ h(x) +h
x
(x)(y −x) for all x, y in X,
(iii) h is concave and convex iff h is afﬁne, i.e. h(x) ≡ α +b

x for some
ﬁxed α ∈ R, b ∈ R
n
.
Suppose that h is twice differentiable. Then
(iv) h is convex iff h
xx
(x) is positive semideﬁnite for all x in X,
(v) h is concave iff h
xx
(x) is negative semideﬁnite for all x in X,
(vi) h is convex and concave iff h
xx
(x) ≡ 0.
Theorem 4: (Sufﬁcient condition) In (5.1) suppose that f
0
is concave and f
i
is convex for
i = 1, . . . , m. Then
(i) Ω is a convex subset of R
n
, and
(ii) if there exist x

∈ Ω, λ

i
≥ 0, i ∈ I(x

), satisfying (5.8), then x

is an optimal solution of
(5.1).
Proof:
(i) Let y, z be in Ω so that f
i
(y) ≤ 0, f
i
(z) ≤ 0 for i = 1, . . . , m. Let 0 ≤ θ ≤ 1. Since f
i
is
convex we have
5.1. QUALITATIVE THEORY OF NONLINEAR PROGRAMMING 55
f
i
(θy + (1 −θ)z) ≤ θf
i
(y) + (1 −θ)f
i
(z) ≤ 0 , 1 ≤ i ≤ m,
so that (θy + (1 −θ)z) ∈ Ω, hence Ω is convex.
(ii) Let x ∈ Ω be arbitrary. Since f
0
is concave, by Exercise 6 we have
f
0
(x) ≤ f
0
(x

) +f
0x
(x

)(x −x

) ,
so that by (5.8)
f
0
(x) ≤ f
0
(x

) +
¸
i∈I(x

)
λ

i
f
ix
(x

)(x −x

) .
(5.13)
Next, f
i
is convex so that again by Exercise 6,
f
i
(x) ≥ f
i
(x

) +f
ix
(x

)(x −x

) ;
but f
i
(x) ≤ 0, and f
i
(x

) = 0 for i ∈ I(x

), so that
f
ix
(x

)(x −x

) ≤ 0 for i ∈ I(x

) . (5.14)
Combining (5.14) with the fact that λ

i
≥ 0, we conclude from (5.13) that f
0
(x) ≤ f
0
(x

), so that
x

is optimal. ♦
Exercise 7: Under the hypothesis of Theorem 4, show that the subset Ω

of Ω, consisting of all the
optimal solutions of (5.1), is a convex set.
Exercise 8: A function h : X → R deﬁned on a convex set X ⊂ R
n
is said to be strictly convex if
h(θy + (1 −θ)z) < θh(y) + (1 −θ)h(z) whenever 0 < θ < 1 and y, z are in X with y = z. h is
said to be strictly concave if −h is strictly convex. Under the hypothesis of Theorem 4, show that
an optimal solution to (5.1) is unique (if it exists) if either f
0
is strictly concave or if the
f
i
, 1 ≤ i ≤ m, are strictly convex. (Hint: Show that in (5.13) we have strict inequality if x = x

.)
5.1.3 Sufﬁcient conditions for CQ.
As stated, it is usually impractical to verify if CQ is satisﬁed for a particular problem. In this
subsection we give two conditions which guarantee CQ. These conditions can often be veriﬁed in
practice. Recall that a function g : R
n
→ R is said to be afﬁne if g(x) ≡ α + b

x for some ﬁxed
α ∈ R and b ∈ R
n
.
We adopt the formulation (5.1) so that
Ω = ¦x ∈ R
n
[f
i
(x) ≤ 0 , 1 ≤ i ≤ m¦ .
Lemma 3: Suppose x

∈ Ω and suppose there exists h

∈ R
n
such that for each i ∈ I(x

), either
f
ix
(x

)h

< 0, or f
ix
(x

)h

= 0 and f
i
is afﬁne. Then CQ is satisﬁed at x

.
Proof: Let h ∈ R
n
be such that f
ix
(x

)h ≤ 0 for i ∈ I(x

). Let δ > 0. We will ﬁrst show that
(h +δh

) ∈ C(Ω, x

). To this end let ε
k
> 0, k = 1, 2, . . . , be a sequence converging to 0 and set
x
k
= x

k
(h +δh

). Clearly x
k
converges to x

, and (1/ε
k
)(x
k
−x

) converges to (h +δh

).
Also for i ∈ I(x

), if f
ix
(x

)h < 0, then
f
i
(x
k
) = f
i
(x

) +ε
k
f
ix
(x

)(h +δh

) +o(ε
k
[h +δh

[)
≤ δε
k
f
ix
(x

)h

+o(ε
k
[h +δh

[)
< 0 for sufﬁciently large k ,
whereas for i ∈ I(x

), if f
i
is afﬁne, then
56 CHAPTER 5. NONLINEAR PROGRAMMING
f
i
(x
k
) = f
i
(x

) +ε
k
f
ix
(x

)(h +δh

) ≤ 0 for all k .
Finally, for i ∈ I(x

) we have f
i
(x

) < 0, so that f
i
(x
k
) < 0 for sufﬁciently large k. Thus we
have also shown that x
k
∈ Ω for sufﬁciently large k, and so by deﬁnition (h + δh

) ∈ C(Ω, x

).
Since δ > 0 can be arbitrarily small, and since C(Ω, x

) is a closed set by Exercise 2, it follows that
h ∈ C(Ω, x

). ♦
Exercise 9: Suppose x

∈ Ω and suppose there exists ˆ x ∈ R
n
such that for each i ∈ I(x

), either
f
i
(x

) < 0 and f
i
is convex, or f
i
(ˆ x) ≤ 0 and f
i
is afﬁne. Then CQ is satisﬁed at x

. (Hint: Show
that h

= ˆ x −x

satisﬁes the hypothesis of Lemma 3.)
Lemma 4: Suppose x

∈ Ω and suppose there exists h

∈ R
n
such that f
ix
(x

)h

≤ 0 for
i ∈ I(x

), and ¦f
ix
(x

)[i ∈ I(x

), f
ix
(x

)h

= 0¦ is a linearly independent set. Then CQ is
satisﬁed at x

.
Proof: Let h ∈ R
n
be such that f
ix
(x

)h ≤ 0 for all i ∈ I(x

). Let δ > 0. We will show that
(h +δh

) ∈ C(Ω, x

). Let J
δ
= ¦i[i ∈ I(x

), f
ix
(x

)(h +δh

) = 0¦, consist of p elements.
Clearly J
δ
⊂ J = ¦i[i ∈ I(x

), f
i
x(x

)h

= 0¦, so that ¦f
ix
(x

, u

)[i ∈ J
δ
¦ is linearly
independent. By the Implicit Function Theorem, there exist ρ > 0, an open set V ⊂ R
n
containing
x

= (w

, u

), and a differentiable function g : U → R
p
, where U = ¦u ∈ R
n−p
[[u −u

[ < ρ¦,
such that
f
i
(w, u) = 0, i ∈ J
δ
, and (w, u) ∈ V
iff
u ∈ U, and w = g(u) .
Next we partition h, h

as h = (ξ, η), h

= (ξ

, η

) corresponding to the partition of x = (w, u).
Let ε
k
> 0, k = 1, 2 . . . , be any sequence converging to 0, and set u
k
= u

+ ε
k
(η +δη

), w
k
=
g(u
k
), and ﬁnally x
k
= (s
k
, u
k
).
We note that u
k
converges to u

, so w
k
= g(u
k
) converges to w

= g(u

). Thus, x
k
converges
to x

. Now (1/ε
k
)(x
k
−x

) = (1/ε
k
)(w
k
−w

, u
k
−u

) = (1/ε
k
)(g(u
k
) −g(u

), ε
k
(η +δη

)).
Since g is differentiable, it follows that (1/ε
k
)(x
k
−x

) converges to (g
u
(u

)(η +δη

), η +δη

).
But for i ∈ J
δ
we have
0 = f
ix
(x

)(h +δh

) = f
iw
(x

)(ξ +δξ

) +f
iu
(x

)(η +δη

) . (5.15)
Also, for i ∈ J
δ
, 0 = f
i
(g(u), u) for u ∈ U so that 0 = f
iw
(x

)g
u
(u

) +f
iu
(x

), and hence
0 = f
iw
(x

)g
u
(u

)(η +δη

) +f
iu
(x

)(η +δη

) . (5.16)
If we compare (5.15) and (5.16) and recall that ¦f
iw
(x

)[i ∈ J
δ
¦ is a basis in R
p
we can conclude
that (ξ +δξ

) = g
u
(u

)(η +δη

) so that (1/ε
k
)(x
k
−x

) converges to (h +hδh

).
It remains to show that x
k
∈ Ω for sufﬁciently large k. First of all, for i ∈ J
δ
, f
i
(x
k
) =
f
i
(g(u
k
), u
k
) = 0, whereas for i ∈ J
δ
, i ∈ I(x

),
f
i
(x
k
) = f
i
(x

) +f
ix
(x

)(x
k
−x

) +o([x
k
−x

[)
f
i
(x

) +ε
k
f
ix
(x

)(h +δh

) +o(ε
k
) +o([x
k
−x

[),
5.2. DUALITY THEORY 57
and since f
i
(x

) = 0 whereas f
ix
(x

)(h + δh

) < 0, we can conclude that f
i
(x
k
) < 0 for sufﬁ-
ciently large k. Thus, x
k
∈ Ω for sufﬁciently large k. Hence, (h +δh

) ∈ C(Ω, x

).
To ﬁnish the proof we note that δ > 0 can be made arbitrarily small, and C(Ω, x

) is closed by
Exercise 2, so that h ∈ C(Ω, x

). ♦
The next lemma applies to the formulation (5.9). Its proof is left as an exercise since it is very
similar to the proof of Lemma 4.
Lemma 5: Suppose x

is feasible for (5.9) and suppose there exists h

∈ R
n
such that the set
¦f
ix
(x

)[i ∈ I(x

), f
ix
(x

)h

= 0¦
¸
¦r
jx
(x

)[j = 1, . . . , k¦ is linearly independent, and f
ix
(x

)h

0 for i ∈ I(x

), r
jx
(x

)h

= 0 for 1 ≤ j ≤ k. Then CQ is satisﬁed at x

.
Exercise 10: Prove Lemma 5
5.2 Duality Theory
Duality theory is perhaps the most beautiful part of nonlinear programming. It has resulted in many
applications within nonlinear programming, in terms of suggesting important computational algo-
rithms, and it has provided many unifying conceptual insights into economics and management
science. We can only present some of the basic results here, and even so some of the proofs are
relegated to the Appendix at the end of this Chapter since they depend on advanced material. How-
ever, we will give some geometric insight. In 2.3 we give some application of duality theory and in
2.2 we refer to some of the important generalizations. The results in 2.1 should be compared with
Theorems 1 and 4 of 4.2.1 and the results in 4.2.3.
It may be useful to note in the following discussion that most of the results do not require differ-
entiability of the various functions.
5.2.1 Basic results.
Consider problem (5.17) which we call the primal problem:
Maximize f
0
(x)
subject to f
i
(x) ≤
ˆ
b
i
, 1 ≤ i ≤ m
x ∈ X ,
(5.17)
where x ∈ R
n
, f
i
: R
n
→ R, 1 ≤ i ≤ m, are given convex functions, f
0
: R
n
→ R is a
given concave function, X is a given convex subset of R
n
and
ˆ
b = (
ˆ
b
1
, . . . ,
ˆ
b
m
)

is a given vector.
For convenience, let f = (f
1
, . . . , f
m
)

: R
n
→ R
m
. We wish to examine the behavior of the
maximum value of (5.17) as
ˆ
b varies. So we deﬁne
Ω(b) = ¦x[x ∈ X, f(x) ≤ b¦, B = ¦b[Ω(b) = φ¦,
and
M : B → R
¸
¦+∞¦ by M(b) = sup¦f
0
(x)[x ∈ X, f(x) ≤ b¦
= sup¦f
0
(x)[x ∈ Ω(b)¦ ,
so that in particular if x

is an optimal solution of (5.17) then M(
ˆ
b) = f
0
(ˆ x). We need to consider
the following problem also. Let λ ∈ R
m
, λ ≥ 0, be ﬁxed.
Maximize f
0
(x) −λ

(f(x) −
ˆ
b)
subject to x ∈ X ,
(5.18)
58 CHAPTER 5. NONLINEAR PROGRAMMING
and deﬁne
m(λ) = sub¦f
0
(x) −λ

(f(x) −
ˆ
b)[x ∈ X¦ .
Problem (5.19) is called the dual problem:
Minimize m(λ)
subject to λ ≥ 0 .
(5.19)
Let m

= inf ¦m(λ)[λ ≥ 0¦.
Remark 1: The set X in (5.17) is usually equal to R
n
and then, of course, there is no reason to
separate it out. However, it is sometimes possible to include some of the constraints in X in such
a way that the calculation of m(λ) by (5.18) and the solution of the dual problem (5.19) become
simple. For example see the problems discussed in Sections 2.3.1 and 2.3.2 below.
Remark 2: It is sometimes useful to know that Lemmas 1 and 2 below hold without any convexity
conditions on f
0
, f, X. Lemma 1 shows that the cost function of the dual problem is convex which
is useful information since there are computation techniques which apply to convex cost functions
but not to arbitrary nonlinear cost functions. Lemma 2 shows that the optimum value of the dual
problem is always an upper bound for the optimum value of the primal.
Lemma 1: m : R
n
+
→ R
¸
¦+∞¦ is a convex function. (Here R
n
+
= ¦λ ∈ R
n
[λ ≥ 0¦.)
Exercise 1: Prove Lemma 1.
Lemma 2: (Weak duality) If x is feasible for (5.17), i.e., x ∈ Ω(
ˆ
b), and if λ ≥ 0, then
f
0
(x) ≤ M(
ˆ
b) ≤ m

≤ m(λ) . (5.20)
Proof: Since f(x) −
ˆ
b ≤ 0, and λ ≥ 0, we have λ

(f(x) −
ˆ
b) ≤ 0. So,
f
0
(x) ≤ f
0
(x) −λ

(f(x) −
ˆ
b), for x ∈ Ω(
ˆ
b), λ ≥ 0 .
Hence
f
0
(x) ≤ sup ¦f
0
(x)[x ∈ Ω(
ˆ
b)¦ = M(
ˆ
b)
≤ sup ¦f
0
(x) −λ

(f(x) −
ˆ
b)[x ∈ Ω(
ˆ
b)¦ and since Ω(
ˆ
b) ⊂ X,
≤ sup ¦f
0
(x) −λ

(f(x) −
ˆ
b)[x ∈ X¦ = m(λ) .
Thus, we have
f
0
(x) ≤ M(
ˆ
b) ≤ m(λ) for x ∈ Ω(
ˆ
b), λ ≥ 0 ,
and since M(
ˆ
b) is independent of λ, if we take the inﬁmum with respect to λ ≥ 0 in the right-hand
inequality we get (5.20). ♦
The basic problem of Duality Theory is to determine conditions under which M(
ˆ
b) = m

in
(5.20). We ﬁrst give a simple sufﬁciency condition.
Deﬁnition: A pair (ˆ x,
ˆ
λ) with ˆ x ∈ X, and
ˆ
λ ≤ 0 is said to satisfy the optimality conditions if
5.2. DUALITY THEORY 59
ˆ x is optimal solution of (5.18) with λ =
ˆ
λ, (5.21)
ˆ x is feasible for (5.17), i.e., f
i
(ˆ x) ≤
ˆ
b
i
for i = 1, . . . , m , (5.22)
ˆ
λ
i
= 0 when f
i
(ˆ x) <
ˆ
b
i
, equivalently,
ˆ
λ

(f(ˆ x) −
ˆ
b) = 0. (5.23)
ˆ
λ ≥ 0 is said to be an optimal price vector if there is ˆ x ∈ X such that (ˆ x,
ˆ
λ) satisfy the optimality
condition. Note that in this case ˆ x ∈ Ω(
ˆ
b) by virtue of (5.22).
The next result is equivalent to Theorem 4(ii) of Section 1 if X = R
n
, and f
i
, 0 ≤ i ≤ m, are
differentiable.
Theorem 1: (Sufﬁciency) If (ˆ x,
ˆ
λ) satisfy the optimality conditions, then ˆ x is an optimal solution to
the primal,
ˆ
λ is an optimal solution to the dual, and M(
ˆ
b) = m

.
Proof: Let x ∈ Ω(
ˆ
b), so that
ˆ
λ

(f(x) −
ˆ
b) ≤ 0. Then
f
0
(x) ≤ f
0
(x) −
ˆ
λ

(f(x) −
ˆ
b)
≤ sup¦f
0
(x) −
ˆ
λ

(f(x) −
ˆ
b)[x ∈ X¦
= f
0
(ˆ x) −
ˆ
λ

(f(ˆ x) −
ˆ
b) by (5.21)
= f
0
(ˆ x) by (5.23)
so that ˆ x is optimal for the primal, and hence by deﬁnition f
0
(ˆ x) = M(
ˆ
b). Also
m(
ˆ
λ) = f
0
(ˆ x) −
ˆ
λ

(f(ˆ x) −
ˆ
b)
f
0
(ˆ x) = M(
ˆ
b) ,
so that from Weak Duality
ˆ
λ is optimal for the dual. ♦
We now proceed to a much more detailed investigation.
Lemma 3: B is a convex subset of R
m
, and M : B → R
¸
¦+∞¦ is a concave function.
Proof: Let b,
˜
b belong to B, let x ∈ Ω(b), ˜ x ∈ Ω(
˜
b), let 0 ≤ θ ≤ 1. Then (θx + (1 − θ)˜ x) ∈ X
since X is convex, and
f
i
(θx + (1 −θ)˜ x) ≤ θf
i
(x) + (1 −θ)f
i
(˜ x)
since f
i
is convex, so that
f
i
(θx + (1 −θ)˜ x) ≤ θb + (1 −θ)
˜
b , (5.24)
hence
(θx + (1 −θ)˜ x) ∈ Ω(θb + (1 −θ)
˜
b)
and therefore, B is convex.
Also, since f
0
is concave,
f
0
(θx + (1 −θ)˜ x) ≥ θf
0
(x) + (1 −θ)f
0
(˜ x) .
60 CHAPTER 5. NONLINEAR PROGRAMMING
Since (5.24) holds for all x ∈ Ω(b) and ˜ x ∈ Ω(
˜
b) it follows that
M(θb + (1 −θ)
ˆ
b) ≥ sup ¦f
0
(θx + (1 −θ)˜ x)[x ∈ Ω(b), ˜ x ∈ Ω(
˜
b)¦
≥ sup¦f
0
(x)[x ∈ Ω(b)¦ + (1 −θ) sup ¦f
0
(˜ x)[˜ x ∈ Ω(
˜
b)¦
= θM(b) + (1 −θ)M(
˜
b). ♦
Deﬁnition: Let X ⊂ R
n
and let g : X → R
¸
¦∞, −∞¦. A vector λ ∈ R
n
is said to be a
g(x) ≤ g(ˆ x) +λ

(x − ˆ x) for x ∈ X.
(g(x) ≥ g(ˆ x) +λ

(x − ˆ x) for x ∈ X.)
(See Figure 5-3.)
,
-
-
-
-
-
.
.
.
.
M(b)
M(
ˆ
b)
b ∈ B
ˆ
b b
M is not stable at
ˆ
b
M(b)
M(
ˆ
b)
ˆ
b
b
M is stable at
ˆ
b
M(b)
M(
ˆ
b) +λ

(b −
ˆ
b)
ˆ
b b
ˆ
b
Figure 5.3: Illustration of supergradient of stability.
Deﬁnition: The function M : B → R
¸
¦∞¦ is said to be stable at
ˆ
b ∈ B if there exists a real
number K such that
M(b) ≤ M(
ˆ
b) +K[b −
ˆ
b[ for b ∈ B .
(In words, M is stable at
ˆ
b if M does not increase inﬁnitely steeply in a neighborhood of
ˆ
b. See
Figure 5.3.)
A more geometric way of thinking about subgradients is the following. Deﬁne the subset A ⊂
R
1+m
by
5.2. DUALITY THEORY 61
A = ¦(r, b)[b ∈ B, and r ≤ M(b)¦ .
Thus A is the set lying ”below” the graph of M. We call A the hypograph
1
of M. Since M is
concave it follows immediately that A is convex (in fact these are equivalent statements).
Deﬁnition: A vector (λ
0
, λ
1
, . . . , λ
m
) is said to be the normal to a hyperplane supporting A at a
point(ˆ r,
ˆ
b) if
λ
0
ˆ r +
m
¸
i=1
λ
i
ˆ
b
i
≥ λ
0
r +
m
¸
i=1
λ
i
b
i
for all (r, b) ∈ A . (5.25)
(In words, Alies below the hyperplane ˆ π = ¦(r, b)[λ
0
r+
¸
λ
i
b
i
= λ
0
ˆ r+
¸
λ
i
b
i
¦.) The supporting
hyperplane is said to be non-vertical if λ
0
= 0. See Figure 5.4.
Exercise 2: Show that if
ˆ
b ∈ B,
˜
b ≥
ˆ
b, and ˜ r ≤ M(
ˆ
b), then
˜
b ∈ B, M(
˜
b), and (˜ r,
˜
b) ∈ A.
Exercise 3: Assume that
ˆ
b ∈ B, and M(
ˆ
b) < ∞. Show that (i) if λ = (λ
1
, . . . , λ
m
)

is a
ˆ
b then λ ≥ 0, and (1, −λ
1
, . . . , −λ
m
)

deﬁnes a non-vertical hyperplane
supporting A at (M(
ˆ
b),
ˆ
b), (ii) if (λ
0
, −λ
1
, . . . , −λ
m
)

deﬁnes a hyperplane supporting A at
(M(
ˆ
b),
ˆ
b) then λ
0
≥ 0, λ
i
≥ 0 for 1 ≤ i ≤ m; futhermore, if the hyperplane is non-vertical then
((λ
1

0
, . . . , (λ
m

0
))

is a supergradient of M at
ˆ
b.
We will prove only one part of the next crucial result. The reader who is familiar with the
Separation Theorem of convex sets should be able to construct a proof for the second part based
on Figure 5.4, or see the Appendix at the end of this Chapter.
Lemma 4: (Gale [1967]) M is stable at
ˆ
b iff M has a supergradient at
ˆ
b. Proof: (Sufﬁciency only)
Let λ be a supergradient at
ˆ
b, then
M(b) ≤ M(
ˆ
b) +λ

(b −
ˆ
b)
≤ M(
ˆ
b) +[λ[[b −
ˆ
b[ . ♦
The next two results give important alternative interpretations of supergradients.
Lemma 5: Suppose that ˆ x is optimal for (5.17). Then
ˆ
λ is a supergradient of M at
ˆ
b iff
ˆ
λ is an
optimal price vector, and then (ˆ x,
ˆ
λ) satisfy the optimality conditions.
Proof: By hypothesis, f(ˆ x) = M(
ˆ
b), ˆ x ∈ X, and f(ˆ x) ≤
ˆ
b. Let
ˆ
λ be a supergradient of M at
ˆ
b.
By Exercise 2, (M(
ˆ
b), f(ˆ x)) ∈ A and by Exercise 3,
ˆ
λ ≥ 0 and
M(
ˆ
b) −
ˆ
λ

ˆ
b ≥ M(
ˆ
b) −
ˆ
λ

f(ˆ x) ,
so that
ˆ
λ

(f(ˆ x) −
ˆ
b) ≥ 0. But then
ˆ
λ

(
ˆ
b − f(ˆ x)) = 0, giving (5.23). Next let x ∈ X. Then
(f
0
(x), f(x)) ∈ A, hence again by Exercise 3
M(
ˆ
b) −
ˆ
λ

ˆ
b ≥ f
0
(x) −
ˆ
λ

f(x) .
Since f
0
(ˆ x) = M(
ˆ
b), and
ˆ
λ

(f(ˆ x) −
ˆ
b) = 0, we can rewrite the inequality above as
1
From the Greek “hypo” meaning below or under. This neologism contrasts with the epigraph of a function which is
the set lying above the graph of the function.
62 CHAPTER 5. NONLINEAR PROGRAMMING
.
.
M(
ˆ
b)
M(b)
ˆ
b
A
b
No non-vertical hyperplane supporting A at (M(
ˆ
b),
ˆ
b)

0
, . . . , λ
m
)
M(
ˆ
b)
M(b)
ˆ π
A
b
ˆ
b
ˆ π is a non-vertical hyperplane supporting A at (M(
ˆ
b),
ˆ
b)
Figure 5.4: Hypograph and supporting hyperplane.
f
0
(ˆ x) +
ˆ
λ

(f(ˆ x) −
ˆ
b) ≥ f
0
(x) −
ˆ
λ

(f(x) −
ˆ
b) ,
so that (5.21) holds. It follows that (ˆ x,
ˆ
λ) satisfy the optimality conditions.
Conversely, suppose ˆ x ∈ X,
ˆ
λ ≥ 0 satisfy (5.21), (5.22), and (5.23). Let x ∈ Ω(b), i.e.,
x ∈ X, f(x) ≤ b. Then
ˆ
λ

(f(x) −b) ≤ 0 so that
f
0
(x) ≤ f
0
(x)
ˆ
λ

(f(x) −b)
= f
0
(x) −
ˆ
λ

(f(x) −
ˆ
b) +
ˆ
λ

(b −
ˆ
b)
≤ f
0
(ˆ x) −
ˆ
λ

(f(ˆ x) −
ˆ
b) +
ˆ
λ

(b −
ˆ
b) by (5.21)
= f
0
(ˆ x) +
ˆ
λ

(b −
ˆ
b) by (5.23)
= M(
ˆ
b) +
ˆ
λ

(b −
ˆ
b) .
Hence
M(b) = sup¦f
0
(x)[x ∈ Ω(b)¦ ≤ M(
ˆ
b) +
ˆ
λ

(b −
ˆ
b) ,
so that
ˆ
λ

is a supergradient of M at
ˆ
b. ♦
Lemma 6: Suppose that
ˆ
b ∈ B, and M(
ˆ
b) < ∞. Then
ˆ
λ is a supergradient of M at
ˆ
b iff
ˆ
λ is an
optimal solution of the dual (5.19) and m(
ˆ
λ) = M(
ˆ
b).
Proof: Let
ˆ
λ be a supergradient of M at
ˆ
b. Let x ∈ X. By Exercises 2 and 3
5.2. DUALITY THEORY 63
M(
ˆ
b) −
ˆ
λ

ˆ
b ≥ f
0
(x) −
ˆ
λ

f(x)
or
M(
ˆ
b) ≥ f
0
(x) −
ˆ
λ

(f(x) −
ˆ
b) ,
so that
M(
ˆ
b) ≥ sup¦f
0
(x) −
ˆ
λ

(f(x) −
ˆ
b)[x ∈ X¦ = m(
ˆ
λ) .
By weak duality (Lemma 2) it follows that M(
ˆ
b) = m(
ˆ
λ) and
ˆ
λ is optimal for (5.19).
Conversely suppose
ˆ
λ ≥ 0, and m(
ˆ
λ) = M(
ˆ
b). Then for any x ∈ X
M(
ˆ
b) ≥ f
0
(x) −
ˆ
λ

(f(x) −
ˆ
b) ,
and if moreover f(x) ≤ b, then
ˆ
λ

(f(x) −b) ≤ 0, so that
M(
ˆ
b) ≥ f
0
(x) −
ˆ
λ

(f(x) −
ˆ
b) +
ˆ
λ

(f(x) −b)
= f
0
(x) −
ˆ
λ

b +
ˆ
λ

ˆ
b for x ∈ Ω(b) .
Hence,
M(b) = sup¦f
0
(x)[x ∈ Ω(b)¦ ≤ M(
ˆ
b) +
ˆ
λ

(b −
ˆ
b) ,
so that
ˆ
We can now summarize our results as follows.
Theorem 2: (Duality) Suppose
ˆ
b ∈ B, M(
ˆ
b) < ∞, and M is stable at
ˆ
b. Then
(i) there exists an optimal solution
ˆ
λ for the dual, and m(
ˆ
λ) = M(
ˆ
b),
(ii)
ˆ
λ is optimal for the dual iff
ˆ
λ is a supergradient of M at
ˆ
b,
(iii) if
ˆ
λ is any optimal solution for the dual, then ˆ x is optimal for the primal iff (ˆ x,
ˆ
λ) satisfy the
optimality conditions of (5.21), (5.22), and (5.23).
Proof: (i) follows from Lemmas 4,6. (ii) is implied by Lemma 6. The “if” part of (iii) follows from
Theorem 1, whereas the “only if” part of (iii) follows from Lemma 5. ♦
Corollary 1: Under the hypothesis of Theorem 2, if
ˆ
λ is an optimal solution to the dual then
(∂M
+
/∂b
i
)(
ˆ
b) ≤
ˆ
λ
i
≤ (∂M

/∂b
i
)(
ˆ
b).
Exercise 4: Prove Corollary 1. (Hint: See Theorem 5 of 4.2.3.)
5.2.2 Interpretation and extensions.
It is easy to see using convexity properties that, if X = R
n
and f
i
, 0 ≤ i ≤ m, are differentiable,
then the optimality conditions (5.21), (5.22), and (5.23) are equivalent to the Kuhn-Tucker condition
(5.8). Thus the condition of stability of M at
ˆ
b plays a similar role to the constraint qualiﬁcation.
However, by Lemmas 4, 6 stability is equivalent to the existence of optimal dual variables, whereas
CQ is only a sufﬁcient condition. In other words if CQ holds at ˆ x then M is stable at
ˆ
b. In particular,
if X = R
n
and the f
i
are differentiable, the various conditions of Section 1.3 imply stability. Here
we give one sufﬁcient condition which implies stability for the general case.
Lemma 7: If
ˆ
b is in the interior of B, in particular if there exists x ∈ X such that f
i
(x) <
ˆ
b
i
for
1 ≤ i ≤ m, then M is stable at
ˆ
b.
64 CHAPTER 5. NONLINEAR PROGRAMMING
The proof rests on the Separation Theorem for convex sets, and only depends on the fact that M
is concave, M(
ˆ
b) < ∞ without loss of generality, and
ˆ
b is the interior of B. For details see the
Appendix.
Much of duality theory can be given an economic interpretation similar to that in Section 4.4.
Thus, we can think of x as the vector of n activity levels, f
0
(x) the corresponding revenue, X as
constraints due to physical or long-term limitations, b as the vector of current resource supplies,
and ﬁnally f(x) the amount of these resources used up at activity levels x. The various convexity
conditions are generalizations of the economic hypothesis of non-increasing returns-to-scale. The
primal problem (5.17) is the short-term decision problem faced by the ﬁrm. Next, if the current
resources can be bought or sold at prices
ˆ
λ = (λ
1
, . . . , λ
m
)

, the ﬁrm faces the decision problem
(5.18). If for a price system
ˆ
λ, an optimal solution of (5.17) also is an optimal solution for (5.18),
then we can interpret
ˆ
λ as a system of equilibrium prices just as in 4.2. Assuming the realistic
condition
ˆ
b ∈ B, M(
ˆ
b) < ∞ we can see from Theorem 2 and its Corollary 1 that there exists
an equilibrium price system iff (∂M
+
/∂b
i
)(
ˆ
b) < ∞, 1 ≤ i ≤ m; if we interpret (∂M
+
/∂b
i
)(
ˆ
b)
as the marginal revenue of the ith resource, we can say that equilibrium prices exist iff marginal
productivities of every (variable) resource is ﬁnite. These ideas are developed in (Gale [1967]).
.
M(b)
A
b
ˆ
b
M(
ˆ
b)
Figure 5.5: If M is not concave there may be no supporting hyperplane at (M(
ˆ
b),
ˆ
b).
Referring to Figure 5.3 or Figure 5.4, and comparing with Figure 5.5 it is evident that if M is not
concave or, equivalently, if its hypograph Ais not convex, there may be no hyperplane supporting A
at (M(
ˆ
b),
ˆ
b). This is the reason why duality theory requires the often restrictive convexity hypoth-
esis on X and f
i
. It is possible to obtain the duality theorem under conditions slightly weaker than
convexity but since these conditions are not easily veriﬁable we do not pursue this direction any fur-
ther (see Luenberger [1968]). A much more promising development has recently taken place. The
basic idea involved is to consider supporting A at (M(
ˆ
b),
ˆ
b) by (non-vertical) surfaces ˆ π more gen-
eral than hyperplanes; see Figure 5.6. Instead of (5.18) we would then have more general problem
of the form (5.26):
Maximize f
0
(x) −F(f(x) −
ˆ
b)
subject to x ∈ X ,
(5.26)
5.2. DUALITY THEORY 65
where F : R
m
→ R is chosen so that ˆ π (in Figure 5.6) is the graph of the function b → M(
ˆ
b) −
F(b −
ˆ
b). Usually F is chosen from a class of functions φ parameterized by µ = (µ
1
, . . . , µ
k
) ≥ 0.
Then for each ﬁxed µ ≥ 0 we have (5.27) instead of (5.26):
Maximize f
0
(x) −φ(µ; f(x) −
ˆ
b)
subject to x ∈ X .
(5.27)
.
M(b)
ˆ π
A
b
ˆ
b
M(
ˆ
b)
Figure 5.6: The surface ˆ π supports A at (M(
ˆ
b),
ˆ
b).
If we let
ψ(µ) =sup¦f
0
(x) −φ(µ; f(x) −
ˆ
b)[x ∈ X¦ .
then the dual problem is
Minimize ψ(µ)
subject to µ ≥ 0 ,
in analogy with (5.19).
The economic interpretation of (5.27) would be that if the prevailing (non-uniform) price system
is φ(µ; ) then the resources f(x) −
ˆ
b can be bought (or sold) for the amount φ(µ; f(x) −
ˆ
b). For
such an interpretation to make sense we should have φ(µ; b) ≥ 0 for b ≥ 0, and φ(µ; b) ≥ φ(µ;
˜
b)
whenever b ≥
˜
b. A relatively unnoticed, but quite interesting development along these lines is
presented in (Frank [1969]). Also see (Arrow and Hurwicz [1960]).
For non-economic applications, of course, no such limitation on φ is necessary. The following
references are pertinent: (Gould [1969]), (Greenberg and Pierskalla [1970]), (Banerjee [1971]). For
more details concerning the topics of 2.1 see (Geoffrion [1970a]) and for a mathematically more
elegant treatment see (Rockafellar [1970]).
5.2.3 Applications.
Decentralized resource allocation.
Parts (i) and (iii) of Theorem 2 make duality theory attractive for computation purposes. In particular
from Theorem 2 (iii), if we have an optimal dual solution
ˆ
λ then the optimal primal solutions are
those optimal solutions of (5.18) for λ =
ˆ
λ which also satisfy the feasibility condition (5.22) and
66 CHAPTER 5. NONLINEAR PROGRAMMING
the “complementary slackness” condition (5.23). This is useful because generally speaking (5.18)
is easier to solve than (5.17) since (5.18) has fewer constraints.
Consider a decision problem in a large system (e.g., a multi-divisional ﬁrm). The system is
made up of k sub-systems (divisions), and the decision variable of the ith sub-system is a vector
x
i
∈ R
n
i
, 1 ≤ i ≤ k. The sub-system has individual constraints of the form x
i
∈ X
i
where x
i
is
a convex set. Furthermore, the sub-systems share some resources in common and this limitation is
expressed as f
1
(x
1
) +. . . +f
k
(x
k
) ≤
ˆ
b where f
i
: R
n
i
→ R
m
are convex functions and
ˆ
b ∈ R
m
is the vector of available common resources. Suppose that the objective function of the large system
is additive, i.e. it is the form f
1
0
(x
1
) + . . . + f
k
0
(x
k
) where f
i
0
: R
n
i
→ R are concave functions.
Thus we have the decision problem (5.28):
Maximize
k
¸
i=1
f
i
0
(x
i
)
subject to x
i
∈ X
i
, 1 ≤ i ≤ k,
k
¸
i=1
f
i
(x
i
) ≤
ˆ
b .
(5.28)
For λ ∈ R
m
, λ ≥ 0, the problem corresponding to (5.19) is
Maximize f
i
0
(x
i
) −λ

f
i
(x
i
) −λ

(
k
¸
i=1
f
i
(x
i
) −
ˆ
b)
subject to x
i
∈ X
i
, 1 ≤ i ≤ k ,
which decomposes into k separate problems:
Maximize f
i
0
(x
i
) −λ

f
i
(x
i
)
subject to x
i
∈ X
i
, 1 ≤ i ≤ k .
(5.29)
If we let m
i
(λ) = sup¦f
i
0
(x
i
) −λ

f
i
(x
i
)[x
i
∈ X
i
¦, and m(λ) =
k
¸
i=1
m
i
(λ) +λ

ˆ
b, then the dual
problem is
Minimize m(λ) ,
subject to λ ≥ 0 .
(5.30)
Note that (5.29) may be much easier to solve than (5.28) because, ﬁrst of all, (5.29) involves fewer
constraints, but perhaps more importantly the decision problems in (5.29) are decentralized whereas
in (5.28) all the decision variables x
1
, . . . , x
k
are coupled together; in fact, if k is very large it may
be practically impossible to solve (5.28) whereas (5.29) may be trivial if the dimensions of x
i
are
small.
Assuming that (5.28) has an optimal solution and the stability condition is satisﬁed, we need to
ﬁnd an optimal dual solution so that we can use Theorem 2(iii). For simplicity suppose that the
f
i
0
, 1 ≤ i ≤ k, are strictly concave, and also suppose that (5.29) has an optimal solution for every
λ ≥ 0. Then by Exercise 8 of Section 1, for each λ ≥ 0 there is a unique optimal solution of (5.29),
say x
i
(λ). Consider the following algorithm.
5.2. DUALITY THEORY 67
Step 1. Select λ
0
≥ 0 arbitrary. Set p = 0, and go to Step 2.
Step 2. Solve (5.29) for λ = λ
p
and obtain the optimal solution x
p
= (x
1

p
), . . . , x
k

p
)).
Compute e
p
=
k
¸
i=1
f
i
(x
i

p
)) −
ˆ
b. If e
p
≥ 0, x
p
is feasible for (5.28) and can easily be seen to be
optimal.
Step 3. Set λ
p=1
according to
λ
p+1
i
=

λ
p
i
if e
p
i
≥ 0
λ
p
i
−d
p
e
p
i
if e
p
i
< 0
where d
p
> 0 is chosen a priori. Set p = p + 1 and return to Step 3.
It can be shown that if the step sizes d
p
are chosen properly, x
p
will converge to the optimum
solution of (5.28). For more detail see (Arrow and Hurwicz [1960]), and for other decentralization
schemes for solving (5.28) see (Geoffrion [1970b]).
Control of water quality in a stream.
The discussion in this section is mainly based on (Kendrick, et al., [1971]). For an informal discus-
sion of schemes of pollution control which derive their effectiveness from duality theory see (Solow
[1971]). See (Dorfman and Jacoby [1970].)
Figure 5.7 is a schematic diagram of a part of a stream into which n sources (industries and
municipalities) discharge polluting efﬂuents. The pollutants consist of various materials, but for
simplicity of exposition we assume that their impact on the quality of the stream is measured in
terms of a single quantity, namely the biochemical oxygen demand (BOD) which they place on the
dissolved oxygen (DO) in the stream. Since the DO in the stream is used to breakdown chemically
the pollutants into harmless substances, the quality of the stream improves with the amount of
DO and decreases with increasing BOD. It is a well-advertized fact that if the DO drops below a
certain concentration, then life in the stream is seriously threatened; indeed, the stream can “die.”
Therefore, it is important to treat the efﬂuents before they enter the stream in order to reduce the
BOD to concentration levels which can be safely absorbed by the DO in the stream. In this example
we are concerned with ﬁnding the optimal balance between costs of waste treatment and costs of
high BOD in the stream.
We ﬁrst derive the equations which govern the evolution in time of BOD and DO in the n areas
of the streams. The ﬂuctuations of BOD and DO will be cyclical with a period of 24 hours. Hence,
it is enough to study the problem over a 24-hour period. We divide this period into T intervals,
t = 1, . . . , T. During interval t and in area i let
z
i
(t) = concentration of BOD measured in mg/liter,
q
i
(t) = concentration of DO measured in mg/liter,
s
i
(t) = concentration of BOD of efﬂuent discharge in mg/liter, and
m
i
(t) = amount of efﬂuent discharge in liters.
The principle of conservation of mass gives us equations (5.31) and (5.32):
z
i
(t + 1) −z
i
(t) = −α
i
z
i
(t) +
ψ
i−1
z
i−1
(t)
v
i

ψ
i
z
i
(t)
v
i
+
s
i
(t)m
i
(t)
v
i
, (5.31)
q
i
(t + 1) −q
i
(t) = β
i
(q
s
i
−q
i
(t)) +
ψ
i−1
q
i−1
(t)
v
i

ψ
i
q
i
(t)
v
i

i
z
i
(t) −η
i
v
i
, t = 1, . . . , T and i = 1, . . . , N.
(5.32)
68 CHAPTER 5. NONLINEAR PROGRAMMING
. . . . . .
. . .
direction of ﬂow
0
z
0
q
0
1
z
1
q
1
i −1
z
i−1
q
i−1
i
z
i
q
i
i + 1
z
i+1
q
i+1
N
z
N
q
N
N + 1
given
(1 −π
1
)s
i
s
i
1 −π
i−1
s
i−a
s
i
(1 −π
i
)s
i
s
i+1
(1 −π
i+1
)s
i+1
s
N
(1 −π
N
)s
N
Figure 5.7: Schematic of stream with efﬂuent discharges.
Here, v
i
= volume of water in area i measured in liters, ψ
i
= volume of water which ﬂows from
area i to are i +1 in each period measured in liters. α
i
is the rate of decay of BOD per interval. This
decay occurs by combination of BOD and DO. β
i
is the rate of generation of DO. The increase in
DO is due to various natural oxygen-producing biochemical reactions in the stream and the increase
is proportional to (q
s
− q
i
) where q
s
is the saturation level of DO in the stream. Finally, η
i
is the
DO requirement in the bottom sludge. The v
i
, ψ
i
, α
i
, η
i
, q
s
are parameters of the stream and are
assumed known. They may vary with the time interval t. Also z
0
(t), q
0
(t) which are the concen-
trations immediately upstream from area 1 are assumed known. Finally, the initial concentrations
z
i
(1), q
i
(1), i = 1, . . . , N are assumed known.
Now suppose that the waste treatment facility in area i removes in interval t a fraction π
i
(t) of
the concentration s
i
(t) of BOD. Then (5.31) is replaced by
z
i
(t + 1) −z
i
(t) = −α
i
z
i
(t) +
ψ
i
z
i−1
v
i

ψ
i
z
i
(t)
v
i
+
(1−π
i
(t))s
i
(t)m
i
(t)
v
i
. (5.33)
We now turn to the costs associated with waste treatment and pollution. The cost of waste treat-
ment can be readily identiﬁed. In period t the ith facility treats m
i
(t) liters of efﬂuent with a BOD
concentration s
i
(t) mg/liter of which the facility removes a fraction π
i
(t). Hence, the cost in period
t will be f
i

i
(t), s
i
(t), m
i
(t)) where the function must be monotonically increasing in all of its
arguments. We further assume that f is convex.
The costs associated with increased amounts of BOD and reduced amounts of DO are much
more difﬁcult to quantify since the stream is used by many institutions for a variety of purposes
(e.g., agricultural, industrial, municipal, recreational), and the disutility caused by a decrease in
the water quality varies with the user. Therefore, instead of attempting to quantify these costs let
us suppose that some minimum water quality standards are set. Let q be the minimum acceptable
DO concentration and let ¯ z be the maximum permissible BOD concentration. Then we face the
5.2. DUALITY THEORY 69
following NP:
Maximize −
N
¸
i=1
T
¸
t=1
f
i

i
(t), s
i
(t), m
i
(t))
subject to (5.32), (5.33), and
−q
i
(t) ≤ −q , i = 1, . . . , N; t = 1, . . . , T,
z
i
(t) ≤ ¯ z , i = 1, . . . , N; t = 1, . . . , T,
0 ≤ π
i
(t) ≤ 1 , i = 1, . . . , N; t = 1, . . . , T.
(5.34)
Suppose that all the treatment facilities are in the control of a single public agency. Then assuming
that the agency is required to maintain the standards (q, ¯ z) and it does this at a minimum cost it will
solve the NP (5.34) and arrive at an optimal solution. Let the minimum cost be m(q, ¯ z). But if
there is no such centralized agency, then the individual polluters may not (and usually do not) have
any incentive to cooperate among themselves to achieve these standards. Furthermore, it does not
make sense to enforce legally a minimum standard q
i
(t) ≥ q, z
i
(t) ≤ ¯ z on every polluter since the
pollution levels in the ith area depend upon the pollution levels on all the other areas lying upstream.
On the other hand, it may be economically and politically acceptable to tax individual polluters in
proportion to the amount of pollutants discharged by the individual. The question we now pose
is whether there exist tax rates such that if each individual polluter minimizes its own total cost
(i.e., cost of waste treatment + tax on remaining pollutants), then the resulting water quality will be
acceptable and, furthermore, the resulting amount of waste treatment is carried out at the minimum
expenditure of resources (i.e., will be an optimal solution of (5.34)).
It should be clear from the duality theory that the answer is in the afﬁrmative. To see this let
w
i
(t) = (z
i
(t), −q
i
(t))

, let w(t) = (w
1
(t), . . . , w
N
(t)), and let w = (w(1), . . . , w(t)). Then we
can solve (5.32) and (5.33) for w and obtain
w = b +Ar , (5.35)
where the matrix A and the vector b depend upon the known parameters and initial conditions, and
r is the NT-dimensional vector with components (1 − π
i
(t))s
i
(t)m
i
(t). Note that the coefﬁcients
of the matrix must be non-negative because an increase in any component of r cannot decrease the
BOD levels and cannot increase the DO levels. Using (5.35) we can rewrite (5.34) as follows:
Maximize −
¸
i
¸
t
f
i

i
(t), s
i
(t), m
i
(t))
subject to b +Ar ≤ ¯ w ,
0 ≤ π
i
(t) ≤ 1 , i = 1, . . . , N; t = 1, . . . , T,
(5.36)
where the 2NT-dimensional vector ¯ w has its components equal to −q or ¯ z in the obvious manner.
By the duality theorem there exists a 2NT-dimensional vector λ

≥ 0, and an optimal solution
π

i
(t), i = 1, . . . , N, t = 1, . . . , T, of the problem:
Maximize −
¸
i
¸
t
f
i

i
(t), s
i
(t), m
i
(t)) −λ

(b +Ar −w)
subject to 0 ≤ π
i
(t) ≤ 1, i = 1, . . . , N; t = 1, . . . , T ,
(5.37)
such that ¦π

i
(t)¦ is also an optimal solution of (5.36) and, furthermore, the optimal values of (5.36)
and (5.37) are equal. If we let p

= A

λ

≥ 0, and we write the components of p

as p

i
(t) to match
70 CHAPTER 5. NONLINEAR PROGRAMMING
with the components (1−π
i
(t))s
i
(t)m
i
(t) of r we can see that (5.37) is equivalent to the set of NT
problems:
Maximize −f
i

i
(t), s
i
(t), m
i
(t)) −p

i
(t)(1 −π
i
(t))s
i
(t)m
i
(t)
0 ≤ π
i
(t) ≤ 1 ,
i = 1, . . . , N; t = 1, . . . , T .
(5.38)
Thus, p

i
(t) is optimum tax per mg of BOD in area i during period t.
Before we leave this example let us note that the optimum dual variable or shadow price λ

plays an important role in a larger framework. We noted earlier that the quality standard (q, ¯ z)
was somewhat arbitrary. Now suppose it is proposed to change the standard in the ith area during
period t to q +∆q
i
(t) and ¯ z +∆z
i
(t). If the corresponding components of λ

are λ
q∗
i
(t) and λ
z∗
i
(t),
then the change in the minimum cost necessary to achieve the new standard will be approximately
λ
q∗
i
(t)∆q
i
(t) + λ
z∗
i
(t)∆z
i
(t). This estimate can now serve as a basis in making a beneﬁts/cost
analysis of the proposed new standard.
An important special case of NP is the quadratic programming (QP) problem:
Maximize c

x −
1
2
x

Px
subject to Ax ≤ b, x ≥ 0 ,
(5.39)
where x ∈ R
n
is the decision variable , c ∈ R
n
, b ∈ R
m
are ﬁxed, A is a ﬁxed m n matrix and
P = P

is a ﬁxed positive semi-deﬁnite matrix.
Theorem 1: A vector x

∈ R
n
is optimal for (5.39) iff there exist λ

∈ R
m
, µ

∈ R
n
, such that
Ax

≤ b, x

≥ 0
c −Px

= A

λ

−µ

, λ

≥ 0, µ

≥ 0 ,

)

(Ax

−b) = 0 , (µ

)

x

= 0 .
(5.40)
Proof: By Lemma 3 of 1.3, CQ is satisﬁed, hence the necessity of these conditions follows from
Theorem 2 of 1.2. On the other hand, since P is positive semi-deﬁnite it follows from Exercise 6
of Section 1.2 that f
0
: x → c

x −1/2 x

Px is a concave function, so that the sufﬁciency of these
conditions follows from Theorem 4 of 1.2. ♦
From (5.40) we can see that x

is optimal for (5.39) iff there is a solution (x

, y

, λ

, µ

) to
(5.41), (5.42), and (5.43):
Ax +I
m
Y = b
−Px −A

λ +I
n
µ = −c ,
(5.41)
x ≥ 0 y ≥ 0, λ ≥ 0, µ ≥ 0 , (5.42)
µ

x = 0 , λ

y = 0 . (5.43)
Suppose we try to solve (5.41) and (5.42) by Phase I of the Simplex algorithm (see 4.3.2). Then we
must apply Phase II to the LP:
Maximize −
m
¸
i=1
z
i

n
¸
j=1
ξ
j
5.4. COMPUTATIONAL METHOD 71
subject to
Ax +I
m
y +z = b
−Px −A

λ +I
n
µ +ξ = −c
x ≥ 0, y ≥ 0, λ ≥ 0, µ ≥ 0, z ≥ 0, ξ ≥ 0,
(5.44)
starting with a basic feasible solution z = b, ξ = −c. (We have assumed, without loss of generality,
that b ≥ 0 and −c ≥ 0.) If (5.41) and (5.42) have a solution then the maximum value in (5.44) is 0.
We have the following result.
Lemma 1: If (5.41), (5.42), and (5.43) have a solution, then there is an optimal basic feasible solution
of (5.44) which is also a solution f (5.41), (5.42), and (5.43).
Proof: Let ˆ x, ˆ y,
ˆ
λ, ˆ µ be a solution of (5.41), (5.42), and (5.43). Then ˆ x, ˆ y,
ˆ
λ, ˆ µ, ˆ z = 0,
ˆ
ξ = 0 is
an optimal solution of (5.44). Furthermore, from (5.42) and (5.43) we see that at most (n + m)
components of (ˆ x, ˆ y,
ˆ
λ, ˆ µ) are non-zero. But then a repetition of the proof of Lemma 1 of 4.3.1 will
also prove this lemma. ♦
This lemma suggests that we can apply the Simplex algorithm of 4.3.2 to solve (5.44), starting
with the basic feasible solution z = b, ξ = −c, in order to obtain a solution of (5.41), (5.42), and
(5.43). However, Step 2 of the Simplex algorithm must be modiﬁed as follows to satisfy (5.43):
If a variable x
j
is currently in the basis, do not consider µ
j
as a candidate for entry into the basis;
if a variable y
i
is currently in the basis, do not consider λ
i
as a candidate for entry into the basis. If
it not possible to remove the z
i
and ξ
j
from the basis, stop.
The above algorithm is due to Wolfe [1959]. The behavior of the algorithm is summarized below.
Theorem 2: Suppose P is positive deﬁnite. The algorithm will stop in a ﬁnite number of steps at an
optimal basic feasible solution (ˆ x, ˆ y,
ˆ
λ, ˆ µ, ˆ z,
ˆ
ξ) of (5.44). If ˆ z = 0 and
ˆ
ξ = 0 then (ˆ x, ˆ y,
ˆ
λ, ˆ µ) solve
(5.41), (5.42), and (5.43) and ˆ x is an optimal solution of (5.39). If ˆ z = 0 or
ˆ
ξ = 0, then there is no
solution to (5.41), (5.42), (5.43), and there is no feasible solution of (5.39).
For a proof of this result as well as for a generalization of the algorithm which permits positive
semi-deﬁnite P see (Cannon, Cullum, and Polak [1970], p. 159 ff).
5.4 Computational Method
Maximize f
0
(x)
subject to f
i
(x) ≤ 0, i = 1, . . . , m ,
(5.45)
where x ∈ R
n
, f
i
: R
n
→ R, 0 ≤ i ≤ m, are differentiable. Let Ω ⊂ R
n
denote the set of
feasible solutions. For ˆ x ∈ Ω deﬁne the function ψ(ˆ x) : R
n
→ R by
ψ(ˆ x)(h) = max¦−f
0x
(ˆ x)h, f
1
(ˆ x) +f
1x
(ˆ x)h, . . . , f
m
(ˆ x) +f
mx
(ˆ x)h¦.
Consider the problem:
Minimize ψ(ˆ x)(h)
subject to −ψ(ˆ x)(h) −f
0x
(ˆ x)h ≤ 0 ,
−ψ(ˆ x)(h) +f
i
(ˆ x)f
ix
h ≤ 0 ,
1 ≤ i ≤ m , −1 ≤ h
j
≤ 1 , 1 ≤ j ≤ n .
(5.46)
72 CHAPTER 5. NONLINEAR PROGRAMMING
.
f
0
(x) = F
0
(x

) > f
0
(x
k
)
f
0
(x) = f
0
(x
k
)
f
2
= 0
f
1
= 0

f
3
= 0
f
2
(x
k
)
x
k
f
3
(x
k
)
f
1
(x
k
)
f
0
(x
k
)
h(x
k
)
Figure 5.8: h(x
k
) is a feasible direction.
Call h(ˆ x) an optimum solution of (5.46) and let h
0
(ˆ x) = ψ(ˆ x)(h(ˆ x)) be the minimum value at-
tained. (Note that by Exercise 1 of 4.5.1 (5.46) can be solved as an LP.)
The following algorithm is due to Topkis and Veinott [1967].
Step 1. Find x
0
∈ Ω, set k = 0, and go to Step 2.
Step 2. Solve (5.46) for ˆ x = x
k
and obtain h
0
(x
k
), h(x
k
). If h
0
(x
k
) = 0, stop, otherwise go to Step
3.
Step 3. Compute an optimum solution µ(x
k
) to the one-dimensional problem,
Maximize f
0
(x
k
+µh(x
k
)) ,
subject to (x
k
+µh(x
k
)) ∈ Ω, µ ≥ 0 ,
and go to Step 4.
Step 4. Set x
k+1
= x
k
+µ(x
k
)h(x
k
), set k = k + 1 and return to Step 2.
The performance of the algorithm is summarized below.
Theorem 1: Suppose that the set
Ω(x
0
) = ¦x[x ∈ Ω, f
0
(x) ≥ f
0
(x
0

is compact, and has a non-empty interior, which is dense in Ω(x
0
). Let x

be any limit point of
the sequence x
0
, x
1
, . . . , x
k
, . . . , generated by the algorithm. Then the Kuhn-Tucker conditions are
satisﬁed at x

.
For a proof of this result and for more efﬁcient algorithms the reader is referred to (Polak [1971]).
Remark: If h
0
(x
k
) < 0 in Step 2, then the direction h(x
k
) satisﬁes f
0x
(x
k
)h(x
k
) > 0, and f
i
(x
k
)+
f
ix
(x
K
)h(x
k
) < 0, 1 ≤ i ≤ m. For this reason h(x
k
) is called a (desirable) feasible direction.
(See Figure 5.8.)
5.5. APPENDIX 73
5.5 Appendix
The proofs of Lemmas 4,7 of Section 2 are based on the following extremely important theorem
(see Rockafeller [1970]).
Separation theorem for convex sets. Let F, Gbe convex subsets of R
n
such that the relative interiors
of F, G are disjoint. Then there exists λ ∈ R
n
, λ = 0, and θ ∈ R such that
λ

g ≤ θ for all g ∈ G
λ

f ≥ θ for all f ∈ F .
Proof of Lemma 4: Since M is stable at
ˆ
b there exists K such that
M(b) −M(
ˆ
b) ≤ K[b −
ˆ
b[ for all b ∈ B . (5.47)
In R
1+m
consider the sets
F = ¦(r, b)[b ∈ R
m
, r > K[b −
ˆ
b[¦ ,
G = ¦(r, b)[b ∈ B, r ≤ M(b) −M(
ˆ
b)¦ .
It is easy to check that F, G are convex, and (5.47) implies that F ∩ G = φ. Hence, there exist

0
, . . . , λ
m
) = 0, and θ such that
λ
0
r +
m
¸
i=1
λ
i
b
i
≤ θ for (r, b) ∈ G ,
λ
0
r +
m
¸
i=1
λ
i
b
i
≥ θ for (r, b) ∈ F .
(5.48)
From the deﬁnition of F, and the fact that (λ
0
, . . . , λ
m
) = 0, it can be veriﬁed that (5.49) can hold
only if λ
0
> 0. Also from (5.49) we can see that
m
¸
i=1
λ
i
ˆ
b
i
≥ θ, whereas from (5.48)
m
¸
i=1
λ
i
ˆ
b
i
≤ θ,
so that
m
¸
i=1
λ
i
ˆ
b
i
= θ. But then from (5.48) we get
M(b) −M(
ˆ
b) ≤
1
λ
0
[θ −
m
¸
i=1
λ
i
b
i
] =
m
¸
i=1
(−
λ
i
λ
0
)(b
i

ˆ
b). ♦
Proof of Lemma 7: Since
ˆ
b is in the interior of B, there exists ε > 0 such that
b ∈ B whenever [b −
ˆ
b[ < ε . (5.49)
In R
1+m
consider the sets
F = ¦(r,
ˆ
b)[r > M(
ˆ

G = ¦(r, b)[b ∈ B, r ≤ M(b)¦ .
Evidently, F, G are convex and F ∩ G = φ, so that there exist (λ
0
, . . . , λ
m
) = 0, and θ such that
λ
0
r +
m
¸
i=1
λ
i
ˆ
b
i
≥ θ , for r > M(
ˆ
b) , (5.50)
74 CHAPTER 5. NONLINEAR PROGRAMMING
λ
0
r +
m
¸
i=1
λ
i
ˆ
b
i
≤ θ , for (r, b) ∈ G . (5.51)
From (5.49), and the fact that (λ
0
, . . . , λ
m
) = 0 we can see that (5.50) and (5.51) imply λ
0
> 0.
From (5.50),(5.51) we get
λ
0
M(
ˆ
b) +
m
¸
i=1
λ
i
ˆ
b
i
= θ ,
so that (5.52) implies
M(b) ≤ (
ˆ
b) +
m
¸
i=1
(−
λ
i
λ
0
)(b
i

ˆ
b
i
) . ♦
Chapter 6
SEQUENTIAL DECISION PROBLEMS:
DISCRETE-TIME OPTIMAL
CONTROL
In this chapter we apply the results of the last two chapters to situations where decisions have to be
made sequentially over time. A very important class of problems where such situations arise is in
the control of dynamical systems. In the ﬁrst section we give two examples, and in Section 2 we
derive the main result.
6.1 Examples
The trajectory of a vertical sounding rocket is controlled by adjusting the rate of fuel ejection which
generates the thrust force. Speciﬁcally suppose that the equations of motion are given by (6.1).
˙ x
1
(t) = x
2
(t)
˙ x
2
(t) = −
C
D
x
3
(t)
ρ(x
1
(t))x
2
2
(t) −g +
C
T
x
3
(t)
u(t)
˙ x
3
(t) = −u(t) ,
(6.1)
where x
1
(t) is the height of the rocket from the ground at time t, x
2
(t) is the (vertical) speed at
time t, x
3
(t) is the weight of the rocket (= weight of remaining fuel) at time t. The “dot” denotes
differentiation with respect to t. These equations can be derived from the force equations under the
assumption that there are four forces acting on the rocket, namely: inertia = x
3
¨ x
1
= x
3
˙ x
2
; drag
force = C
D
ρ(x
1
)x
2
2
where C
D
is constant, ρ(x
1
) is a friction coefﬁcient depending on atmospheric
density which is a function of x
1
; gravitational force = gx
3
with g assumed constant; and thrust
force C
T
˙ x
3
, assumed proportional to rate of fuel ejection. See Figure 6.1. The decision variable at
time t is u(t), the rate of fuel ejection. At time 0 we assume that (x
1
(0), x
2
(0), x
3
(0)) = (0, 0, M);
that is, the rocket is on the ground, at rest, with initial fuel of weight M. At a prescribed ﬁnal time
t
f
, it is desired that the rocket be at a position as high above the ground as possible. Thus, the
75
76 CHAPTER 6. DISCRETE-TIME OPTIMAL CONTROL
decision problem can be formalized as (6.2).
Maximize x
1
(t
f
)
subject to ˙ x(t) = f(x(t), u(t)), 0 ≤ t ≤ t
f
x(0) = (0, 0, M)
u(t) ≥ 0, x
3
(t) ≥ 0, 0 ≤ t ≤ t
f
,
(6.2)
where x = (x
1
, x
2
, x
3
)

, f : R
3+1
→ R
3
is the right-hand side of (6.1). The constraint inequalities
u(t) ≥ 0 and x
3
(t) ≥ 0 are obvious physical constraints.
x
3
¨ x
1
= inertia
C
D
ϕ(x
1
)x
2
2
= drag
gx
3
= gravitational force
C
R
˙ x
3
= thrust
Figure 6.1: Forces acting on the rocket.
The decision problem (6.2) differs from those considered so far in that the decision variables,
which are functions u : [0, t
f
] → R, cannot be represented as vectors in a ﬁnite-dimensional
space. We shall treat such problems in great generality in the succeeding chapters. For the moment
we assume that for computational or practical reasons it is necessary to approximate or restrict
the permissible function u() to be constant over the intervals [0, t
1
), [t
1
, t
2
), . . . , [t
N−1
, t
f
), where
t
1
, t
2
, . . . , t
N−1
are ﬁxed a priori. But then if we let u(i) be the constant value of u() over [t
i
, t
i+1
),
we can reformulate (6.2) as (6.3):
Maximize x
1
(t
N
)(t
N
= t
f
)
subject to x(t
i+1
) = g(i, x(t
i
), u(i)), i = 0, 1, . . . , N −1
x(t
0
) = x(0) = (0, 0, M)
u(i) ≥ 0, x
3
(t
i
) ≥ 0, i = 0, 1, . . . , N .
(6.3)
In (6.3) g(i, x(t
1
), u(i)) is the state of the rocket at time t
i+1
when it is in state x(t
i
) at time t
i
and
u(t) ≡ u(i) for t
i
≤ t < t
i+1
.
As another example consider a simple inventory problem where time enters discretely in a natural
fashion. The Squeezme Toothpaste Company wants to plan its production and inventory schedule
for the coming month. It is assumed that the demand on the ith day, 0 ≤ i ≤ 30, is d
1
(i) for
6.2. MAIN RESULT 77
their orange brand and d
2
(i) for their green brand. To meet unexpected demand it is necessary that
the inventory stock of either brand should not fall below s > 0. If we let s(i) = (s
1
(i), s
2
(i))

denote the stock at the beginning of the ith day, and m(i) = (m
1
(i), m
2
(i))

denote the amounts
manufactured on the ith day, then clearly
s(i + 1) +s(i) +m(i) −d(i) ,
where d(i) = (d
1
(i), d
2
(i))

. Suppose that the initial stock is ˆ s, and the cost of storing inventory s
for one day is c(s) whereas the cost of manufacturing amount mis b(m). The the cost-minimization
decision problem can be formalized as (6.4):
Maximize
30
¸
i=0
(c(s(i)) +b(m(i)))
subject to s(i + 1) = s(i) +m(i) −d(i), 0 ≤ i ≤ 29
s(0) = ˆ s
s(i) ≥ (s, s)

, m(i) ≥ 0, 0 ≤ i ≤ 30 .
(6.4)
Before we formulate the general problem let us note that (6.3) and (6.4) are in the form of non-
linear programming problems. The reason for treating these problems separately is because of their
practical importance, and because the conditions of optimality take on a special form.
6.2 Main Result
The general problem we consider is of the form (6.5).
Maximize
N−1
¸
i=0
f
0
(i, x(i), u(i))
subject to
dynamics : x(i + 1) −x(i) = f(i, x(i), u(i)), i = 0, . . . , N −1 ,
initial condition: q
0
(x(0) ≤ 0, g
0
(x(0)) = 0 ,
ﬁnal condition: q
N
(x(N)) ≤ 0, g
N
(x(N)) = 0 ,
state-space constraint: q
i
(x(i)) ≤ 0, i = 1, . . . , N −1 ,
control constraint: h
i
(u(i)) ≤ 0, i = 0, . . . , N −1 .
(6.5)
Here x(i) ∈ R
n
, u(i) ∈ R
p
, f
0
(i, , ) : R
n+p
→ R, f(i, , ) : R
n+p
→ R
n
, q
i
: R
n

R
m
i
, g
i
: R
n
→ R

i
, h
i
: R
p
→ R
s
i
are given differentiable functions. We follow the control
theory terminology, and refer to x(i) as the state of the system at time i, and u(i) as the control or
input at time i.
We use the formulation mentioned in the Remark following Theorem 3 of V.1.2, and construct
the Lagrangian function L by
L(x(0), . . . , x(N); u(0), . . . , u(N −1); p(1), . . . , p(N);
λ
0
, . . . , λ
N
; α
0
, α
N
; γ
0
, . . . , γ
N−1
)
78 CHAPTER 6. DISCRETE-TIME OPTIMAL CONTROL
=
N−1
¸
i=0
f
0
(i, x(i), u(i)) −

N−1
¸
i=0
(p(i + 1))

(x(i + 1) −x(i) −f(i, x(i), u(i)))+
N
¸
i=0

i
)

q
i
(x(i)) + (α
0
)

g
0
(x(0)) + (α
N
)

g
N
(x(N)) +
N−1
¸
i=0

i
)

h
i
(u(i))
¸
.
Suppose that CQ is satisﬁed for (6.5), and x

(0), . . . , x

(N); u

(0), . . . , u

(N −1), is an optimal
solution. Then by Theorem 2 of 5.1.2, there exist p

(i) in R
n
for 1 ≤ i ≤ N, λ
i∗
≥ 0 in R
m
i
for
0 ≤ i ≤ N, α
i∗
in R

i
for i = 0, N, and γ
i∗
≥ 0 in R
s
i
for 0 ≤ i ≤ N −1, such that
(A) the derivative of L evaluated at these points vanishes,
and
(B) λ
i∗
q
i
(x

(i)) = 0 for 0 ≤ i ≤ N, γ
i∗
h
i
(u

(i)) = 0 for 0 ≤ i ≤ N −1 .
We explore condition (A) by taking various partial derivatives.
Differentiating L with respect to x(0) gives
f
0x
(0, x

(0), u

(0)) −¦−(p

(1))

−(p

(1))

[f
x
(0, x

(0), u

(0))]
+(λ
0∗
)

[q
0x
(x

(0))] + (α
0∗
)

[g
0x
(x

(0))]¦ = 0 ,
or
p

(0) −p

(1) = [f
x
(0, x

(0), u

(x))]

p

(1)
+[f
0x
(0, x

(0), u

(0))]

−[q
0x
(x

(0))]

λ
0∗
,
(6.6)
where we have deﬁned
p

(0) = [g
0x
(x

(x))]

α
0∗
. (6.7)
Differentiating L with respect to x(i), 1 ≤ i ≤ N −1, and re-arranging terms gives
p

(i) −p

(i + 1) = [f
x
(i, x

(i), u

(i))]

p

(i + 1)
+[f
0x
(i, x

(i), u

(i))]

−[q
ix
(x

(i))]

λ
i∗
.
(6.8)
Differentiating L with respect to x(N) gives,
p

(N) = −[g
Nx
(x

(N))]

α
N∗
−[q
Nx
(x

(N))]

λ
N∗
.
It is convenient to replace α
N∗
by −α
N∗
so that the equation above becomes (6.9)
p

(N) = [g
Nx
(x

(N))]

α
N∗
−[q
Nx
(x

(N))]

λ
N∗
. (6.9)
Differentiating L with respect to u(i), 0 ≤ i ≤ N −1 gives
[f
0u
(i, x

(i), u

(i))]

+ [f
u
(i, x

(i), u

(i))]

p

(i +l) −[h
iu
(u

(i))]

γ
i∗
= 0 . (6.10)
We summarize our results in a convenient form in
Table 6.1
Remark 1: Considerable elegance and mnemonic simpliﬁcation is achieved if we deﬁne the
Hamiltonian function H by
6
.
2
.
M
A
I
N
R
E
S
U
L
T
7
9
Suppose x

(0), . . . , x

(N);
u

(0), . . . , u

(N −1) maximizes
N−1
¸
i=0
f
0
(i, x(i), u(i)) subject
to the constraints below
then there exist p

(N); λ
0∗
, . . . , λN

; α
0∗
, α
N∗;
γ
0∗
, . . . , γ
N−1

, such that
dynamics: i = 0, . . . , N −1
x(i + 1) −x(i) = f(i, x(i), u(i))
initial condition:
q
0
(x

(0)) ≤ 0, g
0
(x

(0)) = 0
ﬁnal conditions:
q
N
(x

(N)) ≤ 0, g
N
(x

(N)) = 0
state space constraint:
i = 1, . . . , N −1
q
i
(x

(i)) ≤ 0
control constraint:
i = 0, . . . , N −1
h
i
(u

(i)) ≤ 0
adjoint equations: i = 0, . . . , N −1
p

(i) −p

(i + 1) = [f
x
(i, x

(i), u

(i)]

p

(i + 1)
+[f
0x
(i, x

(i), u

(i)]

−[q
ix
(x

(i)]

γ
i∗
transversality conditions:
p

(0) = [g
0x
(x

(0))]

α
0∗
p

(N) = [g
Nx
(x

(N))]

α
N∗
−[q
Nx
(x

(N))]

λ
N∗
[f
0u
(i, x

(i), u

(i))]

+ [f
u
(i, x

(i)u

(i))]

.
p

(i
1
) = [h
iu
(u

(i))]

γ
i∗
λ
0∗
≥ 0,

0∗
)

q
0
(x

(0)) = 0
λ
N∗
≥ 0,

N∗)

q
N
(x

(N)) = 0
λ
i∗
≥ 0,

i∗
)

q
i
(x

(i)) = 0
γ
i∗
≥ 0

i∗)

h
i
(u

(i) = 0
T
a
b
l
e
6
.
1
:
80 CHAPTER 6. DISCRETE-TIME OPTIMAL CONTROL
H(i, x, u, p) = f
0
(i, x, u) +p

f(i, x, u) .
The dynamic equations then become
x

(i + 1) −x

(i) = [H
p
(i, x

(i), u

(i), p

(i + 1))]

,
0 ≤ i ≤ N −1 .
(6.11)
and the adjoint equations (6.6) and (6.8) become
p

(i) −p

(i + 1) = [H
x
(i, x

(i), u

(i), u

(i), p

(i + 1))]

−[q
ix
(x

(i))]

λ
i∗
,
0 ≤ i ≤ N −1 ,
whereas (6.10) becomes
[h
iu
(u

(i))]

γ
i∗
= [H
u
(i, x

(i), u

(i), p

(i + 1))]

, 0 ≤ i ≤ N −1 . (6.12)
Remark 2: If we linearize the dynamic equations about the optimal solution we obtain
δx(i + 1) −δx(i) = [f
x
(i, x

(i), u

(i))]δx(i) + [f
u
(i, x

(i, x

, (i), u

(i))]δu(i) ,
whose homogeneous part is
z(i + 1) −z(i) = [f
x
(i, x

(i), u

(i))]z(i) ,
which has for it adjoint the system
r(i) −r(i + 1) = [f
x
(i, x

(i), u

(i))]

r(i + 1) . (6.13)
Since the homogeneous part of the linear difference equations (6.6), (6.8) is (6.13), we call (6.6),
(6.8) the adjoint equations, and the p

Remark 3: If the f
0
(i, , ) are concave and the remaining function in (6.5) are linear, then CQ is
satisﬁed, and the necessary conditions of Table 6.1 are also sufﬁcient. Furthermore, in this case we
see from (6.13) that u

(i) is an optimal solution of
Maximize H(i, x

(i), u, p

(i + 1)),
subject to h
i
(u) ≤ 0 .
For this reason the result is sometimes called the maximum principle.
Remark 4: The conditions (6.7), (6.9) are called transversality conditions for the following reason.
Suppose q
0
≡ 0, q
N
≡ 0, so that the initial and ﬁnal conditions read g
0
(x(0)) = 0, g
N
(x(N)) = 0,
which describe surfaces in R
n
. Conditions (6.7), (6.9) become respectively p

(0) = [g
0x
(x

(0))]

α
0∗
, p

(N) =
[g
Nx
(x(N))]

α
N∗
which means that p

(0) and p

(N) are respectively orthogonal or transversal to
the initial and ﬁnal surfaces. Furthermore, we note that in this case the initial and ﬁnal conditions
specify (
0
+
n
) conditions whereas the transversality conditions specify (n−
0
)+(n−
n
) condi-
tions. Thus, we have a total of 2n boundary conditions for the 2n-dimensional system of difference
equations (6.5), (6.12); but note that these 2n boundary conditions are mixed, i.e., some of them
refer to the initial time 0 and the rest refer to the ﬁnal time.
6.2. MAIN RESULT 81
Exercise 1: For the regulator problem,
Maximize
1
2
N−1
¸
i=0
x(i)

Qx(i) +
1
2
N−1
¸
i=0
u(i)

Pu(i)
subject to x(i + 1) −x(i) = Ax(i) +Bu(i), 0 ≤ i ≤ N −1
x(0) = ˆ x(0),
u(i) ∈ R
p
, 0 ≤ i ≤ N −1 ,
where x(i) ∈ R
n
, A and B are constant matrices, ˆ x(0) is ﬁxed, Q = Q

is positive semi-deﬁnite,
and P = P

is positive deﬁnite, show that the optimal solution is unique and can be obtained by
solving a 2n-dimensional linear difference equation with mixed boundary conditions.
Exercise 2: Show that the minimal fuel problem,
Minimize
N−1
¸
i=0

¸
P
¸
j=1
[(u(i))
j
[
¸

,
subject to x(i + 1) −x(i) = Ax(i) +Bu(i), 0 ≤ i ≤ N −1
x(0) = ˆ x(0), x(N) = ˆ x(N) ,
u(i) ∈ R
p
, [u(i))
j
[ ≤ 1, 1 ≤ j ≤ p, 0 ≤ i ≤ N −1
can be transformed into a linear programming problem. Here ˆ x(0), ˆ x(N) are ﬁxed, A and B are as
in Exercise 1.
82 CHAPTER 6. DISCRETE-TIME OPTIMAL CONTROL
Chapter 7
SEQUENTIAL DECISION PROBLEMS:
CONTINUOUS-TIME OPTIMAL
CONTROL OF LINEAR SYSTEMS
We will investigate decision problems similar to those studied in the last chapter with one (math-
ematically) crucial difference. A choice of control has to be made at each instant of time t where
t varies continuously over a ﬁnite interval. The evolution in time of the state of the systems to be
controlled is governed by a differential equation of the form:
˙ x(t) = f(t, x(t), u(t)) ,
where x(t) ∈ R
n
and u(t) ∈ R
p
are respectively the state and control of the system at time t.
To understand the main ideas and techniques of analysis it will prove proﬁtable to study the linear
case ﬁrst. The general nonlinear case is deferred to the next chapter. In Section 1 we present the
general linear problem and study the case where the initial and ﬁnal conditions are particularly
simple. In Section 2 we study more general boundary conditions.
7.1 The Linear Optimal Control Problem
We consider a dynamical system governed by the linear differential equation (7.1):
˙ x(t) = A(t)x(t) +B(t)u(t), t ≥ t
0
. (7.1)
Here A() and B() are n n- and n p-matrix valued functions of time; we assume that they are
piecewise continuous functions. The control u() is constrained to take values in a ﬁxed set Ω ⊂ R
p
,
and to be piecewise continuous.
Deﬁnition: A piecewise continuous function u : [t
0
, ∞) → Ω will be called an admissible control.
| denotes the set of all admissible controls.
Let c ∈ R
n
, x
0
∈ R
n
be ﬁxed and let t
f
≥ t
0
be a ﬁxed time. We are concerned with the
83
84 CHAPTER 7. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL
decision problem (7.2).
Maximize c

x(t
f
),
subject to
dynamics: ˙ x(t) = A(t)x(t) +B(t)u(t) , t
0
≤ t ≤ t
f
,
initial condition: x(t
0
) = x
0
,
ﬁnal condition: x(t
f
) ∈ R
n
,
control constraint: u() ∈ | .
(7.2)
Deﬁnition: (i) For any piecewise continuous function u() : [t
0
, t
f
] → R
p
, for any z ∈ R
n
, and
any t
0
≤ t
1
≤ t
2
≤ t
f
let
φ(t
2
, t
1
, z, u)
denote the state of (7.1) at time t
2
, if a time t
1
it is in state z, and the control u() is applied.
(ii) Let
K(t
2
, t
1
, z) = ¦φ(t
2
, t
1
, z, u)[u ∈ |¦ .
Thus, K(t
2
, t
1
, z) is the set of states reachable at time t
2
starting at time t
1
in state z and using
admissible controls. We call K the reachable set.
Deﬁnition: Let Φ(t, τ), t
0
≤ τ ≤ t ≤ t
f
, be the transition-matrix function of the homogeneous
part of (7.1), i.e., Φ satisﬁes the differential equation
∂Φ
∂t
(t, τ) = A(t)Φ(t, τ) ,
and the boundary condition
Φ(t, t) ≡ I
n
.
The next result is well-known. (See Desoer [1970].)
Lemma 1: φ(t
2
, t
1
, z, u) = Φ(t
2
, t
1
)z +

t
2
t
1
Φ(t
2
, τ)B(τ)u(τ)dτ.
Exercise 1: (i) Assuming that Ω is convex, show that | is a convex set. (ii) Assuming that | is
convex show that K(t
2
, t
1
, z) is a convex set. (It is a deep result that K(t
2
, t
1
, z) is convex even if
Ω is not convex (see Neustadt [1963]), provided we include in | any measurable function
u : [t
0
, ∞) → Ω.)
Deﬁnition: Let K ⊂ R
n
, and let x

∈ K. We say that c is the outward normal to a hyperplane
supporting K at x

if c = 0, and
c

x

≥ c

x for all x ∈ K .
The next result gives a geometric characterization of the optimal solutions of (2).
Lemma 2: Suppose c = 0. Let u

() ∈ | and let x

(t) = φ(t, t
0
, x
0
, u

). Then u

is an optimal
solution of (2) iff
(i) x

(t
f
) is on the boundary of K = K(t
f
, t
0
, x
0
), and
(ii) c is the outward normal to a hyperplane supporting K at x

. (See Figure 7.1.)
Proof: Clearly (i) is implied by (ii) because if x

(t
f
) is in the interior of K there is δ > 0 such
that (x

(t
f
) +δc) ∈ K; but then
7.1. THE LINEAR OPTIMAL CONTROL PROBLEM 85
x
3
c
x
2
x
1
c
x

(t
f
)
π

= ¦x[c

x = c

x

(t
f

K
Figure 7.1: c is the outward normal to π

supporting K at x

(t
f
)
.
c

(x

(t
f
) +δc) = c

x

(t
f
) +δ[c[
2
> c

x

(t
f
) .
Finally, from the deﬁnition of K it follows immediately that u

is optimal iff c

x

(t
f
) ≥ c

x for all
x ∈ K . ♦
The result above characterizes the optimal control u

in terms of the ﬁnal state x

(t
f
). The beauty
and utility of the theory lies in the following result which translates this characterization directly in
terms of u

.
Theorem 1: Let u

() ∈ | and let x

(t) = φ(t, t
0
, x
0
, u

), t
0
≤ t ≤ t
f
. Let p

(t) be the solution
of (7.3) and (7.4):

(t) = −A

(t)p

(t) , t
0
≤ t ≤ t
f
. (7.3)
ﬁnal condition: p

(t
f
) = c . (7.4)
Then u

() is optimal iff
(p

(t))

B(t)u

(t) = sup¦(p

(t))

B(t)v[v ∈ Ω¦ , (7.5)
for all t ∈ [t
0
, t
f
], except possibly for a ﬁnite set.
Proof: u

() is optimal iff for every u() ∈ |
(p

(t
f
))

[Φ(t
f
, t
0
)x
0
+

t
f
t
0
Φ(t
f
, τ)B(τ)u

(τ)dτ]
≥ (p

(t
f
))

[Φ(t
f
, t
0
)x
0
+

t
f
t
0
Φ(t
f
, τ)B(τ)u(τ)dτ] ,
which is equivalent to (7.6).

t
f
t
0
(p

(t
f
))

Φ(t
f
, τ)B(τ)u

(τ)dτ

t
f
t
0
(p

(t
f
))

Φ(t
f
, τ)B(τ)u(τ)dτ
(7.6)
86 CHAPTER 7. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL
Now by properties of the adjoint equation we know that p

(t))

= (p

(t
f
))

Φ(t
f
, t) so that (7.6) is
equivalent to (7.7),

t
f
t
0
(p

(τ))

B(τ)u

(τ)dτ ≥

t
f
t
0
(p

(τ))

B(τ)u(τ)dτ, (7.7)
and the sufﬁciency of (7.5) is immediate.
To prove the necessity let D be the ﬁnite set of points where the function B() or u

() is discon-
tinuous. We shall show that if u

() is optimal then (7.5) is satisﬁed for t ∈ D. Indeed if this is not
the case, then there exists t

∈ [t
0
, t
f
], t

∈ D, and v ∈ Ω such that
(p

(t

))

B(t

)u

(t

) < (p

(t

))

B(t

)v ,
and since t

is a point of continuity of B() and u

(), it follows that there exists δ > 0 such that
(p

(t))

B(t)u

(t) < (p

(t))

B(t)v, for [t −t

[ < δ . (7.8)
Deﬁne ˜ u() ∈ | by
˜ u(t) =

v [t −t

[ < δ, t ∈ [t
0
, t
f
]
u

(t) otherwise .
Then (7.8) implies that

t
f
t
0
(p

(t))

B(t)˜ u(t)dt >

t
f
t
0
(p

(t))

B(t)u

(t)dt .
But then from (7.7) we see that u

() cannot be optimal, giving a contradiction. ♦
Corollary 1: For t
0
≤ t
1
≤ t
2
≤ t
f
,
(p

(t
2
))x

(t
2
) ≥ (p

(t
2
))

x for all x ∈ K(t
2
, t
1
, x

(t
1
)). (7.9)
Exercise 2: Prove Corollary 1.
Remark 1: The geometric meaning of (7.9) is the following. Taking t
1
= t
0
in (7.9), we see that if
u

() is optimal, i.e., if c = p

(t
f
) is the outward normal to a hyperplane supporting K(t
f
, t
0
, x
0
)
at x

(t
f
), then x

(t) is on the boundary of K(t, t
0
, x
0
) and p

(t) is the normal to a hyperplane
supporting K(t, t
0
, x
0
) at x

(t). This normal is obtained by transporting backwards in time, via
the adjoint differential equation, the outward normal p

(t
f
) at time t
f
. The situation is illustrated
in Figure 7.2.
Remark 2: If we deﬁne the Hamiltonian function H by
H(t, x, u, p) = p

(A(t)x +B(t)u) ,
and we deﬁne M by
M(t, x, p) = sup¦H(t, x, u, p)[u ∈ Ω¦,
then (7.5) can be rewritten as
H(t, x

(t), u

(t), p

(t)) = M(t, x

(t), p

(t)) . (7.10)
This condition is known as the maximum principle.
7.2. MORE GENERAL BOUNDARY CONDITIONS 87
Exercise 3: (i) Show that m(t) = M(t, x

(t), p

(t)) is a Lipschitz function of t. (ii) If A(t), B(t)
are constant, show that m(t) is constant. (Hint: Show that (dm/dt) ≡ 0.)
The next two exercises show how we can obtain important qualitative properties of an optimal
control.
Exercise 4: Suppose that Ω is bounded and closed. Show that there exists an optimal control u

()
such that u

(t) belongs to the boundary of Ω for all t.
Exercise 5: Suppose Ω = [α, β], so that B(t) is an n 1 matrix. Suppose that A(t) ≡ A and
B(t) ≡ B are constant matrices and A has n real eigenvalues. Show that there is an optimal
control u

() and t
0
≤ t
1
≤ t
2
≤ . . . ≤ t
n
≤ t
f
such that u

(t) ≡ α or β on [t
i
, t
i+1
), 0 ≤ i ≤ n.
(Hint: ﬁrst show that (p

(t))

B = γ
1
exp(δ
1
t) +. . . +γ
n
exp(δ
n
(t)) for some γ
i
, δ
i
in R.)
Exercise 6: Assume that K(t
f
, t
0
, x
0
) is convex (see remark in Exercise 1 above). Let
f
0
: R
n
→ R be a differentiable function and suppose that the objective function in (7.2) is
f
0
(x(t
f

x(t
f
). Suppose u

() is an optimal control. Show that u

() satisﬁes the
maximum principle (7.10) where p

() is the solution of the adjoint equation (7.3) with the ﬁnal
condition
p

(t
f
) = f
0
(x

(t
f
)) .
Also show that this condition is sufﬁcient for optimality if f
0
is concave. (Hint: Use Lemma 1 of
5.1.1 to show that if u

() is optimal, then f
0x
(x

(t
f
)(x

(t
f
) −x) ≤ for all x ∈ K(t
f
, t
0
, x
0
).)
7.2 More General Boundary Conditions
We consider the following generalization of (7.2). The notion of the previous section is retained.
Maximize c

x(t
f
)
subject to
dynamics: ˙ x(t) = A(t)x(t) +B(t)u(t), t
0
≤ t ≤ t
f
,
initial condition: G
0
x(t
0
) = b
0
,
ﬁnal condition: G
f
x(t
f
) = b
f
,
control constraint: u() ∈ |, i.e., ¯ : [.

, .
{
] → ⊗ and
u()piecewise continuous.
(7.11)
In (7.11) G
0
and G
f
are ﬁxed matrices of dimensions
0
xn and
f
n respectively, while b
0

R

0
, b
f
∈ R

f
are ﬁxed vectors.
We will analyze the problem in the same way as before. That is, we ﬁrst characterize optimality
in terms of the state at the ﬁnal time, and then translate these conditions in terms of the control. For
convenience let
T
0
= ¦z ∈ R
n
[G
0
z = b
0
¦ ,
T
f
= ¦z ∈ R
n
[G
f
z = b
f
¦ .
Deﬁnition: Let p ∈ R
n
. Let z

∈ T
0
. We say that p is orthogonal to T
0
at z

and we write
p ⊥ T
0
(z

) if
88 CHAPTER 7. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL
R
n
x
0
=
x

(
t
0
)
=
K
(
t
0
,
t
0
,
x
0
) R
n
p

(
t
1
)
x

(
t
1
)
K
(
t
1
,
t
0
,
x
0
)
R
n
p

(
t
2
)
x

(
t
2
)
K
(
t
2
,
t
0
,
x
0
)
R
n
p

(
t
f
)
=
c
x

(
t
f
)
K
(
t
f
,
t
0
,
x
0
)
t
t
f
t
2
t
1
t
0
Figure 7.2: Illustration of (7.9) for t
1
= t
0
.
7.2. MORE GENERAL BOUNDARY CONDITIONS 89
p

(z −z

) = 0 for all z ∈ T
0
.
Similarly if z

∈ T
f
, p ⊥ T
f
(z

) if
p

(z −z

) = 0 for all z ∈ T
f
.
Deﬁnition: Let X(t
f
) = ¦Φ(t
f
, t
0
)z +w[z ∈ T
0
, w ∈ K(t
f
, t
0
, 0)¦.
Exercise 1: X(t
f
) = ¦Φ(t
f
, t
0
, z, u)[z ∈ T
0
, u() ∈ |¦.
Lemma 1: Let x

(t
0
) ∈ T
0
and u

() ∈ |. Let x

(t) = φ(t, t
0
, x

(t
0
), u

), and suppose that
x

(t
f
) ∈ T
f
.
(i) Suppose the Ω is convex. If u

() is optimal, there exist ˆ p
0
∈ R, ˆ p
0
≥ 0 and ˆ p ∈ R
n
, not both
zero, such that
(ˆ p
0
c + ˆ p)

x

(t
f
) ≥ (ˆ p
0
c + ˆ p)

x for all x ∈ X(t
f
) , (7.12)
ˆ p ⊥ T
f
(x

(t
f
)) , (7.13)
[Φ(t
f
, t
0
)]

(ˆ p
0
c + ˆ p) ⊥ T
0
(x

(t
0
)) . (7.14)
(ii) Conversely if there exist ˆ p
0
> 0, and ˆ p such that (7.12) and (7.13) are satisﬁed, then u

() is
optimal and (7.14) is also satisﬁed.
Proof: Clearly u

() is optimal iff
c

x

(t
f
) ≥ c

x for all x ∈ X(t
f
) ∩ T
f
. (7.15)
(i) Suppose that u

() is optimal. In R
1+m
deﬁne sets S
1
, S
2
by
S
1
= ¦(r, x)[r > c

x

(t
f
), x ∈ T
f
¦ , (7.16)
S
2
= ¦(r, x)[r = c

x , x ∈ X(t
f
)¦ . (7.17)
First of all S
1
∩S
2
= φ because otherwise there exists x ∈ X(t
f
) ∩T
f
such that c

x > c

x

(t
f
)

() by (7.15).
Secondly, S
1
is convex since T
f
is convex. Since Ω is convex by hypothesis it follows by Exercise
1 of Section 1 that S
2
is convex.
But then by the separation theorem for convex sets (see 5.5) there exists ˆ p
0
∈ R, ˆ p ∈ R
n
, not
both zero, such that
ˆ p
0
r
1
+ ˆ p

x
1
≥ ˆ p
0
r
2
+ ˆ p

x
2
for all (r
i
, x
i
) ∈ S
i
, i = 1, 2. (7.18)
In particular (7.18) implies that
ˆ p
0
r + ˆ p

x

(t
f
) ≥ ˆ p
0
c

x + ˆ p

x for all x ∈ X(t
f
), r > c

x

(t
f
). (7.19)
Letting r → ∞ we conclude that (7.19) can hold only if ˆ p
0
≥ 0. On the other hand letting r →
c

x

(t
f
) we see that (7.19) can hold only if
ˆ p
0
c

x

(t
f
) + ˆ p

x

(t
f
) ≥ ˆ p
0
c

x + ˆ p

x for all x ∈ X(t
f
) , (7.20)
which is the same as (7.12). Also from (7.18) we get
90 CHAPTER 7. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL
ˆ p
0
r + ˆ p/x ≥ ˆ p
0
c

x

(t
f
) + ˆ p

x

(t
f
) for all r > c

x

(t
f
), x ∈ T
f
,
which can hold only if
ˆ p
1
c

x

(t
f
) + ˆ p

x ≥ ˆ p
0
c

x

(t
f
) + ˆ p

x

(t
f
) for all x ∈ T
f
,
or
ˆ p

(x −x

(t
f
)) ≥ 0 for all x ∈ T
f
(7.21)
But ¦x −x

(t
f
)[x ∈ T
f
¦ = ¦z[G
f
z = 0¦ is a subspace of R
n
, so that (7.21) can hold only if
ˆ p

(x −x

(t
f
)) = 0 for all x ∈ T
f
,
which is the same as (7.13). Finally (7.12) always implies (7.14), because by the deﬁnition of X(t
f
)
and Exercise 1, ¦Φ(t
f
, t
0
)(z − x

(t
0
)) + x

(t
f
)¦ ∈ X(t
f
) for all z ∈ T
0
, so that from (7.12) we
get
0 ≥ (ˆ p
0
c + ˆ p)

Φ(t
f
, t
0
)(z −x

(t
0
)) for all z ∈ T
0
,
which can hold only if (7.14) holds.
(ii) Now suppose that ˆ p
0
> 0 and ˆ p are such that (7.12), (7.13) are satisﬁed. Let ˜ x ∈ X(t
f
) ∩ T
f
.
Then from (7.13) we conclude that
ˆ p

x

(t
f
) = ˆ p

˜ x ,
so that from (7.12) we get
ˆ p
0
c

x

(t
f
) ≥ ˆ p
0
c

˜ x ;
but then by (7.15) u

() is optimal. ♦
Remark 1: If it is possible to choose ˆ p
0
> 0 then ˆ p
0
= 1, ˆ p = (ˆ p/ˆ p
0
) will also satisfy (7.12),
(7.13), and (7.14). In particular, in part (ii) of the Lemma we may assume ˆ p
0
= 1.
Remark 2: it would be natural to conjecture that in part (i) ˆ p
0
may be chosen > 0. But in Figure
7.3 below, we illustrate a 2-dimensional situation where T
0
= ¦x
0
¦, T
f
is the vertical line, and
T
f
∩ X(t
f
) consists of just one vector. It follows that the control u

() ∈ | for which
x

(t
f
) = φ(t
f
, t
0
, x
0
, u

) ∈ T
f
is optimal for any c. Clearly then for some c (in particular for the
c in Figure 7.3) we are forced to set ˆ p
0
= 0. In higher dimensions the reasons may be more
complicated, but basically if T
f
is “tangent” to X(t
f
) we may be forced to set ˆ p
0
= 0 (see
Exercise 2 below). Finally, we note that part (i) is not too useful if ˆ p
0
= 0, since then (7.12), (7.13),
and (7.14) hold for any vector c whatsoever. Intuitively ˆ p
0
= 0 means that it is so difﬁcult to satisfy
the initial and ﬁnal boundary conditions in (7.11) that optimization becomes a secondary matter.
Remark 3: In (i) the convexity of Ω is only used to guarantee that K(t
f
, t
0
, 0) is convex. But it is
known that K(t
f
, t
0
, 0) is convex even if Ω is not (see Neustadt [1963]).
Exercise 2: Suppose there exists z in the interior of X(t
f
) such that z ∈ T
f
. Then in part (i) we
must have ˆ p
0
> 0.
We now translate the conditions obtained in Lemma 1 in terms of the control u

.
7.2. MORE GENERAL BOUNDARY CONDITIONS 91
Theorem 1: Let x

(t
0
) ∈ T
0
and u

() ∈ |. Let x

(t) = φ(t, t
0
, x

(t
0
), u

) and suppose that
x

(t
f
) ∈ T
f
.
(i) Suppose that Ω is convex. If u

() is optimal for (7.11), then there exist a number p

0
≥ 0, and a
function p

: [t
0
, t
f
] → R
n
, not both identically zero, satisfying

(t) = −A

(t)p

(t) , t
0
≤ t ≤ t
f
(7.22)
initial condition: p

(t
0
)⊥T
0
(x

(t
0
)) (7.23)
ﬁnal condition: (p

(t
f
) −p

0
c)⊥T
f
(x

(t
f
)) , (7.24)
and the maximum principle
H(t, x

(t), u

(t), p

(t)) = M(t, x

(t), p

(t)) , (7.25)
holds for all t ∈ [t
0
, t
f
] except possibly for a ﬁnite set.
(ii) Conversely suppose there exist p

0
> 0 and p

() satisfying (7.22), (7.23), (7.24), and (7.25).
Then u

() is optimal.
[Here
H(t, x, u, p) = p

(A(t)x +B(t)u), M(t, x, p) = sup¦H(t, x, v, p)[v ∈ Ω¦.]
Proof: A repetition of a part of the argument in the proof of Theorem 1 of Section 1 show that if p

satisﬁes (7.22), then (7.25) is equivalent to (7.26):
(p

(t
f
))

x

(t
f
) ≥ (p

(t
f
))

x for all x ∈ K(t
f
, t
0
, x

(t
0
)) . (7.26)
(i) Suppose u

() is optimal and Ω is convex. Then by Lemma 1 there exist ˆ p ≥ 0, ˆ p ∈ R
n
, not
both zero, such that (7.12), (7.13) and (7.14) are satisﬁed. Let p

0
= ˆ p
0
and let p

() be the solution
of (7.22) with the ﬁnal condition
p

(t
f
) = p

0
c + ˆ p = ˆ p
0
c + ˆ p .
Then (7.14) and (7.13) are respectively equivalent to (7.23) and (7.24), whereas since K(t
f
, t
0
, x

(t
0
)) ⊂
X(t
f
), (7.26) is implied by (7.12).
(ii) Suppose p

0
> 0 and (7.22), (7.23), (7.24), and (7.26) are satisﬁed. Let ˆ p
0
= p

0
and ˆ p =
p

(t
f
) −p

0
c, so that (7.24) becomes equivalent to (7.13). Next if x ∈ X(t
f
) we have
(ˆ p
0
c + ˆ p)

x = (p

(t
f
))

x
= (p

(t
f
))

(Φ(t
f
, t
0
)z +w) ,
92 CHAPTER 7. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL
.
x
0
=
T
0
T
f
x

(
t
f
)
=
X
(
t
f
)
¸
T
f
K
(
t
f
,
t
0
,
x
0
)
=
X
(
t
f
)
c
t
Figure 7.3: Situation where ˆ p
0
= 0
7.2. MORE GENERAL BOUNDARY CONDITIONS 93
for some z ∈ T
0
and some w ∈ K(t
f
, t
0
, 0). Hence
(ˆ p
0
c + ˆ p)

x = (p

(
f
))

Φ(t
f
, t
0
)(z −x

(t
0
))
+(p

(t
f
))

(w +φ(t
f
, t
0
)x

(t
0
))
= (p

(t
0
))

(z −x

(t
0
))
+(p

(t
f
))

(w + Φ(t
f
, t
0
)x

(t
0
)) .
But by (7.23) the ﬁrst term on the right vanishes, and since (w+φ(t
f
, t
0
)x

(t
0
)) ∈ K(t
f
, t
0
, x

(t
0
)),
it follows from (7.26) that the second term is bounded by (p

(t
f
))

x

(t
f
). Thus
(ˆ p
0
c + ˆ p)

x

(t
f
) ≥ (ˆ p
0
c + ˆ p)

x for all x ∈ X(t
f
) ,
and so u

() is optimal by Lemma 1. ♦
Exercise 3: Suppose that the control constraint set is Ω(t) which varies continuously with t, and
we require that u(t) ∈ Ω(t) for all t. Show that Theorem 1 also holds for this case where, in (7.25),
M(t, x, p) =sup¦H(t, x, v, p)[v ∈ Ω(t)¦.
Exercise 4: How would you use Exercise 3 to solve Example 3 of Chapter 1?
94 CHAPTER 7. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL
Chapter 8
SEQUENTIAL DECISION PROBLEMS:
CONTINUOUS-TIME OPTIMAL
CONTROL OF NONLINEAR SYSTEMS
We now present a sweeping generalization of the problem studied in the last chapter. Unfortunately
we are forced to omit the proofs of the results since they require a level of mathematical sophis-
tication beyond the scope of these Notes. However, it is possible to convey the main ideas of the
proofs at an intuitive level and we shall do so. (For complete proofs see (Lee and Markus [1967]
or Pontryagin, et al., [1962].) The principal result, which is a direct generalization of Theorem 1 of
7.2 is presented in Section 1. An alternative form of the objective function is discussed in Section
2. Section 3 deals with the minimum-time problem and Section 4 considers the important special
case of linear systems with quadratic cost. Finally, in Section 5 we discuss the so-called singular
case and also analyze Example 4 of Chapter 1.
8.1 Main Results
8.1.1 Preliminary results based on differential equation theory.
We are interested in the optimal control of a system whose dynamics are governed by the nonlinear
differential equation
˙ x(t) = f(t, x, (t), u(t)) , t
0
≤ t ≤ t
f
, (8.1)
where x(t) ∈ R
n
is the state and u(t) ∈ R
p
is the control. Suppose u

() is an optimal control
and x

() is the corresponding trajectory. In the case of linear systems we obtained the necessary
conditions for optimality by comparing x

() with trajectories x() corresponding to other admis-
sible controls u(). This comparison was possible because we had an explicitly characterization of
x() in terms of u(). Unfortunately when f is nonlinear such a characterization is not available.
Instead we shall settle for a comparison between the trajectory x

() and trajectories x() obtained
by perturbing the control u

() and the initial condition x

(t
0
). We can then estimate the difference
between x() and x

() by the solution to a linear differential equation as shown in Lemma 1 below.
But ﬁrst we need to impose some regularity conditions on the differential equation (8.1). We assume
throughout that the function f : [t
0
, t
f
] R
n
R
p
→ R
n
satisﬁes the following conditions:
95
96 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
1. for each ﬁxed t ∈ [t
0
, t
f
], f(t, , ) : R
n
xR
p
→ R
n
is continuously differentiable in the
remaining variables (x, u),
2. except for a ﬁnite subset D ⊂ [t
0
, t
f
], the functions f, f
x
, f
u
are continuous on [t
0
, t
f
] R
n

R
p
, and
3. for every ﬁnite α, there exist ﬁnite number β and γ such that
[f(t, x, u)[ ≤ β +γ[x[ for all t ∈ [t
0
, t
f
], x ∈ R
n
, u ∈ R
p
with [u[ ≤ α .
The following result is proved in every standard treatise on differential equations.
Theorem 1: For every z ∈ R
n
, for every t
1
∈ [t
0
, t
f
], and every piecewise continuous function
u() : [t
0
, t
f
] → R
p
, there exists a unique solution
x(t) = φ(t, t
1
, z, u()) , t
1
≤ t ≤ t
f
,
of the differential equation
˙ x(t) = f(t, x(t), u(t)) , t
1
≤ t ≤ t
f
,
satisfying the initial condition
x(t
1
) = z .
Furthermore, for ﬁxed t
1
≤ t
2
in [t
0
, t
f
] and ﬁxed u(), the function φ(t
2
, t
1
, , u()) : R
n
→ R
n
is
differentiable. Moreover, the n n matrix-valued function Φ deﬁned by
Φ(t
2
, t
1
, z, u()) =
∂φ
∂z
(t
2
, t
1
, z, u())
is the solution of the linear homogeneous differential equation
∂Φ
∂t
(t, t
1
, z, u, ()) = [
∂f
∂x
(t, x, (t), u(t))]Φ(t, t
1
, z, u()), t
1
≤ t ≤ t
f
,
and the initial condition
Φ(t
1
, t
1
, z, u()) = I
n
.
Now let Ω ⊂ R
p
be a ﬁxed set and let | be set of all piecewise continuous functions u() :
[t
0
, t
f
] → Ω. Let u

() ∈ | be ﬁxed and let D

be the set of discontinuity points of u

(). Let
x

0
∈ R
n
be a ﬁxed initial condition.
Deﬁnition: π = (t
1
, . . . , t
m
;
1
, . . . ,
m
; u
1
, . . . , u
m
) is said to be a perturbation data for u

() if
1. m is a nonnegative integer,
2. t
0
< t
1
< t
2
< . . . t
m
< t
f
, and t
i
∈ D

¸
D, i = 1, . . . , m (recall that D is the set of
discontinuity points of f),
3.
i
≥ 0, i = 1, . . . , m, and
4. u
i
∈ Ω, i = 1, . . . , m.
8.1. MAIN RESULTS 97
Let ε(π) > 0 be such that for 0 ≤ ε ≤ ε(π) we have [t
i
− ε
i
, t
i
] ⊂ [t
0
, t
f
] for all i, and
[t
i
−ε
i
, t
i
]
¸
[t
j
−ε
j
, t
j
] = φ for i = j. Then for 0 ≤ ε ≤ ε(π),the perturbed control u
(π,ε)
() ∈ |
corresponding to π is deﬁned by
u
(π,ε)
(t) =

u
i
for all t ∈ [t
i
−ε
i
, t
i
] , i = 1, . . . , m
u

(t) otherwise .
Deﬁnition: Any vector ξ ∈ R
n
is said to be a perturbation for x

0
, and a function x
(ξ,ε)
deﬁned for
ε > 0 is said to be a perturbed initial condition if
lim
ε→0
x
(ξ,ε)
= x

0
,
and
lim
ε→0
1
ε
(x
(ξ,ε)
−x

0
) = ξ .
Now let x

(t) = φ(t, t
0
, x

0
, u

()) and let x
ε
(t) = φ(t, t
0
, x
(ξ,ε)
, u
(π,ε)
()). Let Φ(t
2
, t
1
) =
Φ(t
2
, t
1
, x

(t
1
), u

()). The following lemma gives an estimate of x

(t) − x
ε
(t). The proof of the
lemma is a straightforward exercise in estimating differences of solutions to differential equations,
and it is omitted (see for example (Lee and Markus [1967])).
Lemma 1: lim
ε→0
[x
ε
(t) −x

(t) −εh
(π,ε)
(t)[ = 0 for t ∈ [t
0
, t
1
], where h
(π,ε)
() is given by
h
(π,ε)
(t) = Φ(t, t
0
)ξ , t ∈ [t
0
, t
1
)
= Φ(t, t
0
)ξ + Φ(t, t
1
)[f(t
1
, x

(t
1
), u
1
) −f(t
1
, x

(t
1
), u

(t
1
))]
1
, t ∈ [t
1
, t
2
)
= Φ(t, t
0
)ξ +
i
¸
j=1
Φ(t, t
j
)[f(t
j
, x

(t
j
), u
j
) −f(t
j
, x

(t
j
), u

(t
j
))]
j
, t ∈ [t
i
, t
i+1
)
= Φ(t, t
0
)ξ +
m
¸
j=1
Φ(t, t
m
)[f(t
j
, x

(t
j
), u
j
) −f(t
j
, x

(t
j
), u

t
j
))]
j
, t ∈ [t
m
, t
f
] .
(See Figure 8.1.)
We call h
(π,ξ)
() the linearized (trajectory) perturbation corresponding to (π, ξ).
Deﬁnition: For z ∈ R
n
, t ∈ [t
0
, t
f
] let
K(t, t
0
, z) = ¦φ(t, t
0
, z, u())[u() ∈ |¦
be the set of states reachable at time t, starting at time t
0
in state z, and using controls u() ∈ |.
Deﬁnition: For each t ∈ [t
0
, t
f
], let
Q(t) = ¦h
(π,0)
(t)[πis a perturbation data for u

(), and
h
(π,0)
()is the linearized perturbation
corresponding to(π, 0)¦ .
Remark: By Lemma 1 (x

(t)+εh
(π,ξ)
) belongs to the set K(t, t
0
, x
(ξ,ε)
) up to an error of order o(ε).
In particular for ξ = 0, the set x

(t) + Q(t) can serve as an approximation to the set K(t, t
0
, x

0
).
More precisely we have the following result which we leave as an exercise.
98 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
| | | | | | |
|
|
| |
u
u
1
u
(πε)
()
u
2
u

()
u
3
ε
3
ε
2
ε
1
t
0
t
1
t
2
t
3
t
f
x
x
(
ξ, ε)
x
ε
()
x

()
εh
πξ
t
1
t
2
t
3
t
f
Figure 8.1: Illustration for Lemma 1.
Exercise 1: (Recall the deﬁnition of the tangent cone in 5.1.1.) Show that
Q(t) ⊂ C(K(t, t
0
, x

0
), x

(t)) . (8.2)
We can now prove a generalization of Theorem 1 of 7.1.
Theorem 2: Consider the optimal control problem (8.3):
Maximize ψ(x(t
f
))
subject to
dynamics: ˙ x(t) = f(t, x(t), u(t)) , t
0
≤ t ≤ t
f
,
initial condition: x(t
0
) = x

0
,
ﬁnal condition: x(t
f
) ∈ R
n
,
control constraint: u() ∈ |, i.e., u : [t
0
, t
f
] → Ω and
u() piecewise continuous ,
(8.3)
where ψ : R
n
→ R is differentiable and f satisﬁes the conditions listed earlier.
Let u

() ∈ | be an optimal control and let x

(t) = φ(t, t
0
, x

0
, u

()), t
0
≤ t ≤ t
f
, be the
corresponding trajectory. Let p

(t), t
0
≤ t ≤ t
f
, be the solution of (8.4) and (8.5):

(t) = −[
∂f
∂x
(t, x

(t), u

(t))]

p

(t), t
0
≤ t ≤ t
f
, (8.4)
8.1. MAIN RESULTS 99
ﬁnal condition: p

(t
f
) = ψ(x

(t
f
)) . (8.5)
Then u

() satisﬁes the maximum principle
H(t, x

(t), u

(t), p

(t)) = M(t, x

(t), p

(t)) (8.6)
for all t ∈ [t
0
, t
f
] except possibly for a ﬁnite set. [Here H(t, x, u, p) = p

f(t, x, u, ), M(t, x, p) =
sup¦H(t, x, v, p)[v ∈ Ω¦].
Proof: Since u

() is optimal we must have
ψ(x

(t
f
)) ≥ ψ(z) for all z ∈ K(t
f
, t
0
, x

0
) ,
and so by Lemma 1 of 5.1.1
ψ(x

(t
f
))h ≤ 0 for all h ∈ C(K(t
f
, t
0
, x

0
), x

(t
f
)) ,
and in particular from (8.2)
ψ
x
(x

(t
f
))h ≤ 0 for all h ∈ Q(t
f
) . (8.7)
Now suppose that (8.6) does not hold from some t

∈ D

∪ D. Then there exists v ∈ Ω such that
p

(t

)

[f(t

, x(t

), v) −f(t

, x(t

), u

(t

))] > 0 . (8.8)
If we consider the perturbation data π = (t

; 1; v), then (8.8) is equivalent to
p

(t

)

h
(π,0)
(t

) > 0 . (8.9)
Now from (8.4) we can see that p

(t

)

= p

(t
f
)

Φ(t
f
, t

). Also h
(π,0)
(t
f
) = Φ(t
f
, t

)h
(π,0)
(t

)
so that (8.9) is equivalent to
p

(t
f
)

h
(π,0)
(t
f
) > 0
8.1.2 More general boundary conditions.
In Theorem 2 the initial condition is ﬁxed and the ﬁnal condition is free. The problem involving
more general boundary conditions is much more complicated and requires more reﬁned analysis.
Speciﬁcally, Lemma 1 needs to be extended to Lemma 2 below. But ﬁrst we need some simple
properties of the sets Q(t) which we leave as exercises.
Exercise 2: Show that
(i) Q(t) is a cone, i.e., if h ∈ Q(t) and λ ≥ 0, then λh ∈ Q(t),
(ii) for t
0
≤ t
1
≤ t
2
≤ t
f
, Φ(t
2
, t
1
)Q(t
1
) ⊂ Q(t
2
) .
Deﬁnition: Let C(t) denote the closure of Q(t).
Exercise 3: Show that
(i) C(t) is a convex cone,
(ii) for t
0
≤ t
1
≤ t
2
≤ t
f
, Φ(t
2
, t
1
)C(t
1
) ⊂ C(t
2
) .
Remark: From Lemma 1 we know that if h ∈ C(t) then (x

(t) +εh) belongs to K(t, t
0
, x

(t
0
)) up
to an error of order o(ε). Lemma 2, below, asserts further that if h is in the interior of C(t) then in
fact (x

(t) +εh) ∈ K(t, t
0
, x

(t
0
)) for ε > 0 sufﬁciently small. The proof of the lemma depends
upon some deep topological results and is omitted. Instead we offer a plausibility argument.
Lemma 2: Let h belong to the interior of the cone C(t). Then for all ε > 0 sufﬁciently small,
100 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
(x

(t) +εh) ∈ K(t, t
0
, x

0
) . (8.10)
Plausibility argument. (8.10) is equivalent to
εh ∈ K(t, t
0
, x

(t
0
)) −¦x

(t)¦ , (8.11)
where we have moved the origin to x

(t). The situation is depicted in Figure 8.2.
0
ˆ
C(ε)
ˆ
K(ε)
o(ε)
K(t
1
, t
0
, x

) −¦x

(t)¦
h
C(t)
δε
εh
Figure 8.2: Illustration for Lemma 2.
Let
ˆ
C(ε) be the cross-section of C(t) by a plane orthogonal to h and passing through εh. Let
ˆ
K(ε) be the cross-section of K(t, t
0
, x

0
) −¦x

(t
0
)¦ by the same plane. We note the following:
(i) by Lemma 1 the distance between
ˆ
C(ε) and
ˆ
K(ε) is of the order o(ε);
(ii) since h is in the interior of C(t), the minimum distance between εh and
ˆ
C(ε) is δε where
δ > 0 is independent of ε.
Hence for ε > 0 sufﬁciently small εh must be trapped inside the set
ˆ
K(ε).
(This would constitute a proof except that for the argument to work we need to show that there
are no “holes” in
ˆ
K(ε) through which εh can “escape.” The complications in a rigorous proof arise
precisely from this drawback in our plausibility argument.) ♦
Lemmas 1 and 2 give us a characterization of K(t, t
0
, x

0
) in a neighborhood of x

(t) when we
perturb the control u

() leaving the initial condition ﬁxed. Lemma 3 extends Lemma 2 to the case
when we also allow the initial condition to vary over a ﬁxed surface in a neighborhood of x

0
.
Let g
0
: R
n
→ R

0
be a differentiable function such that the
0
n matrix g
0
x
(x) has rank

0
for all x. Let b
0
∈ R
n
be ﬁxed and let T
0
= ¦x[g
0
(x) − b
0
¦. Suppose that x

0
∈ T
0
and let
T
0
(x

0
) = ¦ξ[g
0
x
(x

0
)ξ = 0¦. Thus, T
0
(x

0
) + ¦x

0
¦ is the plane through x

0
tangent to the surface
T
0
. The proof of Lemma 3 is similar to that of Lemma 2 and is omitted also.
Lemma 3: Let h belong to the interior of the cone ¦C(t)+Φ(t, t
0
)T
0
(x

0
)¦. For ε ≥ 0 let h(ε) ∈ R
n
be such that lim h(ε) = 0, and lim
ε→0
(
1
ε
)h(ε) = h. Then for ε > 0 sufﬁciently small there exists
x
0
(ε) ∈ T
0
such that
(x

(t) +h(ε)) ∈ K(t, t
0
, x
0
(ε)) .
8.1. MAIN RESULTS 101
We can now prove the main result of this chapter. We keep all the notation introduced above.
Further, let g
f
: R
n
→ R

f
be a differentiable function such that g
f
x
(x) has rank
f
for all x.
Let b
f
∈ R
n
be ﬁxed and let T
f
= ¦x[g
f
(x) − b
f
¦. Finally, if x

(t
f
) ∈ T
f
let T
f
(x

(t
f
)) =
¦ξ[g
f
x
(x

(t
f
))ξ = 0¦.
Theorem 3: Consider the optimal control problem (8.12):
Maximize ψ(x(t
f
))
subject to
dynamics: ˙ x(t) = f(t, x(t), u(t)) , t
0
≤ t ≤ t
f
,
initial conditions: g
0
(x(t
0
)) = b
0
,
ﬁnal conditions: g
f
(x(t
f
)) = b
f
,
control constraint: u() ∈ |, i.e., u : [t
0
, t
f
] → Ω and
u() piecewise continuous .
(8.12)
Let u

() ∈ |, let x

0
∈ T
0
and let x

(t) = φ(t, t
0
, x

0
, u

()) be the corresponding trajectory.
Suppose that x

(t
f
) ∈ T
f
, and suppose that (u

(), x

0
) is optimal. Then there exist a number
p

0
≥ 0, and a function p

: [t
0
, t
f
] → R
n
, not both identically zero, satisfying

(t) = −[
∂f
∂x
(t, x

(t), u

(t))]

p

(t), t
0
≤ t ≤ t
f
, (8.13)
initial condition: p

(t
0
)⊥T
0
(x

0
) , (8.14)
ﬁnal condition: (p

(t
f
) −p

0
∇ψ(x

(t
f
)))⊥T
f
(x

(t
f
)) . (8.15)
Furthermore, the maximum principle
H(t, x

(t), u

(t), p

(t)) = M(t, x

(t), p

(t)) (8.16)
holds for all t ∈ [t
0
, t
f
] except possibly for a ﬁnite set. [Here H(t, x, p, u) = p

f(t, x, u, ), M(t, x, p) =
sup¦H(t, x, v, p)[v ∈ Ω¦].
Proof: We break the proof up into a series of steps.
Step 1. By repeating the argument presented in the proof of Theorem 2 we can see that (8.15) is
equivalent to
p

(t
f
)

h ≤ 0 for all h ∈ C(t
f
) . (8.17)
Step 2. Deﬁne two convex sets S
1
, S
2
in R
1+m
as follows:
S
1
= ¦(y, h)[y > 0, h ∈ T
f
(x

(t
f
))¦,
S
2
= ¦(y, h)[y = ψ
x
(x

(t
f
))h, h ∈ ¦C(t
f
) + Φ(t
f
, t
0
)T
0
(x

0
)¦¦ .
We claim that the optimality of (u

(), x

0
) implies that S
1
∩ Relative Interior (S
2
) = φ. Suppose
this is not the case. Then there exists h ∈ T
f
(x

(t
f
)) such that
ψ
x
(x

(t
f
))h > 0 , (8.18)
102 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
h ∈ Interior¦C(t
f
) + Φ(t
f
, t
0
)T
0
(x

0
)¦ . (8.19)
Now by assumption g
f
x
(x

(t
f
) has maximum rank. Since g
f
x
(x

(t
f
))h = 0 it follows that the
Implicit Function Theorem that for ε > 0 sufﬁciently small there exists h(ε) ∈ R
n
such that
g
f
(x

(t
f
) +h(ε)) = b
f
, (8.20)
and, moreover, h(ε) → 0, (1/ε)h(ε) → h as ε → 0. From (8.18) and Lemma 3 it follows that for
ε > 0 sufﬁciently small there exists x
0
(ε) ∈ T
0
and u
ε
() ∈ | such that
x

(t
f
) +h(ε) = φ(t
f
, t
0
, x
0
(ε), u
ε
()) .
Hence we can conclude from (8.20) that the pair (x
0
(ε), u
ε
()) satisﬁes the initial and ﬁnal condi-
tions, and the corresponding value of the objective function is
ψ(x

(t
f
) +h(ε)) = ψ(x

(t
f
)) +ψ
x
(x

(t
f
))h(ε) +o([h(ε)[) ,
and since h(ε) = εh +o(ε) we get
ψ(x

(t
f
) +h(ε)) = ψ(x

(t
f
)) +ε)ψ
x
(x

(t
f
))h +o(ε) ;
but then from (8.18)
ψ(x

(t
f
) +h(ε)) > ψ(x

(t
f
))
for ε > 0 sufﬁciently small, thereby contradicting the optimality of (u

(), x

0
).
Step 3. By the separation theorem for convex sets there exist ˆ p
0
∈ R, ˆ p
1
∈ R
n
, not both zero, such
that
ˆ p
0
y
1
+ ˆ p

1
h
1
≥ ˆ p
0
y
2
+ ˆ p

1
h
2
for all (y
i
, h
i
) ∈ S
1
, i = 1, 2 . (8.21)
Arguing in exactly the same fashion as in the proof of Lemma 1 of 7.2 we can conclude that (8.21)
is equivalent to the following conditions:
ˆ p
0
≥ 0 ,
ˆ p
1
⊥T
f
(x

(t
f
)) ,
(8.22)
Φ(t
f
, t
0
)

(ˆ p
0
∇ψ(x

(t
f
)) + ˆ p
1
)⊥T
0
(x

0
) , (8.23)
and
(ˆ p
0
ψ
x
(x

(t
f
)) + ˆ p

1
)h ≤ 0 for all h ∈ C(t
f
) . (8.24)
If we let ˆ p

0
= ˆ p
0
and p

(t
f
) = ˆ p
0
∇ψ(x

(t
f
)) + ˆ p
1
then (8.22), (8.23), and (8.24) translate respec-
tively into (8.15), (8.14), and (8.17). ♦
8.2. INTEGRAL OBJECTIVE FUNCTION 103
8.2 Integral Objective Function
In many control problems the objective function is not given as a function ψ(x(t
f
)) of the ﬁnal
state, but rather as an integral of the form

t
f
t
0
f
0
(t, x(t), u(t))dt . (8.25)
The dynamics of the state, the boundary conditions, and control constraints are the same as before.
We proceed to show how such objective functions can be treated as a special case of the problems
of the last section. To this end we deﬁned the augmented system with state variable ˜ x = (x
0
, x) ∈
R
1+m
as follows:
·
˜ x=
¸
˙ x
0
(t)
˙ x(t)

=
˜
f(t, ˜ x(t), u(t)) =
¸
f
0
(t, x(t), u(t))
f(t, x(t), u(t))

.
The initial and ﬁnal conditions which are of the form
g
0
(x) = b
0
, g
f
(x) = b
f
are augmented ˜ g
0
(˜ x) =
¸
x
0
g
0
(x)

=
˜
b
0
=
¸
0
b
0

and ˜ g
f
(˜ x) = g
f
(x) = b
f
. Evidently then the problem of maximizing (8.25) is equivalent to the
problem of maximizing
ψ(˜ x(t
f
)) = x
0
(t
f
) ,
subject to the augmented dynamics and constraints which is of the form treated in Theorem 3 of
Section 1, and we get the following result.
Theorem 1: Consider the optimal control problem (8.26):
Maximize

t
f
t
0
f
0
(t, x(t), u(t))dt
subject to
dynamics: ˙ x(t) = f(t, x(t), u(t)), t
0
≤ t ≤ t
f
,
initial conditions: g
0
(x(t
0
)) = b
0
,
ﬁnal conditions: g
f
(x(t
f
)) = b
f
,
control constraint: u() ∈ | .
(8.26)
Let u

() ∈ |, let x

0
∈ T
o
and let x

(t) = φ(t, t
0
, x

0
, u

()), and suppose that x

(t
f
) ∈ T
f
. If
(u

(), x

0
) is optimal, then there exists a function ˜ p

= (p

0
, p

) : [t
0
, t
f
] → R
1+m
, not identically
zero, and with p

0
(t) ≡ constant and p

0
(t) ≥ 0, satisfying
·
˜ p

(t) = −[

˜
f
∂˜ x
(t, x

(t), u

(t))]

˜ p

(t) ,
initial condition: p

(t
0
)⊥T
0
(x

0
) ,
ﬁnal condition: p

(t
f
)⊥T
f
(x

(t
f
)) .
Futhermore, the maximum principle
˜
H(t, x

(t), ˜ p

(t), u

(t)) =
˜
M(t, x

(t), ˜ p

(t))
holds for all t ∈ [t
0
, t
f
] except possibly for a ﬁnite set. [Here
˜
H(t, x, ˜ p, u) = ˜ p

˜
f(t, x, u) =
p
0
f
0
(t, x, u) +p

f(t, x, u), and
˜
M(t, x, ˜ p) = sup¦
˜
H(t, x, ˜ p, v)[v ∈ Ω¦.]
Finally, if f
0
and f do not explicitly depend on t, then
˜
M(t, x

(t), ˜ p

(t)) ≡ constant.
Exercise 1: Prove Theorem 1. (Hint: For the ﬁnal part show that (d/dt)
˜
M(t, x

(t), ˜ p

(t)) ≡ 0.)
104 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
8.3 Variable Final Time
8.3.1 Main result.
In the problem considered up to now the ﬁnal time t
f
is assumed to be ﬁxed. In many important
cases the ﬁnal time is itself a decision variable. One such case is the minimum-time problem where
we want to transfer the state of the system from a given initial state to a speciﬁed ﬁnal state in
minimum time. More generally, consider the optimal control problem (8.27).
Maximize

t
f
t
0
f
0
(t, x(t), u(t))dt
subject to
dynamics: ˙ x(t) = f(t, x, (t), u(t)), , t
0
≤ t ≤ t
f
,
initial condition: g
0
(x(t
0
)) = b
0
,
ﬁnal condition: g
f
(x(t)f)) = b
f
,
control constraint: u() ∈ | ,
ﬁnal-time constraint: t
f
∈ (t
0
, ∞) .
(8.27)
We analyze (8.27) by converting the variable time interval [t
0
, t
f
] into a ﬁxed-time interval [0, 1].
This change of time-scale is achieved by regarding t as a new state variable and selecting a new
time variable s which ranges over [0, 1]. The equation for t is
dt(s)
ds
= α(s) , 0 ≤ s ≤ 1 ,
with initial condition
t(0) = t
0
.
Here α(s) is a new control variable constrained by α(s) ∈ (0, ∞). Now if x() is the solution of
˙ x(t) = f(t, x(t), u(t)) , t
0
≤ t ≤ t
f
, x(t
0
) = x
0
(8.28)
and if we deﬁne
z(s) = x(t(s)), v(s) = u(t(s)) , 0 ≤ s ≤ 1 ,
then it is easy to see that z() is the solution of
dz
ds
(s) = α(s)f(s, z(s), v(s)) , 0 ≤ s ≤ 1 z(0) = x
0
. (8.29)
Conversely from the solution z() of (8.29) we can obtain the solution x() of (8.28) by
x(t) = z(s(t)) , t
0
≤ t ≤ t
f
,
where s() : [t
0
, t
f
] → [0, 1] is the functional inverse of s(t); in fact, s() is the solution of the
differential equation ˙ s(t) = 1/α(s(t)), s(t
0
) = 0.
8.3. VARIABLE FINAL TIME 105
With these ideas in mind it is natural to consider the ﬁxed-ﬁnal-time optimal control problem
(8.30), where the state vector (t, z) ∈ R
1+m
, and the control (α, v) ∈ R
1+p
:
Maximize

1
0
f
0
(t(s), z(s), v(s))α(s)ds
subject to
dynamics: ( ˙ z(s),
˙
t(s)) = (f(t(s), z(s), v(s))α(s), α(s)),
initial constraint: g
0
(z(0)) = b
0
, t(0) = t
0
,
ﬁnal constraint: g
f
(z(1)) = b
f
, t(1) ∈ R ,
control constraint: (v(s), α(s)) ∈ Ω (0, ∞)
for 0 ≤ s ≤ 1 and v(), α() piecewise continuous.
(8.30)
The relation between problems (8.27) and (8.30) is established in the following result.
Lemma 1: (i) Let x

0
∈ T
0
, u

() ∈ |, t

f
∈ (t
0
, ∞) and let x

(t) = φ(t, t
0
, x

0
, u

()) be the
corresponding trajectory. Suppose that x

(t

f
) ∈ T
f
, and suppose that (u

(), x

0
, t

f
) is optimal for
(8.27). Deﬁne z

0
, v

(), and α

() by
z

0
= x

0
v

(s) = u

(t
0
+s(t

f
−t
0
))
α

(s) = (t

f
−t
0
)
, 0 ≤ s ≤ 1 ,
, 0 ≤ s ≤ 1 .
Then ((v

(), α

()), z

0
) is optimal for (8.30).
(ii) Let z

0
∈ T
0
, and let (v

(), α

()) be an admissible control for (8.30) such that the correspond-
ing trajectory (t

(), z

()) satisﬁes the ﬁnal conditions of (8.30). Suppose that ((v

(), α

()), z

0
)
is optimal for (8.30). Deﬁne x

0
, u

() ∈ |, and t

f
by
x

0
= z

0
,
u

(t) = v

(s

(t)) , t
0
≤ t ≤ t

f
,
t

f
= t

(1) ,
where s

() is functional inverse of t

(). Then (u

(), z

0
, t

f
) is optimal for (8.27).
Exercise 1: Prove Lemma 1.
Theorem 1: Let u

() ∈ |, let x

0
∈ T
0
, let t

f
∈ (0, ∞), and let
x

(t) = φ(t, t
0
, x

0
, u

()), t
0
≤ t ≤ t
f
, and suppose that x

(t

f
) ∈ T
f
. If (u

(), x

0
, t

f
) is optimal
for (8.27), then there exists a function ˜ p

= (p

0
, p

) : [t
0
, t

f
] → R
1+m
, not identically zero, and
with p

0
(t) ≡ constant and p

0
(t) ≥ 0, satisfying
·
˜ p

(t) = −[

˜
f
∂˜ x
(t, x

(t), u

(t))]

˜ p

(t) ,
(8.31)
initial condition: p

(t
0
)⊥T
0
(x

0
) , (8.32)
106 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
ﬁnal condition: p

(t

f
)⊥T
f
(x

(t

f
)) . (8.33)
Also the maximum principle
˜
H(t, x

(t), ˜ p

(t), u

(t)) =
˜
M(t, x

(t), ˜ p

(t)) , (8.34)
holds for all t ∈ [t
0
, t
f
] except possibly for a ﬁnite set. Furthermore, t

f
must be such that
ˆ
H(t

f
, x

(t

f
), ˜ p

(t

f
), u

(t

f
)) = 0 . (8.35)
Finally, if f
0
and f do not explicitly depend on t, then
ˆ
M(t, x

(t), ˜ p

(t)) ≡ 0.
Proof: By Lemma 1, z

0
= x

0
, v

(s) = u

(t
0
+s(t

f
−t
0
)) and α

(s) = (t

f
−t
0
) for 0 ≤ s ≤ 1
constitute an optimal solution for (8.30). The resulting trajectory is
z

(s) = x

(t
0
+s(t

f
−t
0
)), t

(s) = t
0
+s(t

f
−t
0
), 0 ≤ s ≤ 1 , so that in particular
z

(1) = x

(t

f
).
By Theorem 1 of Section 2, there exists a function
˜
λ

= (λ

0
, λ

, λ

n+1
) : [0, 1] → R
1+n+1
, not
identically zero, and with λ

0
(s) ≡ constant and λ

0
(s) ≥ 0, satisfying

˙
λ

0
(t)
˙
λ

(t)
˙
λ

n+1
(t)
¸
¸
¸
¸
¸
¸
= −

0
¦[
∂f
0
∂z
(t

(s), z

(s), v

(s))]

λ

0
(s)
+[
∂f
∂z
(t

(s), z

(s), v

(s))]

λ

(s)¦α

(s)
¦[
∂f
0
∂t
(t

(s), z

(s), v

(s))]

λ

0
(s)
+[
∂f
∂t
(t

(s), z

(s), v

(s))]

λ

(s)¦α

(s)
¸
¸
¸
¸
¸
¸
¸
(8.36)
initial condition: λ

(0)⊥T
0
(z

0
) (8.37)
ﬁnal condition: λ

(1)⊥T
f
(z

(1)) , λ

n+1
(1) = 0 . (8.38)
Furthermore, the maximum principle
λ

0
(s)f
0
(t

(s), z

(s), v

(s))α

(s)

(s)

f(t

(s), z

(s), v

(s))α

(s) +λ

n+1
(s)α

(s)
= sup¦[λ

0
(s)f
0
(t

(s), z

(s), w)β

(s)

f(t

(s), z

(s), w)β +λ

n+1
(s)β][w ∈ Ω, β ∈ (0, ∞)¦
(8.39)
holds for all s ∈ [0, 1] except possibly for a ﬁnite set.
Let s

(t) = (t −t
0
)/(t

f
−t
0
), t
0
≤ t ≤ t

f
, and deﬁne ˜ p

= (p

0
, p

) : [t
0
, t

f
] → R
1+n
by
p

0
(t) = λ

0
(s

(t)), p

(t) = λ

(s

(t)), t
0
≤ t ≤ t

f
. (8.40)
First of all, ˜ p

is not identically zero. Because if ˜ p

≡ 0, then from (8.40) we have (λ

0
, λ

) ≡ 0 and
then from (8.36), λ

n+1
≡ constant, but from (8.38), λ

n+1
(1) = 0 so that we would have
˜
λ

≡ 0
8.3. VARIABLE FINAL TIME 107
which is a contradiction. It is trivial to verify that ˜ p

() satisﬁes (8.31), and, on the other hand (8.37)
and (8.38) respectively imply (8.32) and (8.33). Next, (8.39) is equivalent to
λ

0
(s)f
0
(t

(s), z

(s), v

(s))

(s)

f(t

(s), z

(s), v

(s)) +λ

n+1
(s) = 0
(8.41)
and
λ

0
(s)f
0
(t

(s), z

(s), v

(s)) +λ

(s)

f(t

(s), z

(s), v

(s))
= Sup ¦[λ

0
(s)f
0
(t

(s), z

(s), w) +λ

(s)

f(t

(s), z

(s), w)][w ∈ Ω¦.
(8.42)
Evidently (8.42) is equivalent to (8.34) and (8.35) follows from (8.41) and the fact that λ

n+1
(1) = 0.
Finally, the last assertion of the Theorem follows from (8.35) and the fact that
˜
M(t, x

(t), ˜ p

(t)) ≡
constant if f
0
, f are not explicitly dependent on t. ♦
8.3.2 Minimum-time problems
.
We consider the following special case of (8.27):
Maximize

t
f
t
0
(−1)dt
subject to
dynamics: ˙ x(t) = f(t, x(t), u(t)), t
0
≤ t ≤ t
f
initial condition: x(t
0
) = x
0
,
ﬁnal condition: x(t
f
) = x
f
,
control constraint: u() ∈ | ,
ﬁnal-time constraint: t
f
∈ (t
0
, ∞) .
(8.43)
In (8.43), x
0
, x
f
are ﬁxed, so that the optimal control problem consists of ﬁnding a control which
transfers the system from state x
0
at time t
0
to state x
f
in minimum time. Applying Theorem 1 to
this problem gives Theorem 2.
Theorem 2: Let t

f
∈ (t
0
, ∞) and let u

: [t
0
, t

f
] → Ω be optimal. Let x

() be the corresponding
trajectory. Then there exists a function p

: [t
0
, t

f
] → R
n
, not identically zero, satisfying

(t) = −[
∂f
∂x
(t, x

(t), u

(t))]

p

(t), t
0
≤ t ≤ t

f
,
initial condition: p

(t
0
) ∈ R
n
,
ﬁnal condition: p

(t

f
) ∈ R
n
.
Also the maximum principle
H(t, x

(t), p

(t), u

(t)) = M(t, x

(t), p

(t)) (8.44)
holds for all t ∈ [t
0
, t

f
] except possibly for a ﬁnite set.
Finally,
M(t

f
, x

(t
f
), p

(t
f
)) ≥ 0 (8.45)
and if f does not depend explicitly on t then
M(t, x

(t), p

, (t)) ≡ constant . (8.46)
108 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
Exercise 2: Prove Theorem 2.
We now study a simple example illustrating Theorem 2. Example 1: The motion of a particle is
described by
m¨ x(t) +σ ˙ x(t) = u(t) ,
where m = mass, σ = coefﬁcient of friction, u = applied force, and x = position of the particle. For
simplicity we suppose that x ∈ R, u ∈ R and u(t) constrained by [u(t)[ ≤ 1. Starting with an
initial condition x(0) = x
01
, ˙ x(0) = x
02
we wish to ﬁnd an admissible control which brings the
particle to the state x = 0, ˙ x = 0 in minimum time.
Solution: Taking x
1
= x, x
2
= ˙ x we rewrite the particle dynamics as
¸
˙ x
1
(t)
˙ x
2
(t)

=
¸
0 1
0 −α
¸
x
1
(t)
x
2
(t)

+
¸
0
b

u(t) , (8.47)
where α = (σ/m) > 0 and b = (1/m) > 0. The control constraint set is Ω = [−1, 1].
Suppose that u

() is optimal and x

() is the corresponding trajectory. By Theorem 2 there exists
a non-zero solution p

() of
¸
˙ p

1
(t)
˙ p

2
(t)

= −
¸
0 0
1 −α
¸
p

1
(t)
p

2
(t)

(8.48)
such that (8.44), (8.45), and (8.46) hold. Now the transition matrix function of the homogeneous
part of (8.47) is
Φ(t, τ) =
¸
1
1
α
(1 −e
−α(t−τ)
)
0 e
−α(t−τ)

,
so that the solution of (8.48) is
¸
p

1
(t)
p

2
(t)

=
¸
1 0
1
α
(1 −e
αt
) e
αt
¸
p

1
(0)
p

2
(0)

,
or
p

1
(t) ≡ p

1
(0) ,
and
p

2
(t) =
1
α
p

1
(0) +e
αt
(−
1
α
p

1
(0) +p

2
(0)) . (8.49)
The Hamiltonian H is given by
H(x

(t), p

(t), v) = (p

1
(t) −αp

2
(t))x

2
(t) +bp

2
(t)v
= e
αt
(p

1
(0) −αp

2
(0))x

2
(t) +pb

2
(t)v ,
8.3. VARIABLE FINAL TIME 109
so that from the maximum principle we can immediately conclude that
u

(t) =

+1 if p

2
(t) > 0,
−1 if p

2
(t) < 0,
? if p

2
(t) = 0 .
(8.50)
Furthermore, since the right-hand side of (8.47) does not depend on t explicitly we must also have
e
αt
(p

1
(0) −αp

2
(0))x

2
(t) +bp

2
(t)u

(t) ≡ constant. (8.51)
We now proceed to analyze the consequences of (8.49) and (8.50). First of all since p

1
(t) ≡
p

1
(0), p

2
() can have three qualitatively different forms.
Case 1. −p

1
(0) + αp

2
(0) > 0: Evidently then, from (8.49) we see that p

2
(t) must be a strictly
monotonically increasing function so that from (8.50) u

() can behave in one of two ways:
either
u

(t) =

−1 for t <
ˆ
t and p

2
(t) < 0 for t <
ˆ
t,
+1 for t >
ˆ
t and p

2
(t) > 0 for t >
ˆ
t,
or
u

(t) ≡ +1 and p

2
(t) > 0 for all t.
Case 2. −p

1
(0) +αp

2
(0) < 0 : Evidently u

() can behave in one of two ways:
either
u

(t) =

+1 for t <
ˆ
t and p

2
(t) > 0 for t <
ˆ
t,
−1 for t >
ˆ
t and p

2
(t) < 0 for t >
ˆ
t,
or
u

(t) ≡ −1 and p

(t) < 0 for all t.
Case 3. −p

1
(0) + αp

2
(0) = 0 : In this case p

2
(t) ≡ (1/α)p

1
(0). Also since p

(t) ≡ 0, we must
have in this case p

1
(0) = 0. Hence u

() we can behave in one of two ways:
either
u

(t) ≡ +1 and p

2
(t) ≡
1
α
p

1
(0) > 0 ,
or
u

(t) ≡ −1 and p

2
(t) ≡
1
α
p

1
(0) < 0 ,
110 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
Thus, the optimal control u

is always equal to +1 or -1 and it can switch at most once between
these two values. The optimal control is given by
u

(t) = sgn p

2
(t)
= sgn [
1
α
p

1
(0) +e
αt
(−
1
α
p

1
(0) +p

2
(0))] .
Thus the search for the optimal control reduces to ﬁnding p

1
(0), p

2
(0) such that the solution of the
differential equation
˙ x = x
2
˙ x
2
= −αx
2
+b sgn[
1
α
p

1
(0) +e
αt
(−
1
α
p

1
(0) +p

2
(0))] ,
(8.52)
with initial condition
x
1
(0) = x
10
, x
20
= x
20
(8.53)
also satisﬁes the ﬁnal condition
x
1
(t

f
) = 0, x
2
(t

f
) = 0 , (8.54)
for some t

f
> 0; and then t

f
is the minimum time.
There are at least two ways of solving the two-point boundary value problem (8.52), (8.53), and
(8.54). One way is to guess at the value of p

(0) and then integrate (8.52) and (8.53) forward in time
and check if (8.54) is satisﬁed. If (8.54) is not satisﬁed then modify p

(0) and repeat. An alternative
is to guess at the value of p

(0) and then integrate (8.52) and (8.54) backward in time and check of
(8.53) is satisﬁed. The latter approach is more advantageous because we know that any trajectory
obtained by this procedure is optimal for initial conditions which lie on the trajectory. Let us follow
this procedure.
Suppose we choose p

(0) such that −p

1
(0) = αp

2
(0) = 0 and p

2
(0) > 0. Then we must have
u

(t) ≡ 1. Integrating (8.52) and (8.54) backward in time give us a trajectory ξ(t) where
˙
ξ
1
(t) = −
˙
ξ
2
(t)
˙
ξ
2
(t) = αξ
2
(t) −b ,
with
ξ
1
(0) −ξ
2
(0) = 0 .
This gives
ξ
1
(t) =
b
α
(−t +
e
αt
−1
α
) , ξ
2
(t) =
b
α
(1 −e
αt
) ,
which is the curve OA in Figure 8.3.
On the other hand, if p

(0) is such that −p

1
(0) + αp

2
(0) = 0 and p

2
(0) < 0, then u

(t) ≡ −1
and we get
ξ
1
(t) = −
b
α
(−t +
e
αt
−1
α
) , ξ
2
(t) = −
b
α
(1 −e
αt
) ,
which is the curve OB.
8.3. VARIABLE FINAL TIME 111
B
u

≡ −1
D
u

≡ 1
C
ξ
1
O
ξ
2
u

≡ 1
A
E
u

≡ −1
F
Figure 8.3: Backward integration of (8.52) and (8.54).
Next suppose p

(0) is such that −p

1
(0) + αp

2
(0) > 0, and p

2
(0) < 0. Then [(1/α)p

1
(0) +
e
αt
(−(1/α)p

1
(0) + p

2
(0))] will have a negative value for t ∈ (0,
ˆ
t) and a positive value for t ∈
(
ˆ
t, ∞). Hence, if we integrate (8.52), (8.54) backwards in time we get trajectory ξ(t) where
˙
ξ(t) = −ξ
2
(t)
˙
ξ
2
(t) = αξ
2
(t)+

−b for t <
ˆ
t
b for t >
ˆ
t ,
with ξ
1
(0) = 0, ξ
2
(0) = 0. This give us the curve OCD. Finally if p

(0) is such that −p

1
(0) +
αp

2
(0) < 0, and p

2
(0) < 0, then u

(t) = 1 for t <
ˆ
t and u

(t) = −1 for t >
ˆ
t, and we get the
curve OEF.
We see then that the optimal control u

() has the following characterizing properties:
u

(t) =

1 if x

(t) is above BOA or on OA
−1 if x

(t) is below BOA or on OB .
Hence we can synthesize the optimal control in feedback from: u

(t) = ψ(x

(t)) where the
B
u

≡ −1
x
2
u

≡ 1
x
1
A
u

≡ 1
O
u

≡ −1
Figure 8.4: Optimal trajectories of Example 1.
112 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
function ψ : R
2
→ ¦1, −1¦ is given by (see Figure 8.4)
ψ(x
1
, x
2
) =

1 if (x
1
, x
2
) is above BOA or on OA
−1 if (x
1
, x
2
) is below BOA or on OB .
An important class of problems which arise in practice is the case when the dynamics are linear and
the objective function is quadratic. Speciﬁcally, consider the optimal control problem (8.55):
Minimize

T
0
1
2
[x

(t)P(t)x(t) +u

(t)Q(t)u(t)]dt
subject to
dynamics: ˙ x(t) = A(t)x(t) +B(t)u(t), 0 ≤ t ≤ T ,
initial condition: x(0) = x
0
,
ﬁnal condition: G
f
x(t) = b
f
,
control constraint: u(t) ∈ R
p
, u() piecewise continuous.
(8.55)
In (8.56) we assume that P(t) is an n n symmetric, positive semi-deﬁnite matrix whereas Q(t) is
a p p symmetric, positive deﬁnite matrix. G
f
is a given
f
n matrix, and x
0
∈ R
n
, b
f
∈ R

f
are given vectors. T is a ﬁxed ﬁnal time.
We apply Theorem 1 of Section 2, so that we must search for a number p

0
≥ 0 and a function
p

: [0, T] → R
n
, not both zero, such that
˙ p

(t) = −p

0
(−P(t)x

(t)) −A

(t)p

(t) , (8.56)
and
p

(t)⊥T
f
(x

(t)) = ¦ξ[G
f
ξ = 0¦ . (8.57)
The Hamiltonian function is
H(t, x

(t), ˜ p

(t), v) = −
1
2
p

0
[x

(t)

P(t)x

(t) +v

Q(t)v]
+p

(t)

[A(t)x

(t) +B(t)v]
so that the optimal control u

(t) must maximize

1
2
p

0
v

Q(t)v +p

(t)

B(t)v for v ∈ R
p
. (8.58)
If p

0
> 0, this will imply
u

(t) =
1
p

0
Q
−1
(t)B

(t)p

(t) ,
(8.59)
whereas if p

0
= 0, then we must have
p

(t)

B(t) ≡ 0 (8.60)
because otherwise (8.58) cannot have a maximum.
8.5. THE SINGULAR CASE 113
We make the following assumption about the system dynamics.
Assumption: The control system ˙ x(t) = A(t)x(t) + B(t)u(t) is controllable over the interval
[0, T]. (See (Desoer [1970]) for a deﬁnition of controllability and for the properties we use below.)
Let Φ(t, τ) be the transition matrix function of the homogeneous linear differential equation ˙ x(t) =
A(t)x(t). Then the controllability assumption is equivalent to the statement that for any ξ ∈ R
n
ξ

Φ(t, τ)B(τ) = 0 , 0 ≤ τ ≤ T , implies ξ = 0 . (8.61)
Next we claim that if the system is controllable then p

0
= 0, because if p

0
= 0 then from (8.56)
we can see that
p

(t) = (Φ(T, t))

p

(T)
and hence from (8.60)
(p

(t))

Φ(T, t)B(t) = 0 , 0 ≤ t ≤ T ,
but then from (8.61) we get p

(T) = 0. Hence if p

0
= 0, then we must have ˜ p

(t) ≡ 0 which is a
contradiction. Thus, under the controllability assumption, p

0
> 0, and hence the optimal control is
given by (8.59). Now if p

0
> 0 it is trivial that ˆ p

(t) = (1, (p

(t)/p

0
)) will satisfy all the necessary
conditions so that we can assume that p

0
= 1. The optimal trajectory and the optimal control is
obtained by solving the following two-point boundary value problem:
˙ x

(t) = A(t)x

(t) +B(t)Q
−1
(t)B

(t)p

(t)
˙ p(t) = P(t)x

(t) −A

(t)p

(t)
x

(0) = x
0
, G
f
x

(T) = b
f
, p

(T)⊥T
f
(x

(T)) .
For further details regarding the solution of this boundary value problem and for related topics see
(See and Markus [1967]).
8.5 The Singular Case
In applying the necessary conditions derived in this chapter it sometimes happens that H(t, x

(t), p

(t), v)
is independent of v for values of t lying in a non-zero interval. In such cases the maximum principle
does not help in selecting the optimal value of the control. We are faced with the so-called singular
case (because we are in trouble–not because the situation is rare). We illustrate this by analyzing
Example 4 of Chapter 1.
The problem can be summarized as follows:
Maximize

T
0
c(t)dt =

T
0
(1 −s(t))f(k(t))dt
subject to
dynamics:
˙
k(t) = s(t)f(k(t)) −µk(t) , 0 ≤ t ≤ T
initial constraint: k(0) = k
0
,
ﬁnal constraint: k(t) ∈ R ,
control constraint: s(t) ∈ [0, 1], s() piecewise continuous.
114 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
We make the following assumptions regarding the production function f:
f
k
(k) > 0, f
kk
(K) < 0 for all k , (8.62)
lim
k→0
f
k
(k) = ∞ .
(8.63)
Assumption (8.62) says that the marginal product of capital is positive and this marginal product
decreases with increasing capital. Assumption (8.63) is mainly for technical convenience and can
be dispensed with without difﬁculty.
Now suppose that s

: [0, T] → [0, 1] is an optimal savings policy and let k

(t), 0 ≤ t ≤ T,
be the corresponding trajectory of the capital-to-labor ratio. Then by Theorem 1 of Section 2, there
exist a number p

0
≥ 0, and a function p

: [0, T] → R, not both identically zero, such that
˙ p

(t) = −p

0
(1 −s

(t))f
k
(k

(t)) −p

(t)[s

(t)f
k
(k

(t)) −µ] (8.64)
with the ﬁnal condition
p

(T) = 0 , (8.65)
and the maximum principle holds. First of all, if p

0
= 0 then from (8.64) and (8.65) we must also
have p

(t) ≡ 0. Hence we must have p

0
> 0 and then by replacing (p

0
, p

) by (1/p

0
)(p

0
, p

) we
can assume without losing generality that p

0
= 1, so that (8.64) simpliﬁes to
˙ p

(t) = −1(1 −s

(t))f
k
(k

(t)) −p

(t)[s

(t)f
k
(k

(t)) −µ] . (8.66)
The maximum principle says that
H(t, k

(t), p

(t), s) = (1 −s)f(k

(t)) +p

(t)[sf(k

(t)) −µk

(t)]
is maximized over s ∈ [0, 1] at s

(t), which immediately implies that
s

(t) =

1 if p

(t) > 1
0 if p

(t) < 1
? if p

(t) = 1
(8.67)
We analyze separately the three cases above.
Case 1. p

(t) > 1, s

(t) = 1 : Then the dynamic equations become
˙
k

(t) = f(k

(t)) −µk

(t) ,
˙ p

(t) = −p

(t)[f
k
(k

(t)) −µ] .
(8.68)
The behavior of the solutions of (8.68) is depicted in the (k, p)−, (k, t)− and (p, t)−planes in
Figure 8.5. Here k
G
, k
H
are the solutions of f
k
(k
G
) −µ = 0 and f(k
M
) −µk = 0. Such solutions
exist and are unique by virtue of the assumptions (8.62) and (8.63). Futhermore, we note from
(8.62) that k
G
< k
M
, and f
k
(k) − µ
<
> 0 according as k
<
> k
G
whereas f(k) − µk
>
< 0 according
as k
<
> k
M
. (See Figure 8.6.)
8.5. THE SINGULAR CASE 115
p
f
k
> µ
f
k
< µ
f < µk f > µk
k
k
M
k
G
l
k
k
M
t
p
l
t
Figure 8.5: Illustration for Case 1.
Case 2. p

(t) < 1, s

(t) = 0: Then the dynamic equations are
˙
k

(t) = −µk

(t) ,
˙ p

(t) = −f
k
(k

(t)) +µp

(t) ,
giving rise to the behavior illustrated in Figure 8.7.
Case 3. p

(t) = 1, s

(t) =?: (Possibly singular case.) Evidently if p

(t) = 1 only for a ﬁnite set of
times t then we do not have to worry about this case. We face the singular case only if p

(t) = 1
for t ∈ I, where I is a non-zero interval. But then we have ˙ p

(t) = 0 for t ∈ I so that from (8.66)
we get
−(1 −s

(t))f
k
(k

(t)) −[s

(t)f
k
(k

(t)) −µ] = 0 for t ∈ I ,
so
−f
k
(k

(t)) +µ = 0 for t ∈ I ,
or
k

(t) = k
G
for t ∈ I . (8.69)
In turn then we must have
˙
k

(t) = 0 for t ∈ I so that
s

(t)f(k
G
) −µK
G
= 0 for t ∈ I ,
and hence,
s

(t) = µ
k
G
f(k
G
)
for t ∈ I . (8.70)
116 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
.
f
line of slope µ
µk
f(k)
k
k
M
k
G
Figure 8.6: Illustration for assumptions (8.62), (8.63).
Thus in the singular case the optimal solution is characterized by (8.69) and (8.70), as in Figure 8.8.
We can now assemble separate cases to obtain the optimal control. First of all, from the ﬁnal
condition (8.65) we know that for t close to T, p

(t) < 1 so that we are in Case 2. We face two
possibilities: Either (A)
p

(t) < 1 for all t < [0, T]
and then s

(t) = 0, k

(t) = k
0
e
−µt
, for 0 ≤ t ≤ T, or (B)
there exists t
2
∈ (0, T) such that p

(t
2
) = 1 and p

(t) < 1 for t
2
< t ≤ T .
We then have three possibilities depending on the value of k

(t
2
):
(Bi) k

(t
2
) < k
G
: then ˙ p

(t
2
) < 0 so that p

(t) > 1 for t < t
2
and we are in Case 1 so that
s

(t) = 1 for t < t
2
. In particular we must have k
0
< k
G
.
(Bii) k

(t
2
) > k
G
: then ˙ p

(
2
) > 0 but then p

(t
2
+ ε) > 1 for ε > 0 sufﬁciently small and since
p

(T) = 0 there must exist t
3
∈ (t
2
, T) such that p

(t
3
) = 1. This contradicts the deﬁnition of t
2
so that this possibility cannot arise.
(Biii) k

(t
2
) − k
G
: then we can have a singular arc in some interval (t
1
, t
2
) so that p

(t) =
1, k

(t) = k
G
, and s

(t) = µ(k
G
/f(k
G
)) for t ∈ (t
1
, t
2
). For t < t
1
we either have p

(t) >
1, s

(t) > 1 if k
0
< k
G
, or we have p

(t) < 1, s

(t) = 0 if k > k
G
.
The various possibilities are illustrated in Figure 8.9.
The capital-to-labor ratio k
G
is called the golden mean and the singular solution is called the
golden path. The reason for this term is contained in the following exercise.
Exercise 1: A capital-to-labor ratio
ˆ
k is said to be sustainable if there exists ˆ s ∈ [0, 1] such that
ˆ sf(
ˆ
k) −µ
ˆ
k = 0. Show that k
G
is the unique sustainable capital-to-labor ratio which maximizes
sustainable consumption (1 −s)f(k).
8.6. BIBLIOGRAPHICAL REMARKS 117
p
k
k
G
l
k
t
p
t
l
Figure 8.7: Illustration for Case 2.
8.6 Bibliographical Remarks
The results presented in this chapter appeared in English in full detail for the ﬁrst time in 1962 in the
book by Pontryagin, et al., cited earlier. That book contains many extensions and many examples
and it is still an important source. However, the derivation of the maximum principle given in the
book by Lee and Markus is more satisfactory. Several important generalizations of the maximum
principle have appeared. On the one hand these include extensions to inﬁnite-dimensional state
spaces and on the other hand they allow for constraints on the state more general than merely initial
and ﬁnal constraints. For a uniﬁed, but mathematically difﬁcult, treatment see (Neustadt [1969]).
For a less rigorous treatment of state-space constraints see (Jacobson, et al, [1971]), whereas for a
discussion of the singular case consult (Kelley, et al. [1968]).
For an applications-oriented treatment of this subject the reader is referred to (Athans and Falb
[1966]) and (Bryson and Ho [1969]). For applications of the maximum principle to optimal eco-
nomic growth see (Shell [1967]). There is no single source of computational methods for optimal
control problems. Among the many useful techniques which have been proposed see (Lasdon, et
al., [1967]), (Kelley [1962]), (McReynolds [1966]), and (Balakrishnan and Neustadt [1964]); also
consult (Jacobson and Mayne [1970]), and (Polak [1971]).
118 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
.
p
k
k
G
1
k
t
k
G
p
t
1
Figure 8.8: Case 3. The singular case.
8.6. BIBLIOGRAPHICAL REMARKS 119
. .
.
.
.
.
. .
p

1
t
T
s

1
t
T
k

t
T
p

Case (A)
t
T t
2
t
1
s

1
µk
G
f(k
G
)
t
k

k
G
k
0
t
p

1
t
T
t
2
s

1
t
T
t
2
k

t
T
t
2
p

Case (Bi)
t
T
t
2
t
1
s

t
k

t
Case (Biii)
Figure 8.9: The optimal solution of example.
120 CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
Chapter 9
Dynamic programing
SEQUENTIAL DECISION PROBLEMS: DYNAMIC PROGRAMMING FORMULATION
The sequential decision problems discussed in the last three Chapters were analyzed by varia-
tional methods, i.e., the necessary conditions for optimality were obtained by comparing the op-
timal decision with decisions in a small neighborhood of the optimum. Dynamic programming
(DP is a technique which compares the optimal decision with all the other decisions. This global
comparison, therefore, leads to optimality conditions which are sufﬁcient. The main advantage of
DP, besides the fact that it give sufﬁciency conditions, is that DP permits very general problem for-
mulations which do not require differentiability or convexity conditions or even the restriction to a
ﬁnite-dimensional state space. The only disadvantage (which unfortunately often rules out its use)
of DP is that it can easily give rise to enormous computational requirements.
In the ﬁrst section we develop the main recursion equation of DP for discrete-time problems. The
second section deals with the continuous-time problem. Some general remarks and bibliographical
references are collected in the ﬁnal section.
9.1 Discrete-time DP
We consider a problem formulation similar to that of Chapter VI. However, for notational conve-
nience we neglect ﬁnal conditions and state-space constraints.
Maximize
N−1
¸
i=0
f
0
(i, x(i), u(i)) + Φ(x(N))
subject to
dynamics: x(i + 1) = f(i, x(i), u(i)) , i = 0, 1, . . . , N −1 ,
initial condition: x(0) = x
0
,
control constraint: u(i) ∈ Ω
i
, i = 0, 1, . . . , N −1 .
(9.1)
In (9.1), the state x(i) and the control u(i) belong to arbitrary sets X and U respectively. X and U
may be ﬁnite sets, or ﬁnite-dimensional vector spaces (as in the previous chapters), or even inﬁnite-
dimensional spaces. x
0
∈ X is ﬁxed. The Ω
i
are ﬁxed subsets of U. Finally f
0
(i, , ) : X U →
R, Φ : X → R, f(i, , ) : X U → X are ﬁxed functions.
121
122 CHAPTER 9. DYNAMIC PROGRAMING
The main idea underlying DP involves embedding the optimal control problem (9.1), in which
the system starts in state x
0
at time 0, into a family of optimal control problems with the same
dynamics, objective function, and control constraint as in (9.1) but with different initial states and
initial times. More precisely, for each x ∈ X and k between ) and N − 1, consider the following
problem:
Maximize
N−1
¸
i=k
f
0
(i, x(i), u(i)) + Φ(x(N)) ,
subject to
dynamics: x(i + 1) = f(i, x(i), u(i)), i = k, k + 1, . . . , N −1,
initial condition: x(k) = x,
control constraint: u(i) ∈ Ω
i
, i = k, k + 1, , N −1 .
(9.2)
Since the initial time k and initial state x are the only parameters in the problem above, we will
sometimes use the index (9.2)
k,x
to distinguish between different problems. We begin with an
elementary but crucial observation.
Lemma 1: Suppose u

(k), . . . , u

(N − 1) is an optimal control for (9.2)
k,x
, and let x

(k) =
x, x

(k + 1), . . . , x

(N) be the corresponding optimal trajectory. Then for any , k ≤ ≤ N −
1, u

(), . . . , u

(N −1) is an optimal control for (9.2)
,x

()
.
Proof: Suppose not. Then there exists a control ˆ u(), ˆ u( + 1), . . . , ˆ u(N − 1), with corresponding
trajectory ˆ x() = x

(), ˆ x( + 1), . . . , ˆ x(N), such that
N−1
¸
i=
f
0
(i, ˆ x(i), ˆ u(i)) + Φ(ˆ x(N))
>
N−1
¸
i=
f
0
(i, x

(i), u

(i)) + Φ(x

(N)) .
(9.3)
But then consider the control ˜ u(k), . . . , ˜ u(N −1) with
˜ u(i)

u

(i) , i = k, . . . , −1
ˆ u(i) , i = , . . . , N −1 ,
and the corresponding trajectory, starting in state x at time k, is ˜ x(k), . . . , ˜ x(N) where
˜ x(i) =

x

(i) , i = k, . . . ,
ˆ x(i) , i = + 1, . . . , N .
The value of the objective function corresponding to this control for the problem (9.2)
k,x
is
N−1
¸
i=k
f
0
(i, ˜ x(i), ˜ u(i)) + Φ(˜ x(n))
=
−1
¸
i=k
f
0
(i, x

(i), u

(i)) +
N−1
¸
i=
f
0
(i, ˆ x(i), ˆ u(i)) + Φ(ˆ x(N))
>
N−1
¸
i=k
f
0
(i, x

(i), u

(i)) + Φ(x

(N)) ,
9.1. DISCRETE-TIME DP 123
by (9.3), so that u

(k), . . . , u

(N −1) cannot be optimal for 9.2)
k
(end theorem)
From now on we assume that an optimal solution to (9.2)
k,x
exists for all 0 ≤ k ≤ N −1, and all
x ∈ X. Let V (k, x) be the maximum value of (9.2)
k,x
. We call V the (maximum) value function.
Theorem 1: Deﬁne V (N, ) by (V (N, x) = Φ(x). V (k, x) satisﬁes the backward recursion equa-
tion
V (k, x) = Max¦f
0
, (k, x, u) +V (k
1
, f(k, x, u, ))[u ∈ Ω
k
¦, 0 ≤ k ≤ N −1 . (9.4)
Proof: Let x ∈ X, let u

(k), . . . , u

(N − 1) be an optimal control for (9.2)
k,x
, and let x

(k) =
x, . . . , x

(N) be the corresponding trajectory be x(k) = x, . . . , x(N). We have
N−1
¸
i=k
f
0
(i, x

(i), u

(i)) + Φ(x

(N))

N−1
¸
i=k
f
0
(i, x(i), u(i)) + Φ(x(N)) .
(9.5)
By Lemma 1 the left-hand side of (9.5) is equal to
f
0
(k, x, u

(k)) +V (k + 1, f(k, x

, u

(k)) .
On the other hand, by the deﬁnition of V we have
N−1
¸
i=k
f
0
(i, x(i), u(i)) + Φ(x(N)) = f
0
(k, x, u(k))

N
¸
i=k+1
f
0
(i, x(i), u(i)) + Φ(x(N)) ≤ f
0
(k, x, u, (k)) +V (k + 1, f(k, x, u(k))¦ ,
with equality if and only if u(k +1), . . . , u(N −1) is optimal for (9.2)
k+1,x(k+1)
. Combining these
two facts we get
f
0
(k, xu

(k)) +V (k + 1, f(k, x, u

(k)))
≥ f
0
(k, x, u(k)) +V (k + 1, f(x, k, u(k))) ,
for all u(k) ∈ Ω
k
, which is equivalent to (9.4).(end theorem)
Corollary 1: Let u(k), . . . , u(N − 1) be any control for the problem (9.2)
k,x
and let x(k) =
x, . . . , x(N) be the corresponding trajectory. Then
V (, x()) ≤ f
0
(, x(), u()) +V ( + 1, f(, x(), u()), k ≤ ≤ N −1 ,
and equality holds for all k ≤ ≤ N −1 if and only if the control is optimal for (9.2)
k,x
.
Corollary 2: For k = 0, 1, . . . , N −1, let ψ(k, ) : X → Ω
k
be such that
f
0
(k, x, ψ(k, x)) +V (k + 1, f(k, x, ψ(k, x))
= Max¦f
0
(k, x, u) +V (k + 1, f(k, x, u))[u ∈ Ω
k
¦ .
Then ψ(k, ), k = 0, . . . , N − 1 is an optimal feedback control, i.e., for any k, x the control
u

(k), . . . , u

(N −1) deﬁned by u

() = ψ(, x

()), k ≤ ≤ N −1, where
124 CHAPTER 9. DYNAMIC PROGRAMING
x

( + 1) = f(, x

(), ψ(, x

()), k ≤ ≤ N −1 , x

(k) = x ,
is optimal for (α)
k,x
.
Remark: Theorem 1 and Corollary 2 are the main results of DP. The recursion equation (9.4) al-
lows us to compute the value function, and in evaluating the maximum in (9.4) we also obtain the
optimum feedback control. Note that this feedback control is optimum for all initial conditions.
However, unless we can ﬁnd a “closed-form” analytic solution to (9.4), the DP formulation may
necessitate a prohibitive amount of computation since we would have to compute and store the val-
ues of V and ψ for all k and x. For instance, suppose n = 10 and the state-space X is a ﬁnite set
with 20 elements. Then we have to compute and store 10 20 values of V , which is a reasonable
amount. But now suppose X = R
n
and we approximate each dimension of x by 20 values. Then
for N = 10, we have to compute and store 10x(20)
n
values of V . For n = 3 this number is 80,000,
and for n = 5 it is 32,000,000, which is quite impractical for existing computers. This “curse of
dimensionality” seriously limits the applicability of DP to problems where we cannot solve (9.4)
analytically.
• Exercise 1: An instructor is preparing to lead his class for a long hike. He assumes that each
person can take up to W pounds in his knapsack. There are N possible items to choose from.
Each unit of item i weighs w
i
pounds. The instructor assigns a number U
i
> 0 for each
unit of item i. These numbers represent the relative utility of that item during the hike. How
many units of each item should be placed in each knapsack so as to maximize total utility?
Formulate this problem by DP.
9.2 Continuous-time DP
We consider a continuous-time version of (9.2):
Maximize

t
f
0
f
0
(t, x(t), u(t))dt + Φ(x(t
f
))
subject to
dynamics: ˙ x(t) = f(t, x(t), u(t)) , t
0
≤ t ≤ t
f
initial condition: x(0) = x
0
,
control constraint: u : [t
0
, t
f
] → Ω and u() piecewise continuous.
(9.6)
In (9.6), x ∈ R
n
, u ∈ R
p
, Ω ⊂ R
p
. Φ : R
n
→ R is assumed differentiable and f
0
, f are assumed
to satisfy the conditions stated in VIII.1.1.
As before, for t
0
≤ t ≤ t
f
and x ∈ R
n
, let V (t, x) be the maximum value of the objective
function over the interval [t, t
f
] starting in state x at time t. Then it is easy to see that V must satisfy
V (t, x) = Max¦

t+∆
t
f
0
(τ, x(τ), u(τ))dτ
+V (t + ∆, x(t + ∆))[u : [t, t + ∆] → Ω¦, ∆ ≥ 0 ,
(9.7)
and
V (t
f
, x) = Φ(x) . (9.8)
9.2. CONTINUOUS-TIME DP 125
In (9.7), x(τ) is the solution of
˙ x(τ) = f(τ, x(τ), u(τ)) , t ≤ τ ≤ t + ∆ ,
x(t) = x .
Let us suppose that V is differentiable in t and x. Then from (9.7) we get
V (t, x) = Max¦f
0
(t, x, u)∆ +V (t, x) +
∂V
∂x
f(t, x, u)∆
+
∂V
∂t
(t, x)∆ +o(∆)[u ∈ Ω¦, ∆ > 0 .
Dividing by ∆ > 0 and letting ∆ approach zero we get the Hamilton-Jacobi- Bellman partial
differentiable equation for the value function:
∂V
∂t
(t, x) + Max¦f
0
(t, x, u) +
∂V
∂x
(t, x)f(t, x, u)[u ∈ Ω¦ = 0. (9.9)
Theorem 1: Suppose there exists a differentiable function V : [t
0
, t
f
] R
n
→ R which satisﬁes
(9.9) and the boundary condition (9.8). Suppose there exists a function ψ : [t
0
, t
f
] R
n
→ Ω
with ψ piecewise continuous in t and Lipschitz in x, satisfying
f
0
(t, x, ψ(t, x)) +
∂V
∂x
f(t, x, ψ(t, x))
= Max¦f
0
(t, x, u) +
∂V
∂x
f(t, x, u)[u ∈ Ω¦ .
(9.10)
Then ψ is an optimal feedback control for the problem (9.6), and V is the value function.
Proof: Let t ∈ [t
0
, t
f
] and x ∈ R
n
. Let ˆ u : [t, t
f
] → Ω be any piecewise continuous control and
let ˆ x(τ) be the solution of
·
ˆ x (τ) = f(τ, ˆ x(τ), ˆ u(τ)) , t ≤ τ ≤ t
f
,
ˆ x(t) = x .
(9.11)
Let x

(τ) be the solution of
˙ x

(τ) = f(τ, x

(τ), ψ(τ, x

(τ))) , t ≤ τ ≤ t
f
,
x

(τ) = x .
(9.12)
Note that the hypothesis concerning ψ guarantees a solution of (9.12). Let u

(τ) = ψ(τ, x

, (τ)), t ≤
τ ≤ t
f
. To show that ψ is an optimal feedback control we must show that

t
f
t
f
0
(tτ, x

(τ), u

(τ))dτ + Φ(x

(τ))

t
f
t
f
0
(τ, x

(τ), ˆ u(τ))dτ + Φ(ˆ x(t
f
)) .
(9.13)
To this end we note that
V (t
f
, x

(t
f
)) −V (t, x

(t)) =

t
f
f
dV

(τ, x

(τ))dτ
=∈
t
f
t
¦
∂V
∂τ
(τ, x

(τ) +
∂V
∂x
˙ x

(τ)¦dτ
= −

t
f
t
F −0(τ, x

(τ), u

(τ))dτ ,
(9.14)
126 CHAPTER 9. DYNAMIC PROGRAMING
using (9.9), (9.10). On the other hand,
V (t
f
, ˆ x(t
f
)) −V (t, ˆ x, (t)) =

t
f
t
¦
∂V
∂τ
(τ, ˆ x(τ)) +
∂V
∂x
·
˜ x (τ)¦dτ
≤ −

t
f
t
f
0
(τ, ˆ x(τ), ˆ u

(τ))dτ ,
(9.15)
using (9.9). From (9.14), (9.15), (9.8) and the fact that x

(t) = ˆ x(t) = x we conclude that
V (t, x) = Φ(x

(t
f
)) +

t
f
t
f
0
(τ, x

(τ), u

(τ))
≥ Φ(ˆ x(t
f
)) +

t
f
t
f
0
(τ, ˆ x(τ), ˆ u(τ))dτ
so that (9.13) is proved. It also follows that V is the maximum value function. ♦
• Exercise 1: Obtain the value function and the optimal feedback control for the linear regula-
tory problem:
Minimize
1
2
x

(T)P(T)x(t) +
1
2

T
0
¦x

(t)P(t)x(t)
+u

(t)Q(t)u(t)¦dt
subject to
dynamics: ˙ x(t) = A(t)x(t) +B(t)u(t) , 0 ≤ t ≤ T ,
initial condition: x(0) = x
0
,
control constraint: u(t) ∈ R
p
,
where P(t) = P

(t) is positive semi-deﬁnite, and Q(t) = Q

(t) is positive deﬁnite. [Hint:
Obtain the partial differential equation satisﬁed by V (t, x) and try a solution of the form
V (t, x) = x

R(t)x where R is unknown.]
9.3 Miscellaneous Remarks
There is vast literature dealing with the theory and applications of DP. The most elegant applications
of DP are to various problems in operations research where one can obtain “closed-form” analytic
solutions to be recursion equation for the value function. See (Bellman and Dreyfus [1952]) and
(Wagner [1969]). In the case of sequential decision-making under uncertainties DP is about the
only available general method. For an excellent introduction to this area of application see (Howard
[1960]). For an important application of DP to computational considerations for optimal control
problems see (Jacobson and Mayne [1970]). Larson [1968] has developed computational tech-
niques which greatly increase the range of applicability of DP where closed-form solutions are not
available. Finally, the book of Bellman [1957] is still excellent reading. []
Bibliography
[1] J.J. Arrow and L. Hurwicz. Decentralization and Computation in Resource in resource Allo-
cation. Essays in Economics and Econometrics. University of North Carolina Press, 1960. in
Pfouts R.W. (ed.).
[2] M. Athans and P.L. Falb. Optimal Control. McGraw-Hill, 1966.
[3] A.V. Balakrishnan and L.W. Neustadt. Computing Methods in Optimization Problems. Aca-
demic Press, 1964.
[4] K. Banerjee. Generalized Lagrange Multipliers in Dynamic Programming. PhD thesis, Col-
lege of Engineering, University of California, Berkeley, 1971.
[5] R.E. Bellman. Dynamic Programming. Princeton University Press, 1957.
[6] R.E. Bellman and S.E. Dreyfus. Applied Dynamic Programming. Princeton University Press,
1962.
[7] D. Blackwell and M.A. Girshick. Theory of Games and Statistical Decisions. John Wiley,
1954.
[8] J.E. Bruns. The function of operations research specilists in large urban schools. IEEE Trans.
on Systems Science and Cybernetics, SSC-6(4), 1970.
[9] A.E. Bryson and Y.C. Ho. Applied Optimal Control. Blaisdell, 1969.
[10] J.D. Cannon, C.D. Cullum, and E. Polak. Theory of Optimal Control and Mathematical Pro-
gramming. McGraw-Hill, 1970.
[11] G. Dantzig. Linear Programming and Extensions. Princeton University Press, 1963.
[12] C.A. Desoer. Notes for a Second Course on Linear Systems. Van Nostrand Reinhold, 1970.
[13] S.W. Director and R.A. Rohrer. On the design of resistance n-port networks by digital com-
puter. IEEE Trans. on Circuit Theory, CT-16(3), 1969a.
[14] S.W. Director and R.A. Rohrer. The generalized adjoint network and network sensitivities.
IEEE Trans. on Circuit Theory, CT-16(3), 1969b.
[15] S.W. Director and R.A. Rohrer. Automated network design–the frequency-domain case. IEEE
Trans. on Circuit Theory, CT-16(3), 1969c.
127
128 BIBLIOGRAPHY
[16] R. Dorfman and H.D. Jacoby. A Model of Public Decisions Illustrated by a Water Pollution
Policy Problem. Public Expenditures and Policy Analysis. Markham Publishing Co, 1970. in
Haveman, R.H. and Margolis, J. (eds.).
[17] R. Dorfman, P.A. Samuelson, and R.M.Solow. Linear Programming and Economic Analysis.
McGraw-Hill, 1958.
[18] R.I. Dowell and R.A. Rohrer. Automated design of biasing circuits. IEEE Trans. on Circuit
Theory, CT-18(1), 1971.
[19] Dowles Foundation Monograph. The Economic Theory of Teams. John Wiley, 1971. to appear.
[20] W.H. Fleming. Functions of Several Variables. Addison-Wesley, 1965.
[21] C.R. Frank. Production Theory and Indivisible Commodities. Princeton University Press,
1969.
[22] D. Gale. A geometric duality theorem with economic applications. Review of Economic
Studies, XXXIV(1), 1967.
[23] A.M. Geoffrion. Duality in Nonlinear Programming: a Simpliﬁed Application-Oriented Treat-
ment. The Rand Corporation, 1970a. Memo RM-6134-PR.
[24] A.M. Geoffrion. Primal resource directive approaches for optimizing nonlinear decomposable
programs. Operations Research, 18, 1970b.
[25] F.J. Gould. Extensions of lagrange multipliers in nonlinear programming. SIAM J. Apl. Math,
17, 1969.
[26] J.J. Greenberg and W.P. Pierskalla. Surrogate mathematical programming. Operations Re-
search, 18, 1970.
[27] H.J.Kushner. Introduction to Stochastic Control. Holt, Rinehart, and Winston, 1971.
[28] R.A. Howard. Dynamic Programming and Markov Processes. MIT Press, 1960.
[29] R. Isaacs. Differential Games. John Wiley, 1965.
[30] D.H. Jacobson, M.M. Lele, and J.L. Speyes. New necessary condtions of optimality for prob-
lems with state-variable inequality constraints. J. Math. Analysis and Applications, 1971. to
appear.
[31] D.H. Jacobson and D.Q. Mayne. Differential Dynamic Programming. American Elsevier
Publishing Co., 1970.
[32] S. Karlin. Mathematical Methods and Theory in Games, Programming, and Economics, vol-
[33] H.J. Kelley. Method of Gradients. Optimization Techniques. Academic Press, 1962. in Leit-
mann, G.(ed.).
BIBLIOGRAPHY 129
[34] J.H. Kelley, R.E. Kopp, and H.G. Mayer. Singular Extremals. Topics in Optimization. Aca-
demic Press, 1970. in Leitman, G. (ed.).
[35] D.A. Kendrick, H.S. Rao, and C.H. Wells. Water quality regulation with multiple polluters. In
Proc. 1971 Jt. Autom. Control Conf., Washington U., St. Louis, August 11-13 1971.
[36] T.C. Koopmans. Objectives, constraints, and outcomes in optimal growth models. Economet-
rica, 35(1), 1967.
[37] H.W. Kuhn and A.W. Tucker. Nonlinear programming. In Proc. Second Berkeley Symp. on
Math. Statistics and Probability. University of California Press, Berkeley, 1951.
[38] R.E. Larson. State Increment Dynamic Programming. American Elsevier Publishing Co.,
1968.
[39] L.S. Lasdon, S.K. Mitter, and A.D. Waren. The conjugate gradient method for optimal control
problems. IEEE Trans. on Automatic Control, AC-12(1), 1967.
[40] E.B. Lee and L. Markus. Foundation of Optimal Control Theory. John Wiley, 1967.
[41] R. Luce and H. Raiffa. Games and Decisions. John Wiley, 1957.
[42] D.G. Luenberger. Quasi-convex programming. Siam J. Applied Math, 16, 1968.
[43] O.L. Mangasarian. Nonlinear Programming. McGraw-Hill, 1969.
[44] S.R. McReynolds. The successive sweep method and dynamic programming. J. Math. Analy-
sis and Applications, 19, 1967.
[45] J.S. Meditch. Stochastic Optimal Linear Estimation and Control. McGraw-Hill, 1969.
[46] M.D. Mesarovic, D. Macho, and Y. Takahara. Theory of Hierarchical, Multi-level Systems.
[47] C.E. Miller. The Simplex Method for Local Separable Programming. Recent Advance Pro-
gramming. McGraw-Hill, 1963. in Graves, R.L. and Wolfe, P. (eds.).
[48] L.W. Neustadt. The existence of optimal controls in the absence of convexity conditions. J.
Math. Analysis and Applications, 7, 1963.
[49] L.W. Neustadt. A general theory of extremals. J. Computer and System Sciences, 3(1), 1969.
[50] H. Nikaido. Convex Structures and Economic Theory. Academic Press, 1968.
[51] G. Owen. Game Theory. W.B. Saunders & Co., 1968.
[52] E. Polak. Computational Methods in Optimization: A Uniﬁed Approach. Academic Press,
1971.
[53] L.S. Pontryagin, R.V. Boltyanski, R.V. Gamkrelidze, and E.F.Mischenko. The Mathematical
Theory of Optimal Processes. Interscience, 1962.
130 BIBLIOGRAPHY
[54] R.T. Rockafeller. Convex Analysis. Princeton University Press, 1970.
[55] M. Sakarovitch. Notes on Linear Programming. Van Nostrand Reinhold, 1971.
[56] L.J. Savage. The Foundation of Statistics. John Wiley, 1954.
[57] K. Shell. Essays in the Theory of Optimal Economic Growth. MIT Press, 1967.
[58] R.M. Solow. The economist’s approach to pollution and its control. Science, 173(3996), 1971.
[59] D.M. Topkis and A. Veinnott Jr. On the convergence of some feasible directions algorithms
for nonlinear programming. SIAM J. on Control, 5(2), 1967.
[60] H.M. Wagner. Principles of Operations Research. Prentice-Hall, 1969.
[61] P. Wolfe. The simplex method for quadratic programming. Econometrica, 27, 1959.
[62] W.M. Wonham. On the seperation theorem of optimal control. SIAM J. on Control, 6(2), 1968.
[63] E. Zangwill. Nonlinear Programming: A Uniﬁed Approach. Prentice-Hall, 1969.
Index
Active constraint, 50
augmented, 98
continuous-time, 85
augmented, 105
continuous-time, 91
discrete-time, 80
Afﬁne function, 54
Basic feasible solution, 39
basic variable, 39
Certainty-equivalence principle, 5
Complementary slackness, 34
Constraint qualiﬁcation
deﬁnition, 53
sufﬁcent conditions, 55
Continuous-time optimal control
necessary condition, 101, 103
problem formulation, 101, 103
sufﬁcient condition, 91, 125
Control of water quality, 67
Convex function
deﬁnition, 37
properties, 37, 54, 55
Convex set, 37
Derivative, 8
Design of resistive network, 15
Discrete-time optimal control
necessary condition, 78
problem formulation, 77
sufﬁcient condition, 123
Discrete-time optimality control
sufﬁcient condition, 80
Dual problem, 33, 58
Duality theorem, 33, 63
Dynamic programming, DP
optimality conditions, 123, 125
problem formulation, 121, 124
Epigraph, 61
Equilibrium of an economy, 45, 64
Farkas’ Lemma, 32
Feasible direction, 72
algorithm, 71
Feasible solution, 33, 49
Game theory , 5
Hamilton-Jacobi-Bellman equation, 125
Hamiltonian H,
˜
H, 78, 99
Hamiltonian H
˜
H, 101
Hypograph, 61
Knapsack problem, 124
Lagrange multipliers, 37
Lagrangian function, 35
Langrangian function, 21, 54
Langrangian multipliers, 21
Linear programming, LP
duality theorem, 33, 35
problem formulation, 31
theory of the ﬁrm, 42
Linear programming,LP
optimality condition, 34
Maximum principle
continuous-time, 86, 91, 101, 103
discrete-time, 80
Minimum fuel problem, 81
Minimum-time problem, 107
131
132 INDEX
example, 108
Non-degeneracy condition, 39
Nonlinear programming, NP
duality theorem, 63
necessary condition, 50, 53
problem formulation, 49
suﬁcient condition, 54
Optimal decision, 1
Optimal economic growth, 2, 113, 117
Optimal feedback control, 123, 125
Optimization over open set
necessary condition, 11
sufﬁcient condition, 13
Optimization under uncertainty, 4
Optimization with equality constraints
necessary condition, 17
sufﬁcient condition, 21
Optimum tax, 70
Primal problem, 33
optimality condition, 70
problem formulation, 70
Wolfe algorithm, 71
Recursion equation for dynamic programming,
124
Regulator problem, 81, 112
Resource allocation problem, 65
Separation theorem for convex sets, 73
Separation theorem for stochastic control, 5
Simplex algorithm, 37
Phase I, 41
Phase II, 39
Singular case for control, 113
Slack variable, 32
State-space constraint
continuous-time problem, 117
discrete-time problem, 77
Supporting hyperplane, 61, 84
Tangent, 50
Transversality condition
continuous-time problem, 91
discrete-time problem, 80
Value function, 123
Variable ﬁnal time, 103
Vertex, 38
Weak duality theorem, 33, 58
Wolfe algorithm, 71

ii

Contents
1 INTRODUCTION 2 OPTIMIZATION OVER AN OPEN SET 3 Optimization with equality constraints 4 Linear Programming 5 Nonlinear Programming 6 Discrete-time optimal control 7 Continuous-time linear optimal control 8 Coninuous-time optimal control 9 Dynamic programing 1 7 15 27 49 75 83 95 121

iii

iv CONTENTS .

the World Wide Web has again made it possible to publish cheaply. edited by George L. accessible treatments of graduate-level material in inexpensive books (the price of a book in the series was about ﬁve dollars). I would appreciate knowing if you ﬁnd any mistakes in the book. However. Berkeley. Notes on Optimization has been out of print for 20 years.PREFACE to this edition Notes on Optimization was published in 1971 as part of the Van Nostrand Reinhold Notes on System Sciences. Turin. several people have been using it as a text or as a reference in a course. They have urged me to re-publish it.P. Books have since become expensive. I thank Kate Klohe for doing just that. The effort was successful for several years. Varaiya v . Our aim was to publish short. or if you have suggestions for (small) changes that would improve it. The only obstacle was to retype the manuscript in LaTex. The idea of making it freely available over the Web was attractive because it reafﬁrmed the original aim. California September. However. 1998 P. Van Nostrand Reinhold was then purchased by a conglomerate which cancelled Notes on System Sciences because it was not sufﬁciently proﬁtable.

vi CONTENTS .

and linear differential equations (transition matrix. Jacob. an understanding of this material should enable the reader to follow much of the recent technical literature on nonlinear programming.PREFACE These Notes were developed for a ten-week course I have taught for the past three years to ﬁrst-year graduate students of the University of California at Berkeley. Finally. the main concepts and techniques of mathematical programming and optimal control to students having diverse technical backgrounds. and Mr. Billie Vrtiak for her marvelous typing in spite of starting from a not terribly legible handwritten manuscript. Chapter V must be read before Chapter VI. Turin for his encouraging and patient editorship. J-P. has been inﬂuenced by many of my students and colleagues. as well as their presentation. A. adjoint solution) is sufﬁcient for the reader to follow the Notes. Ripper. To facilitate the use of these Notes as a textbook. who have read and criticized earlier drafts. A reasonable knowledge of advanced calculus (up to the Implicit Function Theorem). The examples and exercises given in the text form an integral part of the Notes and most readers will need to attend to them before continuing further. and Chapter VII before Chapter VIII. The selection of topics. M. Berkeley.A.L. However. basis. linear algebra (linear independence. in a compact and uniﬁed manner. 1971 P. I also want to thank Mrs. and mathematical economics. The treatment of the topics presented here is deep. I want to thank Professor G. Varaiya vii . C. I would especially like to acknowledge the help of Professors M. E. My objective has been to present. Desoer.P. Athans. Cohen. I have incurred the cost of some repetition in order to make almost all chapters self-contained. matrix inverse). (deterministic) optimal control. California November. Although the coverage is not encyclopedic. Polak.

viii CONTENTS .

namely. In order to model a decision-making situation in mathematical terms. Some illustrations will help. In an agreement signed with Tough’s undergraduates and graduates (TUGs). The cost of each decision is known. “quality” is 1 . An optimal decision is then any decision which incurs the least cost among the set of permissible decisions. and brieﬂy introduce two more general models which we cannot discuss further in these Notes. 1. When these conditions are satisﬁed. and what price p should it set? Example 2: Tough University provides “quality” education to undergraduate and graduate students. The phrase complete information is given means that the following requirements are met: 1. Example 1: The Pot Company (Potco) manufacturers a smoking blend called Acapulco Gold. The blend is made up of tobacco and mary-john leaves. If Potco wants to maximize its proﬁts. For legal reasons the fraction α of maryjohn in the mixture must satisfy 0 < α < 1 . and 2. we present our model of the optimal decision-making problem. certain further requirements must be satisﬁed. From extensive market research Potco has determined 2 their expected volume of sales as a function of α and the selling price p. Furthermore. illustrate decisionmaking situations by a few examples. The set of all decisions can be adequately represented as a subset of a vector space with each vector representing a decision. whereas the cost of mary-john is a function of the amount purchased.Chapter 1 INTRODUCTION In this chapter.1 The Optimal Decision Problem These Notes show how to arrive at an optimal decision assuming that complete information is given. how much mary-john and tobacco should it purchase. tobacco can be purchased at a ﬁxed price. The set of all permissible decisions is known. 1. The cost corresponding to these decisions is given by a real-valued function. and 2. the decisions can be ranked according to whether they incur greater or lesser cost.

Mr. 1] is the fraction of output which is saved and invested. W (t) = C(t) + I(t) = (1 − s(t))W (t) where s(t) = I(t)/W (t) . one of which is a seminar and the rest of which are lecture courses. The University has a faculty of 1000.1) = −δK(t) + s(t)F (K(t). wine. and the remainder I(t) to investment in capital goods.2 CHAPTER 1.1. INTRODUCTION deﬁned as follows: every year. The labor force is growing at a constant birth rate of β > 0. I. Subject to the agreements with the TUGs and WORs how many u’s and g’s should the President admit to maximize his rating? Example 3: (See Figure 1. Hence.) Thus. each u (undergraduate) must take eight courses. If it costs \$c per cubic foot to excavate or ﬁll the ground. L(t)). Suppose that the capital stock decays exponentially with time at a rate δ > 0. then the rate of output of wine W (t) at time is given by the production function W (t) = F (K(t). The only requirement is that the ﬁnal road should not have a slope exceeding 0.1: Admissable set of example.) An engineer is asked to construct a road (broken line) connection point a to point b. whereas every senior faculty member (there are 250 of these) shall teach three lecture courses and three seminars each year. There are two factors of production. whereas each g (graduate) must take two seminars and ﬁve lecture courses. Shell is the manager of an economy which produces one output. The current proﬁle of the ground is given by the solid line. (Obviously. how should he design the road to meet the speciﬁcations at minimum cost? Example 4: Mr. so that the net rate of growth of capital is given by the following equation: ˙ K(t) = d K(t) dt = −δK(t) + s(t)W (t) (1. The Weary Old Radicals (WORs) have a contract with the University which stipulates that every junior faculty member (there are 750 of these) shall be required to teach six lecture courses and two seminars each year. A seminar cannot have more than 20 students and a lecture course cannot have more than 40 students. ∈ [0. C.001. Figure 1. The Regents of Touch rate Tough’s President at α points per u and β points per g “processed” by the University. L(t)) As Manager. If K(t) and L(t) respectively are the capital stock used and the labor employed at time t. b a . and K are being measured in a common currency. W . . capital and labor. Shell allocates some of the output rate W (t) to the consumption rate C(t).

whence the consumption per capita of labor becomes c(t) = (l − s(t))f (k(t)). L) for all λ > 0. In the second example.1. maximizing or minimizing) a real-valued function over a vector space subject to constraints. constrained by the number of faculty and the agreements with the TUGs and WORs is a permissible decision. Shell’s savings policy s(t). L)−LF (K/L. Example 2 is caricature (see also a faintly related but more more elaborate formulation in Bruno [1970]). p) satisfying the constraints 0 < α < 1 and 0 < 2 p. constrained by 0 ≤ s(t) ≤ 1. In the last example. any vector (u. (In connection with this example and related models see the critique by Koopmans [1967]. If we deﬁne the relevant variable in terms of per capita of labor. For instance. in the ﬁrst example. then we see that F (K. ˙ k(t) = s(t)f (k(t)) − µk(t) (1.e. T ] is identiﬁed with total consumption 0 c(t)dt.e. a permissible decision is any two-dimensional vector (α.1) and (1. w = W/L. Shell starts with capital-to-labor T ratio ko . Thus. the value of the mathematical exercise is greater the more insensitive are the optimum savings policies with respect to the simplifying assumptions of the mathematical model. We must always remember that a mathematical formulation is inevitably an abstraction and the gain in precision may have occurred at a great loss of realism.. In the latter case. the set of permissible decisions is represented by the set of all points in some vector space which satisfy certain constraints. The ﬁrst term of the right-hand side in (3) is the increase in the capital-to-labor ratio due to investment whereas the second terms is the decrease due to depreciation and increase in the labor force. whereas Example 4 is light-years away from reality.3). and at time 0 Mr. g ≥ 0. what should Mr. The constraints themselves are presented in terms of functional inequalities or equalities. Using these deﬁnitions and equations (1. 0 ≤ t ≤ T .3) where µ = (δ + β). (1. If “welfare” over the planning period [0. λL) = λF (K. k = K/l. . (It is of mathematical but not conceptual interest to note that in this case a decision is represented by a vector in a function space which is inﬁnite-dimensional. g) with u ≥ 0. 1) = Lf (k). a permissible decision is any real-valued function s(t). F (λK. c = C/L. i. Suppose there is a planning horizon time T .) More concisely then. and if we let f (k) = F (k. the welfare function becomes 0 e− αt c(t)dt. What is the optimum policy corresponding to this criterion? These examples illustrate the kinds of decision-making problems which can be formulated mathematically so as to be amenable to solutions by the theory presented in these Notes. be so as to maximize welfare? What savings policy maximizes welfare subject to the additional restriction that the capital-to-labor ratio at time T should be at least kT ? If future consumption is discounted at rate α > 0 and if time horizon ∞ is ∞.2) Suppose that the production function F exhibits constant returns to scale.2) it is easy to see that K(t) satisﬁes the differential equation (1. these Notes are concerned with optimizing (i. 0 ≤ t ≤ T . l).) In the examples above. THE OPTIMAL DECISION PROBLEM 3 ˙ L(t) = βL(t).1.

We can then redeﬁne the problem as minimizing the ﬁrst cost function (total time for trips) subject to the constraint that the waiting time for any trip is less than some reasonable bound (say one minute). Since we cannot study these more general models in these Notes. it may be the case that the optimum decision according to the ﬁrst criterion may be lead to very long waiting times for a few trips. How should \$1. probability 0. the hypothesis of complete information can be relaxed by allowing that decision-making occurs in an uncertain environment.25 that it drops by \$100. It is customary to model this uncertainty stochastically. For instance. we merely point out here some situations where such models arise naturally and give some references. etc. Usually there are impurities in the chemicals and disturbances in the heating process which may be regarded as additional inputs of a .25 that the price is unchanged. Indeed.4 CHAPTER 1. In the second place. input rates of various chemicals. we need a criterion by which we can compare different decisions. may be quite artiﬁcial in practice. This interchangeability of goal and constraints also appears at a deeper level in much of the mathematical theory. although convenient for presenting the mathematical theory. and probability 0. One way of doing this is to assign as cost to each decision the total amount of time taken to make all the trips within this section. we can replace the single decision-maker by a group of two or more agents whose collective decision determines the outcome. In this way.1? As another example.2 Some Other Models of Decision Problems Our model of a single decision-maker with complete information can be generalized along two very important directions.000 be invested so as to maximize the expected value of the capital gains subject to the constraint that the probability of losing more than \$100 is less than 0. Thus. Before we can begin to suggest a design.5 to the event that the price of shares in Glamor company increases by \$100. 1.2. We will see that in most of the results the objective function and the functions describing the constraints are treated in the same manner. He wants to maximize his capital gains. INTRODUCTION At this point.1 Optimization under uncertainty. In the ﬁrst place. it is important to realize that the distinction between the function which is to be optimized and the functions which describe the constraints. The two objectives are incompatible. An alternative and equally plausible goal may be to minimize the maximum waiting time (that is the total time spent at stop lights) in each trip. A similar model is made for all the other stocks that the investor is willing to consider. and a decision problem can be formulated as follows. The situation is different from our previous examples in that the outcome (future stock prices) is uncertain. since the stock which is likely to have higher gains is also likely to involve greater risk. Now it may happen that these two objective functions may be inconsistent in the sense that they may give rise to different orderings of the permissible decisions. the second goal (minimum waiting time) has been modiﬁed and reintroduced as a constraint.000 in the stock market. the investor may assign probability 0. which in this case are different patterns of trafﬁc-light durations. consider the design of a controller for a chemical process where the decision variable are temperature. A person wants to invest \$1. so that this decision is far from optimum according to the second criterion. we need a criterion to determine what is meant by “optimum trafﬁc ﬂow. suppose we have to choose the durations of various trafﬁc lights in a section of a city so as to achieve optimum trafﬁc ﬂow.” More abstractly. and at the same time minimize the risk of losing his money. 1. Let us suppose that we know the transportation needs of all the people in this section.

The place is a large circular ﬁeld. just as in the case of the portfolioselection problem. it is of great significance to know that. however.. If the uncertainties are modelled stochastically as in the example above. an optimal decision problem under uncertainty is equivalent to another optimal decision problem under complete information.) In the author’s opinion. Blackwell and Girshick [1954]) and on Stochastic Optimal Control (Meditch [1969].. known as the Certainty-Equivalence principle in economics has been extended and baptized the Separation Theorem in the control literature. heavy car which does not maneuver easily. Agent Alpha is chasing agent Beta. the number of results available is pitifully small. Here there are two decision-makers with opposing objectives. The control theorist will probably be most interested in Isaacs [1965]. We need a new concept of rational (optimal) decision-making.) Unfortunately.2. What should Alpha do to get as close to Beta as possible? What should Beta do to stay out of Alpha’s reach? This situation is fundamentally different from those discussed so far. since a particular decision taken by our agent may be better or worse than another decision depending upon the (unknown) decisions taken by the other agents. The difﬁculty caused by the lack of knowledge of the actions of the other decision-making agents arises even if all the agents have the same objective. See Wonham [1968]. For instance. yet the effectiveness of his decision depends crucially upon the other’s decision. [1969]. Although problems involving many decision-makers are present in any system of large size. Although the practical impact of this theory is not great. it has proved to be among the most fruitful sources of unifying analytical concepts in the social sciences. et al.1. . 1. then in many cases the techniques presented in these Notes can be usefully applied to the resulting optimal decision problem. Alpha is driving a fast. To do justice to these decision-making situations. it is necessary to give great attention to the various ways in which the uncertainties can be modelled mathematically. so that optimality cannot be deﬁned as we did earlier. (This result. known as the Theory of Games. whereas Beta is riding a motor scooter. we can formulate a decision problem in such a way as to take into account these random disturbances.2 The case of more than one decision-maker. and Blaquiere. notably economics and political science. these problems represent one of the most important and challenging areas of research in decision theory. whereas the mathematical content of the theory is concisely displayed in Owen [1968]. et al. slow but with good maneuverability. (See Mesarovic. to be able to deal with these models. exists which describes and prescribes behavior in these situations.2. we need a good background in Statistics and Probability Theory besides the material presented in these Notes. [1970] and Marschak and Radner [1971]. It is of crucial importance to invent schemes to coordinate the actions of the individual decision-makers in a consistent manner. SOME OTHER MODELS OF DECISION PROBLEMS 5 random nature and modeled as stochastic processes. given appropriate conditions. Kushner [1971]). After this. The best single source for Game Theory is still Luce and Raiffa [1957]. We can only refer the reader to the extensive literature on Statistical Decision Theory (Savage [1954]. We also need to worry about ﬁnding equivalent but simpler formulations. Each agent does not know what the other is planning to do. Situations such as these have been studied extensively and an elaborate structure.

INTRODUCTION .6 CHAPTER 1.

this entry is sometimes also denoted by the lower case letter aij . If x ∈ R 2.1. its size will be clear from the context. . . . .Chapter 2 OPTIMIZATION OVER AN OPEN SET In this chapter we study in detail the ﬁrst example of Chapter 1. . we write In to denote the n × n identity matrix. . 2. . and different vectors denoted by the same symbol are distinguished by superscripts as in xj and xk . If confusion is likely. . . n. . n. . xn ) and y = (y1 . Then we study our example.2 If x = (x1 . and Ai denotes the ith row of A. If A is an m × n matrix. We ﬁrst establish some notation which will be in force throughout these Notes. but no confusion will result. . This will generalize to a canonical problem. then x ≥ 0. . . xn ).1.1. .1 All vectors are column vectors. Note that Ai is a row vector. . . . Some additional properties are mentioned in the last section. . .5 below and some other minor and convenient exceptions in the text. 2.1. i = 1. .3 and 2. i = 1. Aj denotes the entry i of A in the ith row and jth column. and then we also write A = {aij }. .1 Notation 2. yn ) then x ≥ y means xi ≥ yi . In particular if x ∈ Rn . 7 . . matrix multiplication. the ith component of a vector x ∈ Rn is denoted xi . . + xn yn as in ordinary n we deﬁne |x| = + x x.1. yn ) √ then x y = x1 y1 + . . the properties of whose solution are stated as a theorem. then Aj denotes the jth column of A.3 Matrices are normally denoted by capital letters. 0 denotes both the zero vector and the real number zero. Vectors are normally denoted by lower case letters. if xi ≥ 0. . Prime denotes transpose so that if x ∈ Rn then x is the row vector x = (x1 . Thus if x = (x1 . . . . . . I denotes the identity matrix. . . xn ) and y = (y1 . and x = (x1 . with two consistent exceptions mentioned in 2. . xn ) .

y ) = ((∂f /∂y1 )(ˆ.  . For example. Sometimes we describe a function by specifying a rule to calculate f (x) for every x.8 CHAPTER 2. (∂f /∂xn )(ˆ)).. =  . If f : (x. . .1. Finally. in terms of the notation in Section 2. then its derivative at x is the m × n matrix ˆ  x f1x (ˆ) ∂f   . x x i x fxx (ˆ) = (∂/∂x)(fx ) (ˆ). . .2 Example We consider in detail the ﬁrst example of Chapter 1. y) is a differentiable function from ˆ Rn × Rm into R. . ˆ x x This derivative is denoted by (∂f /∂x)(ˆ) or fx (ˆ) or ∂f /∂x|x=ˆ or fx |x=ˆ . p).5 If f : Rn → R is a differentiable function. if f : Rn → Rm is x ˆ x ˆ x ˆ x ˆ a differentiable function with components f1 . m. ∂fm x ∂xn (ˆ) 2. . . In this case we write f : x → f (x). x  ∂x1. . we can write F : x → Ax to denote the function f : Rn → Rm whose value at a point x ∈ Rn is Ax. ∂x fmx (ˆ) x  ∂f1 (ˆ) . x x and is called the gradient of f at x. ∂fm (ˆ) x ∂x1  ∂f1 x ∂xn (ˆ)     . y ) = (∂f /∂y)(ˆ. its second derivative at x is the n×n matrix (∂ 2 f /∂x∂x)(ˆ) = ˆ x fxx (ˆ) where (fxx (ˆ))j = (∂ 2 f /∂xj ∂xi )(ˆ). Note that fi : Rn → R. y ) = ((∂f /∂x1 )(ˆ. and if the argument x x x ˆ x x is clear from the context it may be dropped. y )).1. . . y ). p) = expected sales volume (as determined by market research) of mixture as a function of(α. . y ) is the n-dimensional x ˆ row vector fx (ˆ. Thus.1. . (∂f /∂ym )(ˆ. the partial derivative of f with respect to x at the point (ˆ. . fm . . if A is an m × n matrix.6 If f : Rn → R is twice differentiable. . i = 1. (∂f /∂xn )(ˆ. f (α. y ). Deﬁne the following variables and functions: α = fraction of mary-john in proposed mixture. its ith component is written fi . . . p = sale price per pound of mixture. ˆ (ˆ) = fx x =  x  . . the derivative of f at x is the row vector ((∂f /∂x1 )(ˆ). . The column vector (fx (ˆ)) is also denoted x f (ˆ). . .. y )).4 If f : Rn → Rm is a function. 2.1. . . and similarly x ˆ x ˆ x ˆ x ˆ fy (ˆ. . x x 2. . y) → f (x. v = total amount of mixture produced. y ) = (∂f /∂x)(ˆ. OPTIMIZATION OVER AN OPEN SET 2. .5 above. .

p) − (α∗ . p) = R(α.3) . p)) + P2 (1 − α)f (α. p) − C(α. (2. and t = amount (in pounds) of tobacco purchased. i.. p) ∈ Ω whenever |(α. The revenue is R(α. Formally. p∗ ) + δ(h1 . p)|0 < α < 1 . so that the net proﬁt is N (α.1) We are going to establish some properties of (a∗ . p) ∈ Ω. p∗ ) ∈ Ω and N (α∗ . p is C(α. p) = pf (α. p). Hence there exits ε > 0 such that (α. First of all we note that Ω is an open subset of R2 . p) = P1 (αf (α. p) for all (α. 0 < p < ∞}.2. h2 )) ∈ Ω for 0 ≤ δ ≤ η (2. and P2 = purchase price per pound of tobacco. EXAMPLE Since it is not proﬁtable to produce more than can be sold we must have: v = f (α. Let P1 (m) = purchase price of m pounds of mary-john. and t = (l − α)v.2) implies that for every vector h = (h1 . m = αv. where Ω = {(α. Suppose that (α∗ . p). p) ∈ Ω. 9 Evidently.e. p). p∗ )| < ε (2. we 2 have the the following decision problem: Maximize N (α. h2 ) in R2 there exists η > 0 (η of course depends on h) such that ((α∗ . p).2. (α∗ . Then the total cost as a function of α. subject to (α. p∗ ) ≥ N (α. p). The set of admissible decisions is Ω. p∗) is an optimal decision. p∗ ). m = amount (in pounds) of mary-john purchased.2) In turn (2.

OPTIMIZATION OVER AN OPEN SET 1 2 (α∗ . (2. p∗ )h1 + ∂α ∂N ∗ ∗ ∂p (α . (α∗ . Combining (2.5) → 0 as δ → 0. p )h2 ] ∂N ∗ ∗ ∂p (α . (2.1) we obtain (2. h2 ) .7). p∗ ) ≥ N (α∗ + δh1 . Ω δh (a∗ . using the facts that N is differentiable. ∂N ∗ ∗ ∂p (α . p ) = 0.4): N (α∗ . . we have concluded that the inequality (2.5) into (2.6) Substitution of (2. p∗ ) + δ(h1 . p )h2 ] + o(δ). p∗ ) | h p Figure 2.1: Admissable set of example. let us prove a direct generalization. p∗ )h1 + ∂α Dividing by δ > 0 gives 0 ≥ [ ∂N (α∗ .4) (2. p∗ )h1 + ∂α ∂N ∗ ∗ ∂p (α . p∗ + δh2 ) = +δ[ ∂N (δ∗ .7) Letting δ approach zero in (2.9) holds for every vector h ∈ R2 . p∗ ) N (α∗ + δh1 . Clearly this is possible only if ∂N ∗ ∗ ∂α (α . (2. p∗ )h1 + ∂α +o(δ). p ) = 0.6) we get 0 ≥ [ ∂N (α∗ . p )h2 ] (2.3) with (2.4) yields 0 ≥ δ[ ∂N (α∗ .8). where oδ δ ∂N ∗ ∗ ∂p (α .9) Before evaluating the usefulness of property (2. + o(δ) δ .10 α CHAPTER 2. and using (2. p∗ + δh2 ) for 0 ≤ δ ≤ η Now we assume that the function N is differentiable so that by Taylor’s theorem N (α∗ . (2.8) Thus. p∗ ) is optimal. p )h2 ]. and δ is open.

17) Letting δ approach zero in (2.3 The Main Result and its Consequences 2.10) = 0. Then ∂f ∗ ∂x (x ) (2. by Taylor’s theorem we have f (x∗ + δh) = f (x∗ ) + where o(δ) δ ∂f ∗ ∂x (x )δh (2. we see that 0≥ ∂f ∗ ∂x (x )h.11) Proof: Since x∗ ∈ Ω and Ω is open.14) + o(δ). ♦ .12) In turn. there exists ε > 0 such that x ∈ Ω whenever |x − x∗ | < ε.3. (2. Let x∗ be an optimal solution of the following decision-making problem: Maximize f (x) subject to x ∈ Ω.12) implies that for every vector h ∈ Rn there exits η > 0 (η depending on h) such that (x∗ + δh) ∈ Ω whenever 0 ≤ δ ≤ η.15) into (2. (2. Since x∗ is optimal. we must have 0= and the theorem is proved.13) (2. Let Ω be an open subset of Rn .17) and taking (2.14) yields 0 ≥ δ ∂f (x∗ )h + o(δ) ∂x and dividing by δ > 0 gives 0≥ ∂f ∗ ∂x (x )h + o(δ) δ (2.16) Substitution of (2.18) Since the inequality (2.15) → 0 as δ → 0 (2. ∂f ∗ ∂x (x ).16) into account. Since f is differentiable. (2.3. THE MAIN RESULT AND ITS CONSEQUENCES 11 2.2. (2.1 Theorem .18) must hold for every h ∈ Rn . (2. we must then have f (x∗ ) ≥ f (x∗ + δh) whenever 0 ≤ δ ≤ η. Let f : Rn → R be a differentiable function.

p∗ ) ∂R ∗ ∗ ∂α (α . ∂f ∗ ∂xn (x ) =0 (2. marginal revenue = marginal cost. In each case Ω = (−1.4. Suppose that R and C are differentiable.6 below).1? Yes Yes No No No Further Consequences x∗ is the unique optimal 2. 2. . Note that in the last three ﬁgures there is no optimal decision since the limit points -1 and +1 are not in the set of permissible decisions Ω = (−1. . p ).3.2. p ). . in which case (2. . . There may exist decisions x ∈ Ω ˜ such that fx (˜) = 0 but x is not optimal. The diagrams in Figure 2. Let us evaluate the usefulness of (2. In practice this may be a very difﬁcult problem since these may be nonlinear equations and it may be necessary to use a digital computer.1 may x ˜ occur. .12 CHAPTER 2. In summary. the theorem does not give us any clues concerning the existence of an optimal decision. 1). OPTIMIZATION OVER AN OPEN SET Table 2. We have obtained an important economic insight. and it does not give us sufﬁcient conditions either. p ) = ∂C ∗ ∗ ∂p (α . We return to the example and recall the N = R − C.11) gives us n equations which must be satisﬁed at any optimal decision x∗ = (x∗ . More generally. ∂R ∗ ∗ ∂p (α . p ) = ∂C ∗ ∗ ∂α (α . in the language of economic analysis. Equation (2. . .4 Remarks and Extensions 2.11) and its special case (2.18) implies that at every optimal decision (α∗ . x∗ ) . The theorem may also have conceptual signiﬁcance.18). ∂f ∗ ∂x2 (x ) = 0. or. any one of the ﬁve cases in Table 2. .2. every optimal decision must be a solution of these n simultaneous equations of n variables.1 A warning.19).2 Consequences.4. 1).2 satisﬁed? Exactly one point.11) is only a necessary condition for x∗ to be optimal. Equation (2. so that the search for an optimal decision from Ω is reduced to searching among the solutions of (2. n 1 These are ∂f ∗ ∂x1 (x ) = 0. However.19) Thus.1 illustrate these cases. say x∗ More than one point None Exactly one point More than one point Case 1 2 3 4 5 Does there exist an optimal decision for 2. in these Notes we shall not be overly concerned with numerical solution techniques (but see 2.1 At how many points in Ω is 2.

3 Local optimum. and if f is continuous. so that dividing δ by δ2 > 0 yields 0 ≥ 1 h fxx (x∗ )h + 2 o(δ2 ) δ2 and letting δ approach zero we conclude that h fxx (x∗ )h ≤ 0 for all h ∈ Rn . Thus. This means that fxx (x∗ ) is a negative semi-deﬁnite matrix. Then fx (x∗ ) = 0.. 2.2. then +1 is the optimal decision but the derivative is positive at that point. if we have a twice differentiable objective function. . 2.4. Suppose f is twice-differentiable and let x∗ ∈ Ω be optimal or even locally optimal.11) for local optima also. REMARKS AND EXTENSIONS 13 -1 Case 1 1 -1 Case 2 1 -1 Case 3 1 -1 Case 4 1 -1 Case 5 1 Figure 2. we get an additional necessary condition. 2. 2 2 (2. 1]. Indeed.e. if Ω = [−1. then it follows by the Weierstrass Theorem that there exists an optimal decision. in the third ﬁgure above. Suppose at x∗ ∈ Ω. It is easy to see that the theorem holds (i.4.2 Existence.4. But then from the expansion (2.20) we can conclude that x∗ is a local optimum. But if Ω is closed we cannot assert that the derivative of f vanishes at the optimum.20) where o(δ2 ) → 0 as δ → 0.4.1.4 Second-order conditions. 2. and by Taylor’s theorem f (x∗ + δh) = f (x∗ ) + 1 δ2 h fxx (x∗ )h + o(δ2 ).5 Sufﬁciency for local optimal. Now for δ > 0 sufﬁciently small f (x∗ + δh) ≤ f (x∗ ). If the set of permissible decisions Ω is a closed and bounded subset of Rn .2: Illustration of 4.4. 2. We say that x∗ ∈ Ω is a locally optimal decision if there exists ε > 0 such that f (x∗ ) ≥ f (x) whenever x ∈ Ω and |x∗ − x| ≤ ε. fx (x∗ ) = 0 and fxx is strictly negative deﬁnite.

otherwise let di = 1/k di−1 where k is the smallest positive integer such that f (xi + 1/k di−1 x f (xi )) > f (xi ). Another choice is to let di = di−1 if f (xi + di−1 x f (xi )) > f (xi ). f (˜ + ε x ˜ x x f (˜)) > f (˜) for all ε > 0 sufﬁciently small.11. We can formalize the scheme as an algorithm. one choice is to take di to be an optimal decision for the following problem: Max{f (xi + d x f (xi ))|d > 0. Set i = 0. OPTIMIZATION OVER AN OPEN SET 2. f (xi+1 ) > f (xi ) if xi+1 = xi . The step size di can be selected in many ways. Exercise: Let f be continuous differentiable. Step 3. This requires a one-dimensional search.6 A numerical procedure. At any point x ∈ Ω the gradient x f (˜) is a direction along which f (x) increases.e. For instance. Step 2. Go to Step 2. fx (x∗ ) = 0. (xi + d x f (xi )) ∈ Ω}. Let {di } be produced by either of these choices and let {xi } be the resulting sequence. . i.14 CHAPTER 2. For other numerical procedures the reader is referred to Zangwill [1969] or Polak [1971]. stop. Then 1. To start the process we let d−1 > 0 be arbitrary. Step 1. i 2. Calculate x f (xi ). If x f (xi ) = 0. Set i = i + 1 and return to Step 2. if x∗ ∈ Ω is a limit point of the sequence {xi }. Otherwise let xi+1 = xi + di x f (xi ) and go to Step 3.4.. This observation suggests the following scheme for x x ﬁnding a point x∗ ∈ Ω which satisﬁes 2. Pick x0 ∈ Ω.

y ∗ ). Y ∗ ) = 0. y) for all (x. suppose (x∗ . y) = 4xy subject to (x.1) The problem can be formalized as follows (see Figure 3. y ∗ ) is an optimal decision. and that f1 (x. g(x)) = α whenever |x − x∗ | < ε. Hence. y)|f1 (x. y) in an open set containing (x∗ . Clearly then either x∗ = 0 or y ∗ = 0. y) ∈ V iff f y = g(x). y ∗ ) ≥ f0 (x. and (iii) a differentiable function g : (x∗ − ε. Since Note that y ∗ = 0 implies f1y (x∗ . Additional properties are summarized in Section 3 and a numerical scheme is applied to determine the optimal design of resistive networks. The assertion is false if y ∗ = 0.1): Maximize f0 (x.2). Let us suppose y ∗ = 0. 1 15 .2) The main difference between problem (3.1 Example We want to ﬁnd the rectangle of maximum area inscribed in an ellipse deﬁned by f1 (x. y) = α}.1 it is evident that there exist (i)ε > 0.3) In particular this implies that y ∗ = g(x∗ ). and the properties of its optimal decisions are stated in the form of a theorem. Returning to problem (3. This will generalize to a canonical problem. In the present case let 0 < ε ≤ a − x∗ and g(x) = +b[α − (x/a)2 ]1/2 . (3. (ii) an open set V containing (x∗ . y ∗ ) is an optimal decision we cannot assert that f0 (x∗ .Chapter 3 OPTIMIZATION OVER SETS DEFINED BY EQUALITY CONSTRAINTS We ﬁrst study a simple example and examine the properties of an optimal decision. x∗ + ε) → V such that f1 (x. Then from ﬁgure 3. y) ∈ Ω = {(x. y) = x2 a2 y2 b2 + = α. y) = α and (x. 3.1 (3. if (x∗ . (3.2) and the decisions studied in the last chapter is that the set of permissible decisions Ω is not an open set. y ∗ ). so that this assertion follows from the Implicit Function Theorem.

4) ˆ But the constraint set in (3.6): −1 f0x − f0y f1y f1x = 0 at (x∗ . which we can also express as ˆ so that by Theorem 2. (3. (x∗ . g(x)) ≡ α for |x − x∗ | < ε. Solving these yields x∗ = + 1/2 a − (α/2) . y ∗ )gx (x∗ ) = 0. y∗ = + 1/2 b. g(x)) subject to |x − x∗ | < ε. y ∗ ).5) to obtain the condition (3. OPTIMIZATION WITH EQUALITY CONSTRAINTS Tangent plane to Ω at (x∗ .6). − (α/2) . we see that f1x (x∗ . (3.4): ˆ Maximize f0 (x) = f0 (x. y ∗ ) = α and (3.2). y ∗ ) = 0 we can evaluate gx (x∗ ).1: Illustration of example.16 CHAPTER 3. y ∗ ) + f0y (x∗ . y ∗ ).4) is an open set (in R1 ) and the objective function f0 is differentiable. (3. f1y ) y∗ g(x) V ( x∗ x | ) Ω Figure 3.1. and since f1y (x∗ . y ∗ ) = (x∗ .6) Thus an optimal decision (x∗ .5) and substitute in (3.3. f0x (x f0x (x∗ . it follows that x∗ is an optimal solution for (3. y ∗ ) (f1x . ∗ ) = 0. y ∗ ) must satisfy the two equations f1 (x∗ . y ∗ ) + f1y (x∗ . g(x∗ )) is optimum for (3. −1 gx (x∗ ) = −f1y f1x (x∗ . y ∗ )gx (x∗ ) = 0 Using the fact that f1 (x.

1 Theorem. Then there exists a vector λ∗ = (λ∗ . f0y ) = λ∗ (f1x . are linearly independent. are linearly independent. .12): Maximize f0 (x) subject to fi (x) = αi .7) (3.12) as a function of α = (α1 . . .10) (3. The condition (3. (3. . b).14) Proof. Deﬁne −1 λ∗ = f0y f1y (x∗ .6) and (3. Finally we note that λ∗ = where m(α) = maximum area. . . i = 1. If x∗ (α) is a differentiable function of α then m(α) is a differentiable function of α.2. j ≤ m. . . + λ∗ fmx (x∗ ) m 1 (3. . . . m.13) Furthermore. . i = 1. i = 0.2 General Case 3.12) Suppose that at x∗ the derivatives fix (x∗ ). λ∗ ) such that m 1 f0x (x∗ ) = λ∗ f1x (x∗ ) + . . y ∗ )]λ∗ . αm ) be the maximum value of (3. y ∗ ) In terms of the gradients of f0 .9) is equivalent to f0 (x∗ . .2. let m(α1 . (3.8) can be rewritten as (3. . . . then by re-labeling the coordinates of x if necessary. m. .12). GENERAL CASE Evidently there are two optimal decisions. f1 . m (m < n). y ∗ ) = m(α) = 2αab. (x∗ . Let fi : Rn → R. y ∗ ) = [ f1 (x∗ . . be continuously differentiable functions and let x∗ be an optimal decision of problem (3. . (3. . . (3. − (α/2) 17 and the maximum area is (3. 1. i = 1.3. (ii) an . 1 ≤ i. . .9): (f0x . αm ) . ∂m ∂α . we can assume that the m × m matrix [(∂fi /∂xj )(x∗ )]. . .6) can be interpreted differently. .9) which means that at an optimal decision the gradient of the objective function f0 is normal to the plane tangent to the constraint set Ω. . is nonsingular. m. Let x∗ (α) be an optimal decision for (3. and (λ∗ ) = ∂m ∂α (3.11) 3. y ∗ ). + 1/2 (a. . Since fix (x∗ ). . f1y ) at (x∗ .8) Then (3. By the Implicit Function Theorem (see Fleming [1965]) it follows that there exist (i) ε > 0.

and n m+1 j fi (g(xm+1 . fm ) . . deﬁne the m-dimensional column vector λ∗ by −1 (λ∗ ) = f0w fw |x∗ . m.20) Then (3. . and substitute in (3. . . . In particular this implies that x∗ = gj (x∗ . xm ) . . xn )]| |xm+ − x∗ | < ε. . and since the m × m matrix fw (x∗ ) is nonsingular we can evaluate gu (u∗ ). .18) to obtain the condition −1 −f0w fw fu + f0u = 0 at x∗ = (w∗ . . . (3. xn ) and f = (f1 . = 1. . u) subject to u ∈ U. . u∗ ). . . . . . . . . u = (xm+1 . . . . . it follows that u∗ is an optimal decision for (3. . 1 ≤ j ≤ m. xn ) . 1 ≤ i ≤ m. such that m+ fi (x1 . . fu (x∗ )). and (x1 . Since x = (w. . . this is the same as f0x (x∗ ) = (λ∗ ) fx (x∗ ) = λ∗ f1x (x∗ ) + . f0u (u∗ ) = 0. . . x∗ ). f0u (x∗ )) = (λ∗ ) (fw (x∗ ). where U = [(xm+1 .18 CHAPTER 3.17) ˆ But U is an open subset of Rn−m and f0 is a differentiable function on U (since f0 and g are ˆ differentiable). xn ) = αi . gu (u∗ ) = −[fw (x∗)]−1 fu (x∗ ).21) . u∗ ) is optimal for (3. and (xm+1 . which we can also express using the chain rule for derivatives as ˆ f0u (u∗ ) = f0w (x∗ )gu (u∗ ) + f0u (x∗ ) = 0. . .2). we see that fw (x∗ )gu (u∗ ) + fu (x∗ ) = 0. Then. . + λ∗ fmx (x∗ ). . (3.16) (3. OPTIMIZATION WITH EQUALITY CONSTRAINTS open set V in Rn containing x∗ . . . xn ) ∈ U (see Figure 3.15) For convenience. . . xn ). . . . u∗ ) = (g(u∗ ). 1 ≤ j ≤ m. . m 1 (3. n − m]. . since x∗ = (w∗ . let us deﬁne w = (x1 .19) Next. . xn ).19) and (3.1 . xm+1 . . so that by Theorem 2. .17): ˆ Maximize f0 (u) = f0 (g(u). . xn ) ∈ V iff xj = gj (xm+1 . .18) (3. . .3. . xn ) = αi . (3. Differentiating (3. .12).16) with respect to u = (xm+1 . . .20) can be written as (3. . . i = 1.21): (f0w (x∗ ). u). . (3. . and (iii) a differentiable function g : U → Rm .

.13). it follows that f is x w nonsingular at x∗ (α) in a neighborhood of α.14).22) (3. x m 19 g(xm+1 . m(α) = f0 (x∗ (α)). . . xn ) (x∗ . α . u∗ (α)). To prove (3.23) for α ∈ N . . (xm+1 . . . .24) Differentiating (3. . x∗ ) n m+1 (3. . . .x . Also.2. fw is nonsingular at m 1 m+1 ( ∗ (α). . . . . .22) with respect to α gives ∗ fw wα + fu u∗ = I. GENERAL CASE x1 . . .3. We deﬁne w∗ (α) = (x∗ (α). . . we vary α in a neighborhood of a ﬁxed value. x∗ ) m 1 Ω= {x|f i (x) = αi } i = 1. . . which is equation (3. . . . . By hypothesis. x∗ (α)) and u∗ (α) = (x∗ (α). Since f (x) and x∗ (α) are continuously differentiable by hypothesis. u∗ (α)) = α. so that ∗ mα = f0w wα + f0u u∗ α (3. −1 −f0w fw fu + f0u = 0 at (w∗ (α). . say α. . . . . α so that ∗ −1 −1 wα + fw fu u∗ = fw . We have the equation f (w∗ (α). . xn ) (x∗ . . . m . say N .2: Illustration of theorem. ∗ V xm+1 U xn 2 Figure 3. x∗ α)) .

if we substitute from (3. in a neighborhood of x. We shall use (3. i = 1. i = 1. together with (3. m}. x (3. Then the Implicit Function ˜ ˜ ˜ Theorem enables us to solve. m}. . 3. . . .2 Geometric interpretation. α Using (3. . As u varies. The hypothesis of linear independence of {fix (x∗ )|1 ≤ i ≤ m} guarantees that the tangent plane through Ω at x∗ is described by {h|fix (x∗ )h = 0 .20) and (3. 3.27) ˜ and (3. u) in Ω which is not necessarily optimal. Condition (3.12 deﬁne a n − m dimensional surface Ω = {x|fi (x) = αi .2. w must change according to w = g(u) (in order to ˜ ˆ maintain f (w. . . . .23).3 Algebraic interpretation.26) f1 (x∗ ) + .25) ♦ In (3. i = 1. Suppose that fw (˜) is nonsinx gular at some point x = (w.28) ˆ u and if u is optimal.13). x ˜ x (3. OPTIMIZATION WITH EQUALITY CONSTRAINTS and multiplying on the left by f0w gives ∗ −1 −1 f0w wα + f0w fw fu u∗ = f0w fw . Let us again deﬁne w = (x1 . . The ˆ at u is derivative of f0 ˜ ˆ u ˜ f0u (˜) = f0w gu + f0u˜ = −λ fu (˜) + f0u (˜). x x x where −1 ˜ λ = f0w fw˜ . . the m equations f (w. .20 CHAPTER 3. u) = α). .24). . .25). xm ) and u = (xm+1 . so that the set of (column vectors orthogonal to this tangent surface is {λ1 x (3. + λm x fm (x∗ )|λi ∈ R.12). .28) in the last section. the gradient of the objective function x f0 (x∗ ) is normal to the tangent surface (3. . . . u can then vary ˜ arbitrarily in a neighborhood of u. . The equality constraints of the problem in 3. xn ) .27) ˆ Therefore. . m}. this equation can be rewritten as ∗ −1 f0w wα + f0u u∗ = f0w fw . and the objective function changes according to f0 (u) = f0 (g(u). u) = α.27) is equation (3. u f0 (˜) = 0 which. α (3.2. . the direction of steepest increase of f0 at u is ˜ u u f0 (˜) ˆ = −fu (˜)λ + fOu (˜) .13) is therefore equivalent to saying that at an optimal decision x∗ . we obtain (3. u).14) and the theorem is proved. .

. ∗ is a local optimum.2 An alternative condition.. 0 ≤ i ≤ m. Furthermore. Let x∗ be optimal for (3. 1 ≤ i ≤ m.12) and (3.1. Then there exists λ∗ ∈ Rm such that (x∗ . ˆ Since we can convert the problem (3.3.3. L(x) = f0 (x) − i=1 λ∗ fi (x).. However.13) need not hold if the derivatives fix (x∗ ).16) in a neighborhood of x∗ . are not linearly independent. i. and suppose that fix (x∗ ). it is useful to translate these the comments of Section 2. and its proof is left as i=1 an exercise.3.I] where m Lww Lwu Luw Luu  gu  . so is g (see Fleming [1965]). This can be checked in the following example Minimize subject to sin(x2 + x2 ) 1 2 π 2 2 2 (x1 + x2 ) = 1. Lx (x∗ .12) into a problem of maximizing f0 over an open set. all ˆ . and a necessary condition for x∗ to be optimal for (3.1 The condition of linear independence.2. 3. The following is a reformulation of 3. REMARKS AND EXTENSIONS 21 3. 1 ≤ i ≤ m.12). The necessary condition (3. u∗ ) gu (u∗ ) = −[fw (x∗ )]−1 fu (x∗ ). Exercise: Show that  .12.3 Second-order conditions. λ∗ ) = 0.3 Remarks and Extensions 3. are twice continuously ˆ differentiable.. λ) → f0 (x) − m λi fi (x). if f is twice differentiable. i . deﬁne the Lagrangian function L : Rn+m → R by L : (x. the following exercise expresses if this matrix is negative deﬁnite then x ˆ f f0uu (u∗ ) in terms of derivatives of the functions fi . then so is f0 .4 will apply to the function f0 remarks in terms of the original function f0 and f . Keeping the notation of Theorem 3. Furthermore.3. ˆ f0uu (u∗ ) = [gu .e. It follows that if the functions fi . This is possible because the function g is uniquely speciﬁed by (3.  I (w∗ .3.29) 3. λ∗ ) is a stationary point of L. λ∗ ) = 0 and Lλ (x∗ . (3.13) and ˆ the condition that the (n − m) × (n − m) matrix f0uu (u∗ ) is negative semi-deﬁnite. are linearly independent.

We assume that the derivatives fix (x). uk ) = α. Set uk = uk + dk f0 (uk ). Then the following algorithm is a straightforward adaptation of the procedure in Section 2. . denote the vectors of branch voltages and branch currents. ˆk Step 3. . x ˜ ˜ Remarks. . . and f0 (uk ) = −fu (xk )λk + f0u (xk ). The practical applicability of the algorithm depends upon two crucial factors: the ease with which we can ﬁnd a partition x = (w. As before. . p ) which are under our control.3. 2 (3.4 A numerical procedure.32) Although (3. vs ∈ Rb for the sources.31): j − js = g(v − vs ) = g(vr ). uk ) = 0. . . no essential simpliﬁcation is achieved. We choose one of the nodes as datum and denote by e = (e1 . . so that jk − jsk = gk (vrk ) = gk (vk − vsk ). let us suppose that there are design parameters p = (p1 .32): j − jx = g(vr . vb ) and j = (j1 . Otherwise go to Step 3. vr ∈ Rb for the resistor voltages. Find x0 arbitrary so that fi (x0 ) = αi . vsk are the source current and voltage in the kth branch. so that (3. set k = k + 1. The w variable may consist of any m components of x. .31) is replaced by (3. . . Orient the network graph and let v = (v1 . .30) as (3. 1 ≤ i ≤ m. gb ) . Here jsk . .30) implies that the current (jk −j s k) through the kth resistor depends only on the voltage vrk = (vk −v sk ) across itself. .4. stop. Let A be the n × b reduced incidence matrix of the network graph. In the next section we apply this algorithm to a practical ˜ ˜ ˜ problem where these two steps can be carried out without too much difﬁculty. in (3.31) where vrk is the voltage across the resistor. Set k = 0 and go to Step 2. thus enabling us to calculate λk . p) = g(v−v s . Step 1. Find wk such that fi (wk . u)2 of the variables such that fw (xk ) is nonsingular. 1 ≤ i ≤ m.30) Next we suppose that each branch k contains a (possibly nonlinear)resistive element with the form shown in Figure 3.22 CHAPTER 3. Furthermore. jb ) respectively. If f0 (uk ) = 0. . p). Step 2. and return to Step 2. 3. Calculate λk −1 ˆk ˆk by (λk ) = f0w fw(xk). are linearly independent for all x. . and g = (g1 . Find a partition x = (w. . u) so that fw (xk ) is nonsingular. the step sizes dk > 0 can be selected various ways. and the ease with which we can ﬁnd wk so that f (wk .33) This is just a notational convenience. 1 ≤ i ≤ m.3. we can rewrite (3. Hence.5 Design of resistive networks. Set ˜ ˜ ˜ ˜ k+1 = (wk . Using the obvious vector notation js ∈ Rb . . en ) the vector of node-to-datum voltages.3.31) we shall assume that gk is a function of vr . (3. .6. 1 ≤ k ≤ b. uk ). and gk is the characteristic of the resistor. . . Then the Kirchhoff current and voltage laws respectively yield the equations Aj = 0 and A e = v (3. OPTIMIZATION WITH EQUALITY CONSTRAINTS 3. This allows us to include coupled resistors and voltagecontrolled current sources. (3. Consider a network N with n + 1 nodes and b branches.

x −1 x x −1 x From the deﬁnition of f we have fe (˜) = AG(˜r . REMARKS AND EXTENSIONS jsk jk − jsk jk o 23 + + vsk - vrk + - o vk - Figure 3. If we compare (3. p) is called the adjoint network (of N ) at (˜r .33) ˜ with (3. If we combine (3. if we let x = (e.4 to this problem. then assumption (b) allows us to identify w = e. p)) v ˜ v ˜ of the original network N .27)) ﬁxed point. p)A is nonsingular for all v ∈ Rb . is ) subject to Ag(A e − vs . moreover. p). Assumption: (a) f0 is differentiable. p). p) = (∂g/∂vr )(˜r .29) and (3. vs .34) where we have deﬁned is = Ajs . is so as to minimize some speciﬁed function f0 (e. vs . Furthermore. (b) g is differentiable and the n×n matrix A(∂g/∂v)(v. In terms of the notation of 3. p.33).36) we see immediately that λ is the node-to-datum response voltages of a linear network N (˜r . Formally. vs . p) driven by the current sources f0e (˜). To this end let x = (˜.37) (3.. is ).3. vs . p. for every value of (p. is ) = Ag(A e−v s . G (˜r . v ˜ v ˜ . Then the corresponding λ = λ ˜ λ = f0w (˜)fw (˜) = f0e (˜)fe (˜). p) − is .3.36) has the following extremely interesting physical interpretation.3: The kth branch. Also let f (x) = f (e. is ).33): Ag(A e − vs .33) is determinate i. x v ˜ ˜ where vr = A e − vs . λ is the solution (unique by ˜ ˜ ˜ v ˜ v ˜ assumption (b)) of the following linear equation: AG (˜r . vs . p. v ˜ ˜ x (3. (3.3. is the transpose of the incremental branch admittance matrix (evaluated at (˜r . p)A λ = f0e (˜). ˜s ) be a ˜ e ˜ ˜ i ˜ is given by (see (3. (3. p. and G(˜r . this network has the same graph as v ˜ x the original network (since they have the same incidence matrix).36) Now (3. (c) The network N described by (3. To do this we make the following assumption. is ) satisfying (3. p. vs .32) we obtain (3.35) We shall apply the algorithm 3. p)A .4. we have the optimization problem (3. The network design problem can then be stated as ﬁnding p. Now the crucial part in the algorithm is to obtain λk at some point xk . For this reason. p ∈ R .3. is ) there is a unique e = E(p. p) = is .e. and u = (p. vs . N (˜r . vs . p) − is = 0. p). Therefore. vs .34): Minimize f0 (e. its branch admittance matrix. is ).

(ii) How do the formulas change if the network equations are written using an arbitrary cutset matrix instead of the incidence matrix? . t (i) Specialize the algorithm above for this particular case. REMARKS AND EXTENSIONS f0 (e) = t∈T 25 αt |vt −v d |2 .3.3.

OPTIMIZATION WITH EQUALITY CONSTRAINTS .26 CHAPTER 3.

In the second section we present the duality theory for linear programming and use it to obtain some sensitivity results. For a detailed and readily accessible treatment of the material presented in this chapter see the companion volume in this Series (Sakarovitch [1971]). Additional miscellaneous comments are collected in the last section. 27 . and then we deﬁne the general linear programming problem. On the supply side of our accounting. 4.1. 000 and 5g+7u 40 ≤ 5250 or 5g + 7u ≤ 210.1 The Linear Programming Problem 4. and the number of 20 lecture courses demanded per year is 5g+7u . Recall Example 2 of Chapter I. the President must satisfy 2g+u 20 ≤ 2250 or 2g + u ≤ 45. Because of his contractual agreements. Let g and u respectively be the number of graduate and undergraduate students admitted. the faculty can 40 offer 2(750) + 3(250) = 2250 seminars and 6(750) + 3(250) = 5250 lecture courses.1 Example.Chapter 4 OPTIMIZATION OVER SETS DEFINED BY INEQUALITY CONSTRAINTS: LINEAR PROGRAMMING In the ﬁrst section we study in detail Example 2 of Chapter I. Then the number of seminars demanded per year is 2g+u . 000 . In section 4 we apply the results of Sections 2 and 3 to study the linear programming theory of competitive economy. In Section 3 we present the Simplex algorithm which is the main procedure used to solve linear programming problems.

β) . LINEAR PROGRAMMING Since negative g or u is meaningless.3) as an exercise. Then the set Ω of all vectors x which satisfy the constraints in (4. Therefore.2.2) is given by Ω = {x|Ai x ≤ bi . . A=  −1 0  0 −1 Then (4.28 CHAPTER 4. the surface of constant payoff k say. Since c x∗ ≥ c y for all y ∈ K ∗ we conclude that c h ≤ 0 for all h such that A1 h ≤ 0. u) . 0. From this condition we can immediately draw two very important conclusions: (i) at least one of the vertices of Ω is an optimal decision. Now x∗ satisﬁes Ax x∗ = b1 . (4. A2 x∗ = b2 .1 we can see that x∗ = Q is an optimal decision and the cone K ∗ is shown in Figure 4. and A3 x∗ < b3 . the President receives the payoff c x.2) Let Ai . u ≥ 0. since K ∗ lies “below” π ∗ .1) can be rewritten as (4. We can rephrase this by saying that x∗ ∈ Ω is an optimal decision if and only if the plane π ∗ through x∗ does not intersect the interior of Ω. denote the rows of A. 210000. The ﬁrst conclusion is the foundation of the powerful Simplex algorithm which we present in Section 3. 0) and let A be the 4×2 matrix   2 1  5 7  .) Evidently an optimal decision is any point x∗ ∈ Ω which lies on a hyperplane π(k) which is farthest along the direction c. These hyperplanes for different values of k are parallel to one another since they have the same normal c. u ≥ 0 . Formally then the President faces the following decision problem: Maximize αg + βu subject to 2g + u ≤ 45. For each choice x. so that K ∗ is given by K ∗ = {x∗ + h|A1 h ≤ 0 . A2 h ≤ 0} . For the situation depicted in Figure 4. b = (45000. Furthermore. 000 5g + 7u ≤ 210. (4. c = (α. is the hyperplane π(k) = {x|c x = k}. and (ii) x∗ yields a higher payoff than all points in the cone K ∗ consisting of all rays starting at x∗ and passing through Ω.1. 1 ≤ i ≤ 4} and is the polygon OP QR in Figure 4. (Obviously we are assuming in this discussion that c = 0. We pause to formulate the generalization of (4. Here we pursue consequences of the second conclusion.1) It is convenient to use a more general notation. So let x = (g.2.1.3) Recall the notation introduced in 1. A4 x∗ < b4 . so that x ≤ y means xi ≤ yi for all i. 000 g ≥ 0. as k increases π(k) moves in the direction c. there are also the constraints g ≥ 0. 1 ≤ i ≤ 4. and futhermore at x∗ the direction c points away from Ω. 1 (4.2)1 Maximize c x subject to Ax ≤ b . A2 h ≤ 0 .

i ∈ I(x). be n-dimensional row vectors. THE LINEAR PROGRAMMING PROBLEM x2 29 - π∗ - π(k) = {x|c x = k} direction of increasing payoff k - P - Q = x∗ - c ⊥ π∗ A2 ⊥ P Q A1 ⊥ QR A3 .1.1: Ω = OP QR. its generalization to n dimensions is a deep theorem known as Farkas’ lemma (see Section 2). . 2 .1): Although this statement is intuitively obvious. For any x satisfying the constraints. 1 ≤ i ≤ k. let I(x) ⊂ {1. . Show that x∗ is optimal if an only if / c h ≤ 0 for all h such that Ai h ≤ 0 . Consider the problem Maximize c x subject to Ai x ≤ bi . We can see from our analysis that the situation is as follows (see Figure 4. λ∗ ≥ 0 such that 1 2 c = λ∗ . . 1 ≤ i ≤ k. O A4 R {x|A1 x = b1 } Figure 4. it is clear that (4. i ∈ I(x). i ∈ I(x∗ ). Let c ∈ Rn . the optimal decision will change. Mathematically this means that (4.3) is satisﬁed if and only if there exist λ∗ ≥ 0.3) is satisﬁed as long as c lies between A1 and A2 . 1 ≤ i ≤ k . be real numbers. 1 2 (4. Returning to our problem. Suppose x∗ satisﬁes the constraints. and let bi . Exercise 1: Let Ai . Ai x < bi . A1 + λ∗ A2 .4) - - {x|A2 x = b2 } - x1 2 As c varies. n} be such that Ai (x) = bi .4. .

5) For purposes of application it is useful to separate those constraints which are of the form xi ≥ 0. (b) if Ai x∗ < bi then λ∗ = 0 . 2 2 3. 2. LINEAR PROGRAMMING P x∗ = Q K∗ A2 c A1 A3 O R π∗ A4 Figure 4. 2. x∗ ∈ QP is optimal iff c lies along A2 iff c = λ∗ A2 for some λ∗ ≥ 0. i = 1.) x∗ ∈ Ω is optimal iff there exist λ∗ ≥ 0 . ai2 ).6) . 2 1i i (4. j = 1. 2. 2. 1 2 j (c) if ci < λ∗ + λ∗ a2i then x∗ = 0.5) accordingly We leave this as an exercise.30 CHAPTER 4. These statements can be made in a more elegant way as follows: x∗ ∈ Ω is optimal iff there exists λ∗ ≥ 0 . 1 ≤ i ≤ 4. x∗ = Q is optimal iff c lies between A1 and A2 iff c = λ∗ A1 + λ∗ A2 for some λ∗ ≥ 0.6). i i (4. 1. x∗ = P is optimal iff c lies between A3 and A2 iff c = λ∗ A2 + λ∗ A3 for some λ∗ ≥ 0. such that i 4 (a) c = i=1 λ∗ ai . 1 2 (b) if aj1 x∗ + aj2 x∗ < bj then x∗ = 0.5) is equivalent to (4. (Here Ai = (ai1 .2: K ∗ is the cone generated by Ω at x∗ . i = 1. λ∗ ≥ 0 such that 1 2 (a) ci ≤ λ∗ a1i + λ∗ a2i . from the rest. and to reformulate (4. Exercise 2: Show that (4. λ∗ ≥ 1 2 1 2 0. below. etc. λ∗ ≥ 2 3 2 3 0.

where the cj . There are two important special cases: Case I: (4. . j=1 subject to 1≤i≤m. l ≤ i ≤ k . . + ain xn = bi . . . . . THE LINEAR PROGRAMMING PROBLEM 31 4. xj ≥ 0 .1. . . . . . . k + 1 ≤ i ≤ . ail x1 + .7.2 Problem formulation. . Step 3: Replace each variable xj which is constrained xj ≤ 0 by a variable yj = −xj constrained yj ≥ 0 and then replace aij xj by (−aij )yj for every i and cj xj by (−cj )yj . . 1≤j≤n. .8). (4. such is not the case.8): n (4. Proposition: Every LP of the form (4. Maximize c1 x1 + c2 x2 + . p + 1 ≤ j ≤ q. A linear programming problem (or LP in brief) is any decision problem of the form 4.7) appears to be more general than (4. + ain xn ≤ bi . Step 1: Replace each inequality constraint aij xj ≥ bi by (−aij )xj ≤ (−bi ).7) Maximize j=1 n cj xj aij xj ≤ bi . .4. .1.7) is of the form (4. . bi are ﬁxed real numbers. 1≤j≤p. Proof.9). . . .8) xj ≥ 0 Case II: (4. ail x1 + . xj arbitary .9) xj ≥ 0 Although (4.8) and (4.7) is of the form (4. j=1 subject to 1≤i≤m. (−aij )xj ≤ (−bi ). 1≤j≤n (4. + cn xn subject to ail x1 + ai2 x2 + . and xj ≥ 0 . aij . + 1 ≤ i ≤ m . . + ain xn ≥ bi .9): n Maximize j=1 n cj xj aij xj = bi . . q + 1 ≤ j ≤ n .7) can be transformed into an equivalent LP of the form (4. Step 2: Replace each equality constraint aij xj = bi by two inequality constraints: aij xj ≤ bi .

Let Ai . .1). Ax ≤ 0 implies c x ≤ 0. . (The new variables added in these steps are called slack variables. 1 ≤ j ≤ m. For a proof the reader is referred to (Mangasarian [1969]). An algebraic version of this result is sometimes more convenient.7) can be transformed into an equivalent LP of the from (4. We begin by quoting a fundamental result. . Use the previous exercise to show that x∗ is optimal iff there exist λ∗ ≥ 0. 1 ≤ i ≤ k.11) . ♦ Proposition: Every LP of the form (4.32 CHAPTER 4. . Let c ∈ Rn . . Exercise 1: With the same hypothesis and notation of Exercise 1 in 4. Using this result it is possible to derive the main results following the intuitive reasoning of (4.10) and (4. Step 4: Repeat these steps from the previous proposition.9) and is equivalent to the original one. The following statements are equivalent: (i) for all x ∈ Rn . Evidently the resulting LP has the form (4. whereas x ∈ Rn and λ ∈ Rm will be variable. LINEAR PROGRAMMING Step 4: Replace each variable xj which is not constrained in sign by a pair of variables yj −z j = xj constrained yj ≥ 0. . b ∈n are ﬁxed vectors. 1 ≤ j ≤ n i m i=1 (b) if j=1 aij x∗ < bi then λ∗ = 0 . 1 ≤ i ≤ m (c) if j i λ∗ aij > cj then x∗ = 0 .9) Proof.1. λ∗ ≥ 0 such that m 1 m i∈I(x∗ ) (a) cj ≤ n i=1 λ∗ aij . . λ ∈ Rk . We leave this development as two exercises and follow a more elegant but less intuitive approach. Evidently the new LP has the form (4. k (ii) there exists λ1 ≥ 0. be n-dimensional row vectors.) Step 3.17). Farkas’ Lemma. Let c ∈ Rn be a column vector. zj ≥ 0 and then replace aij xj by aij yj + (−aij )zj for every i and cj xj by cj yj + (−cj )zj .1 Main results. use the ﬁrst version of Farkas lemma to show that there exist λ∗ ≥ 0 for i ∈ I(x∗ ) such that λ∗ Ai = c . Farkas’ Lemma (algebraic version). (ii) there exists λ ≥ 0. Step 1: Replace each inequality constraint aij xj ≤ bi by the equality constraint aij xj + yi = bi where yi is an additional variable constrained yi ≥ 0.2. Step 2: Replace each inequality constraint aij xj ≥ bi by the equality constraint aij xj − yi = bi where yi is an additional variable constrained by yi ≥ 0. i i Exercise 2: Let x∗ satisfy the constraints for problem (4. ♦ 4. .8) and is equivalent to the original one. Let A be a k × n matrix. Ai x ≤ 0 for 1 ≤ i ≤ k implies c x ≤ 0. c ∈ Rn . such that A λ = c. Consider the pair of LPs (4. and A = {aij } is a ﬁxed m × n matrix.2 Qualitative Theory of Linear Programming 4. λk ≥ 0 such that c = i=1 λi Ai . i j In the remaining discussion. (i) for all x ∈ Rn . The following statements are equivalent.

x ≥ 0} be the set of all points satisfying the constraints of the primal problem. (4. then x∗ is optimal for (4. + λm amj ≥ cj . this is equivalent to the existence of x ≥ 0. QUALITATIVE THEORY OF LINEAR PROGRAMMING below. r ∈ R. λ ≥ 0. ♦ Corollary 1: If x∗ ∈ Ω and λ∗ ∈ Ωd such that c x∗ = (λ∗ ) b. . 33 (4. . Then there exists x∗ which is optimum for (4. i. 1 ≤ j ≤ n λi ≥ 0 . . 1 ≤ j ≤ n . y ≥ 0. Furthermore. . Aw = bθ ≤ 0 . λ ≥ 0}. (4. + λm bm subject to λ1 a1j + . By introducing slack variables y ∈ Rm .11).10) and λ∗ is optimal for (4. 1 ≤ i ≤ m . Theorem 1: (Strong duality) Suppose Ωp = φ and Ωd = φ. . A λ ≥ c and b λ−c x ≤ 0.10) is called the primal problem and (4. + cn xn subject to ai1 x1 + . −w ≤ 0 .10) and λ∗ which is optimum for (4. Maximize c1 x1 + . .10) Maximize λ1 b1 + .12) Proof: x ≥ 0 and λ A − c ≥ 0 implies (λ A−c )x ≥ 0 giving the ﬁrst inequality.e. µ ≤ 0. c x∗ = (λ∗ ) b. A point x ∈ Ωp (λ ∈ Ωd ) is said to be a feasible solution or feasible decision for the primal (dual).11) Deﬁnition: Let Ωp = {x ∈ Rn |Ax ≤ b. + ain xn ≤ bi . λ ≥ 0. Similarly let Ωd = {λ ∈ Rm |λ A ≥ c . Lemma 1: (Weak duality) Let x ∈ Ωp . b−Ax ≥ 0 and λ ≥ 0 implies λ (b−Ax) ≥ 0 giving the second inequality. 1 ≤ i ≤ m xj ≥ 0 . such that Ax ≤ b. . (4. Proof: Because of the Corollary 1 it is enough to prove the last statement. we must show that there exist x ≥ 0.11) is called the dual problem.4. λ ∈ Ωd .11)..14) (4. θ≤0 implies b ξ + c w ≤ 0.2. (4. µ ∈ Rm . The next result is trivial. . r ≤ 0 such that   A −c Im A b −I n     1    x y λ µ r     b  = c    0 By the algebraic version of Farkas’ Lemma.13) . Then c x ≤ λ Ax ≤ λ b. this is possible only if A ξ − cθ ≤ 0 . ξ ≤ 0 .

µ ≤ 0 such that   λ | A −In  − − −  = c | µ By Farkas’ Lemma there exists w ∈ Rn such that Aw ≤ 0. Proof Because of the symmetry of the primal and dual it is enough to prove only (i). By Lemma 1 we always have c x∗ ≤ (λ∗ ) Ax∗ ≤ (λ∗ ) b so that we must have c x∗ = (λ∗ ) Ax∗ = (λ∗ ) b. and c w ≤ (A λ) w = λ (Aw) ≤ 0. (ii) Suppose Ωd = φ. So that b ξ + c w ≤ 0. ♦ .16): (λ∗ ) (Ax∗ − b) = 0. ξ. A(x + θw) ≤ b. ξ. so that c x∗ = (λ∗ ) b. By hypothesis. there does not exist λ ≥ 0. −b ξ = b (−ξ) ≥ (Ax) (−ξ) = x (−A ξ) ≥ 0. so that (x + θw) ∈ Ωp .15) is known as the condition of complementary slackness. so that −A ξ ≥ 0. ♦ Remark: In Theorem 2(i). θ) satisﬁes (4. By hypothesis. λ∗ ∈ Ωd . the hypothesis that Ωp = φ is essential. Hence.15) λ∗ aij < cj implies x∗ = 0 . but then for any θ > 0. Exercise 3: Exhibit a pair of primal and dual problems such that neither has a feasible solution. Suppose x∗ ∈ Ωp is optimal. But (4. Then (ξ/θ) ∈ Ωd . Consider the following exercise. so there exists x ≥ 0 such that Ax ≤ b. and c w > 0.13) and θ < 0. LINEAR PROGRAMMING Case (i): Suppose (w. and (A λ∗ − c) x∗ = 0 . that Ωd = φ.34 CHAPTER 4. Ωd = φ. The ﬁrst equality in (4. Equivalently. Then there exists an optimum decision for the dual LP iff Ωp = φ. j i (4. so that by Theorem 1 there exists λ∗ ∈ Ωd such that c x∗ = (λ∗ ) b. Evidently then. Now. so that by Lemma 1 c w/(−θ) ≤ b ξ/θ. Aw ≤ 0. c (x + θw) = c x + θc w. Then there exists an optimum decision for the primal LP iff Ωd = φ. Also. Theorem 2: (i) Suppose Ωp = φ. which is equivalent to (4. λ∗ ∈ Ωd . while the second yields (A λ∗ ) x∗ = c x∗ . We will show that sup {c x|x ∈ Ωp } = +∞. x∗ is optimal. θ) satisﬁes (4.14) since θ < 0. sup {c x|x ∈ Ωp } = +∞ so that there is no optimal decision for the primal. Suppose (4. Theorem 3: (Optimality condition) x∗ ∈ Ωp is optimal if and only if there exists λ∗ ∈ Ωd such that m j=1 aij x∗ < bi implies λ∗ = 0 . (w/−θ) ∈ Ωp . i j and m i=1 ((4. Suppose. −ξ ≥ 0. By Corollary 1. Then from Theorem 2. Ωd = φ means there does not exist λ ≥ 0 such that A λ ≥ c. (x + θw) ≥ 0. Ωp = φ. ♦ The existence part of the above result can be strengthened. Sufﬁciency.15) is equivalent to (4.16) holds for some x∗ ∈ Ωp . The sufﬁciency part of (i) follows from Theorem 1. λ ∈ Ωd . so that only the necessity remains. Case (ii): Suppose (w.16) Necessity.) Proof: First of all we note that for x∗ ∈ Ωp . (4.16) is just an equivalent rearrangement of these two equalities. w ≥ 0.13) and θ = 0. (4. there exist x ∈ Ωp . in contradiction.16) yields (λ∗ ) b = (λ∗ ) Ax∗ = (A λ∗ ) x∗ . −w ≤ 0.

.20) respectively. λ ∈ Rm .19) . Ωd denote the set of all x.9). Apply Theorems 1 and 2. This is now of the form (4. λ ≥ 0}.4. (−A)x ≤ (−b). λ satisfying the constraints of (4. λ∗ ) satisfying (4.) Exercise 8: Formulate a dual for (4. Replace (4. . Remark.18) (4. Again (4. 4.19) is called the primal and (4. λ∗ ) ≤ L(x∗ . We let Ωp . λ∗ ≥ 0 provided we strengthen (4. λ) = c x − λ (Ax − b) (4.17) is said to form a saddle-point of L over the set {x|x ∈ Rn . The function L is called the Lagrangian. QUALITATIVE THEORY OF LINEAR PROGRAMMING 35 The conditions x∗ ∈ Ωp . Minimize λ1 b1 + . λ is not restricted in sign. . We begin with a pair of LPs: Maximize c1 x1 + . . (Note that.18). Exercise 5: Prove Theorems 1 and 2 with Ωp and Ωd interpreted as above. (4.) Exercise 6: Show that x∗ ∈ Ωp is optimal iff there exists λ∗ ∈ Ωd such that m x∗ j > 0 implies i=1 λ∗ aij = cj . .9).2. 1 ≤ i ≤ m .17) Exercise 4: Prove Theorem 4. λ) for all x ≥ 0.7).10). λ∗ ) ≤ L(x∗ . 1≤j≤n . We state these results as exercises. i Exercise 7: x∗ ≥ 0 is optimal iff there exists λ∗ ∈ Rm such that L(x. . x∗ ∈ Ωd in Theorem 3 can be replaced by the weaker x∗ ≥ 0. subject to Ax ≤ b.2.20) the dual.15) as in the following result. where L is deﬁned in (4. (4. 1 ≤ j ≤ n . (Hint.2 Results for problem (4.20) Note that in (4. λ∗ ) ≤ L(x∗ . x ≥ 0. + ain xn = bi . . . It is possible to derive analogous results for LPs of the form (4. + λm amj ≥ cj (4. xj ≥ 0 . unlike (4. .17). A pair (x∗ . λ∗ ) ≤ L(x∗ . where L: Rn xRm → R is deﬁned by L(x. + cn xn subject to ail x1 + . indicating how to use the results already obtained. whose proof is left as an exercise. and allλ ≥ 0. Theorem 4: (Saddle point) x∗ ≥ 0 is optimal for the primal if and only if there exists λ∗ ≥ 0 such that L(x. x ≥ 0} × {λ|λ ∈ Rm .19).20) the λi are unrestricted in sign. λ) for all x ≥ 0. and obtain the result analogous to Exercise 5.19) by the equivalent LP: maximize c x. + λm bm subject to λ1 a1j + .

b(i. cj + ε. The matrix A will remain ﬁxed. ˆ c ∂M + ˆ ˆ ∂bi (b.22) .20). . LINEAR PROGRAMMING 4.11) or for the pair (4. if b. . . cj+1 .10) or (4. c) − M (ˆ −ε). . c) = lim ε→0 ε>0 1 ˆ ˆ ε {M (b. ε) = (b1 .2. ◦ ◦ Theorem 5: At each (ˆ c) ∈B × C .19) and (4. Let B = {b ∈ Rm |Ωp (b) = φ} and C = {c ∈ Rn |Ωd (c) = φ}. the partial derivatives given above exist.10) and (4. . c) (4. (4. . . 1≤i≤m. bm ) . ε ∈ R. bi−1 . We write Ωp (b) and Ωd (c) to denote the explicit dependence on b and c respectively. C denote the interiors of B. c ∈ Rn denote c(j.36 CHAPTER 4. . . bi + ε. . and for (b. We investigate how the maximum value of (4. ˆ Let B . c) = lim ε→0 ε>0 1 ˆ ˆ ε {M (b. b.19) changes as the vectors b and c change. ε) = (c1 . ˆ as follows: ∂M + ˆ ˆ ∂bi (b. b ∈ Rm denote b(i. ˆ ∂M − ˆ ˆ ∂bi (b. Furthermore. ε)) − M (ˆ c} . . b. and for 1 ≤ j ≤ n.21) = lim ε→0 ε>0 1 ˆ ˆ ε {M (b(i. ˆ x ∈ Ωp (ˆ λ ∈ Ωd (ˆ) are optimal. ˆ ∂M − ˆ ˆ ∂cj (b. . We deﬁne in the usual way the right and left hand partial derivatives of M at a point (ˆ c) ∈ B × C b. c − M (ˆ c(j. C respectively. c) ∈ B × C deﬁne M (b. c) = max {c x|x ∈ Ωp (b)} = min {λ b|λ ∈ Ωd (c)} . c) = lim ε→0 ε>0 1 ˆ ˆ ε {M (b. b. c(j. For 1 ≤ i ≤ m. . ε). . cj−1 . ε ∈ R. c ) − M (ˆ c)} . cn ) . c) ◦ ◦ ˆ ≤ λi ≤ ∂M − ˆ ˆ ∂bi (b. c2 . . Let Ωp and Ωd be the sets of feasible solutions for the pair (4. −ε))} . b2 . . c)} . c) . bi+1 .3 Sensitivity analysis. ˆ ∂M + ˆ ˆ ∂cj (b. then ˆ b).

−ε)} x = xj . 9 below.3. ˆ ≤ ≥ 1ˆ ˆ ˆ ˆ ε λ {b(i.e. . ˆ ε λ {b − b(i. c)} b(i. so that b. ·) are piecewise linear (more accurately. C ⊂ Rn deﬁned above are convex sets.4. c) .23) as ε → 0.22).3 The Simplex Algorithm 4. 1≤j≤n. 1 ≤ i ≤ m xj ≥ 0 . so that b. M (b. Finally. 1 ≤ j ≤ n . c) ≥ xj ≥ ˆ ∂M − ˆ ˆ ∂cj (b. ˆ b(i. ·) : C → R is convex. c) ≤ λ ˆ ε). Ωd . i. (Hint: First show that for ε2 > ε1 > 0 > δ1 > δ2 . (4. {M (b. > 0. c) and M (b.23) assuming that the partial derivatives exist. (b) Show that for ﬁxed c ∈ C. 0 ≤ θ ≤ 1 implies f (θx + (1 − θ)y) ≤ θf (x) + (1 − θ)f (y). Then the result follows immediately.23) Proof: We ﬁrst show (4. Deﬁnition: X ⊂ Rn is said to be convex if x.) x x x x Remark 1: Clearly if (∂M/∂bi )(ˆ exists.1 Preliminaries We now present the celebrated Simplex algorithm for ﬁnding an optimum solution to any LP of the form (4. . for ε > 0. On the other hand. ˆ 1 ˆ ˆ ˆ ˆ ε {M (b(i. ε)) ≥ (ˆ(j. ˆ c ˆ ˆ ˆ ε ε which give (4. Remark 2: We can also show without difﬁculty that M (·. y ∈ X. ˆ c ˆ 1 1 ˆ ˆ ˆ ˆ c ˆ ˆ ˆ ε {M (b. c(j. Exercise 8: (a) Show that Ωp . and then this result b) compares with 3. y ∈ X. c) : B → R is concave and for ﬁxed b ∈ B. for ε > 0. ε). ˆ b. + cn xn subject to ail x1 + . ε2 )) − f (ˆ)} ≥ (1/ε1 ){f (ˆ(i. ε) − b}λi . gives (4. c)} ≥ ε {ˆ(j. THE SIMPLEX ALGORITHM 37 ∂M + ˆ ˆ ∂cj (b. ε > 0. Remark 3: The variables of the dual problem are called Lagrange variables or dual variables or shadow-prices. c)} 1 ˆ c) − M (ˆ −ε). ε)) − M (b. Taking limits as ε → 0. M (·.. M (ˆ c) = c x. y ∈ X and 0 ≤ θ ≤ 1 implies (θx + (1 − θ)y) ∈ X. the left and right hand partial derivatives of f exist.24) . (4.22). ♦ We recall some fundamental deﬁnitions from convex analysis. δ1 )) − f (ˆ))} ≥ (1/δ2 ){f (ˆ(i. for ε 1ˆ ˆ ˆ −ε)} = λi . 4. . 1 ˆ c) − M (ˆ c(j. (ii) f is said to be concave if −f is convex. Show that at each point x in the interior of ˆ X. M (ˆ c) = λ ˆ and by weak duality M (ˆ ε). The reason behind the last name will be clear in Section 4. ˆ ε {M (b. ε > 0. x. Exercise 9: Let X ⊂ Rn .22). then we have equality in (4. δ2 )) − f (ˆ)}. . ε1 )) − f (ˆ))} ≥ x x x x (1/δ1 ){f (ˆ(i.24): Maximize c1 x1 + . c ) − M (b. −ε))} ≤ 1 {ˆ − c(j. (4. Deﬁnition: Let X ⊂ Rn and f : X → R.(1/ε2 ){f (ˆ(i. the existence of the right and left partial derivatives follows from Exercises 8. ˆ ˆˆ b. for ε > 0. (i) f is said to be convex if X is convex. and the sets B ⊂ Rm .14). and f : X → R be convex. and M (ˆ c(j. + ain xn = bi . This is useful in some computational problems. ˆ b(i. 0 ≤ θ ≤ 1 implies f (θx + (1 − θ)y) ≥ θf (x) + (1 − θ)f (y). and x. By strong duality ˆ b. ε) − c} x = xj .3. linear plus constant) functions on B and C respectively. ε)) x.

it follows that z(θ) ≥ 0 when j |θ| ≤ min x∗ j |γj | j ∈ I(x∗ ) = θ ∗ say . Hence z(θ) ∈ Ωp whenever |θ| ≤ θ ∗ .1 the algorithm rests upon the observations that if an optimal exists. cj γj = J∈I(x∗ ) Since θ can take on positive and negative values. such that γj Aj = 0 . m n! Corollary 1: Ωp has at most vertices. not all zero. the inequality above can hold on if 0. . Hence suppose {Aj |j ∈ I(x∗ )} is linearly dependent so that there exist γj . j∈I(x∗ ) j=1 For θ ∈ R deﬁne z(θ) ∈ Rn by zj (θ) = x∗ = θγj . j ∈ I(x∗ ) j x∗ = 0 . Deﬁnition: For x ∈ Ωp . Since x∗ is optimal we must have c x∗ ≥ c z(θ) = c x∗ + θ j∈I(x∗ ) cj yj for −∗ θ ≤ θ ≤ θ ∗ . z in Ωp and 0 < λ < 1. i. and then c x∗ = c z(θ). . We begin with a precise deﬁnition of a vertex. (n − j)! Lemma 2: Let x∗ be an optimal decision of (4. The practicability of this investigation depends on the ease with which we can characterize the vertices of Ωp . LINEAR PROGRAMMING As mentioned in 4. Lemma 1: Let x ∈ Ωp . Then x is a vertex of Ωp iff {Aj |j ∈ I(x)} is a linearly independent set. let I(x) = {j|xj > 0}. .24). . we only have to investigate a ﬁnite set.. Since x∗ > 0 for j ∈ I(x∗ ). implies x = y = z. Since Ωp has only ﬁnitely many vertices (see Corollary 1 below). . let z ∗ = x∗ and we are done. Exercise 1: Prove Lemma 1. Then there is a vertex z ∗ of Ωp which is optimal. with y. I(z(θ0 )) ⊂ I(x∗ ) − {j0 } . Deﬁnition: x ∈ Ωp is said to be a vertex of Ωp if x = λy + (1 − λ)z. amj ) . This is done in Lemma 1. j ∈ I(x∗ ) . then at least one vertex of the feasible set Ωp is an optimal solution.38 CHAPTER 4. j x∗ Aj + θ j j∈I(x∗ ) j∈I(x∗ ) Az(θ) = j∈I(x∗ ) zj (θ)Aj = γj Aj =b+θ·0=b. Proof: If {Aj |j ∈ I(x∗ )} is linearly independent. In the following we let Aj denote the jth column of A. Aj = (a1j .e. But from the deﬁnition of z(θ) it is easy to see that we can pick θ0 with |θ0 | = θ ∗ such that zj (θ0 ) = x∗ +θ0 γj = 0 j for at least one j = j0 in I(x∗ ). so that z(θ) is also an optimal solution for |θ| ≤ θ ∗ . Then.

and {Aj |j ∈ I(z)} is linearly independent. j ∈ I(z) . z is optimal iff there exists λ such that λ Aj = cj .A constitute I(z).3. 4. j ∈ I(z) are called the non-basic variables at z. . Let I(z k ) consist of j1 < j2 < . and if not. Notation: Let z be a non-degenerate basic feasible solution. Compute θ = min {(zj γj )|j ∈ i(z). If all these numbers are ≤ 0. j ∈ I(z). because by Lemma 4 below. We make the following simplifying assumption. and let j1 < j2 < . jm ]. stop.A . and if not obtains another basic feasible solution with a higher value.4. Set k = 0 and go to Step 2. (4. .25). . cjm ) and deﬁne λ(z) by λ (z) = c (z)[D(z)]−1 . We call λ(z) the shadow-price vector at z. xj . if {Aj |j ∈ I(z(θ0 ))} is linearly independent. Phase II starts with a basic feasible solution and determines if it is optimal or not. Phase II: Step 1. Deﬁnition: A basic feasible solution z is said to be non-degenerate if I(z) has m elements.. Assumption of non-degeneracy. . .27) ♦ But since z is non-degenerate. Deﬁnition: (i) z is said to be a basic feasible solution if z ∈ Ωp . . in a ﬁnite number of steps. Lemma 3: Let z be a non-degenerate basic feasible solution. let c(z) denote the m-dimensional column vector c(z) = (cj1 . Otherwise pick any ˆ ∈ I(z k ) such that cˆ − λ (z k )Aj > 0 and go to Step 3. for .26) holds iff λ = λ(z) and then (4. γjm ) = [D(z k )]−1 Aj . and xj . for all . We shall discuss Phase II ﬁrst. . sup {c x|x ∈ Ωp } = +∞. are called the basic variables at z. If γ k ≤ 0. Clearly. stop. then we let z ∗ = z(θ0 ) and we are done. j ∈ I(z) . Every basic feasible solution is non-degenerate. . Compute the vector ˆ k k γ k = (γj1 . Proof: By Exercise 6 of Section 2. Let z 0 be a basic feasible solution obtained from Phase I or by any other means. < jm . γj > 0}. . j ∈ I(z) .3. .e. Then z is optimal if and only if λ (z)A ≥ cj . for . Step 2. Otherwise we repeat the procedure above with z(θ0 ).25) (4. j j Step 3. . and the shadow-price vector λ (z k ) = c (z k )[D(z k )]−1 . . ♦ At this point we abandon the geometric term “vertex” and how to established LP terminology. . Otherwise go to Step 4. Let D(z) denote the m × m non-singular matrix D(z) = [Aj1 .c(z k ).2. . i. . k k k Step 4. we obtain a basic feasible solution.26) (4. (4. because z k is optimal ˆ by Lemma 3. Evidently 0 < θ < ∞. either we obtain an optimum solution or we discover that no optimum exists. . < jm . λ Aj ≥ cj . Calculate [D(z k )]−1 . there is no ﬁnite optimum. The algorithm is divided into two parts: In Phase I we determine if Ωp is empty or not. in a ﬁnite number of steps we will ﬁnd an optimal decision z ∗ which is also vertex. . . We will comment on it later.2 The Simplex Algorithm. THE SIMPLEX ALGORITHM 39 Again. Iterating on this procedure. The set I(z) is then called the basis at z. Deﬁne z k+1 by . j2 .27) is the same as (4. . For each j ∈ I(z k ) calculate cj − λ (z k )Aj .

Lemma 4: If γ k ≤ 0. Proof: Deﬁne z(θ) by  k  zj − θγj . j ∈ I(z) and j = ˆ . sup {c x|x ∈ Ωp } = ∞. Finally if we compare (4.24).30) that c zk+1 − c zk = θ{cˆ − γ (z k )Aj } . z(θ) ∈ Ωp for θ ≥ 0. Hence.30) = c z + θ{cˆ − λ j ˆ (z k )Aj }i . we must have γ˜ j from (4. j=ˆ and j ∈ I(z) . Next. k = 0.31) we must have equality. zj = 0 j First of all. j j j j j hence I(z k+1 ) ⊂ (I(z) − {˜ j}) {ˆ . We see then that D(z k+1 ) is obtained from D(z k ) by replacing the column Aj by .29). Az(θ) = Az − θ j∈I(z) (4. j=ˆ j =  k zj = 0 . since γ k ≤ 0 it follows that z(θ) ≥ 0 for θ ≥ 0. ♦ ˆ Corollary 2: In a ﬁnite number of steps Phase II will obtain an optimal solution or will determine that sup{c x|x ∈ Ωp } = ∞. LINEAR PROGRAMMING  k k  zj − θγj . Set k = k + 1 and return to Step 2. j = ˜ But if this is not the j}. Finally. Remark 1: By the non-degeneracy assumption.29) k γj Aj + θA = Az by deﬁnition of ˆ j γk.28) and (4. so that in (4. Then γ(z ∗ ) is an optimal solution of the dual of (4.28) By Lemma 5 below. k+1 k k k Proof: Let ˜ ∈ I(z k ) be such that γ˜ > 0 and z˜ = θγ˜ . j Lemma 5: z k+1 is a basic feasible solution and c z k+1 > c z k . z k+1 is a basic feasible solution with c z k+1 > c z k . j =ˆ θ j  . I(z k+1 ) has m elements.31) ˜ so that it is enough to prove that Aj is independent of {Aj |j ∈ I(z).40 CHAPTER 4. Then from (4.28) we see that z˜ = 0. Corollary 3: Suppose Phase II terminates at an optimal basic feasible solution z ∗ . k k c z(θ) = c z − θc (z )γ + θcˆ j ˆ = c z + θ{cˆ − c (z k )[D(z k )]−1 Aj } j (4. j ∈ I(z) θ . Exercise 2: Prove Corollaries 2 and 3. j ∈ I(z) zj (θ) = . so that c z(θ) → ∞ as θ → ∞. giving a contradiction. we see case. ♦ ˆ But from step 2 {cˆ − λ (z k )Aj } > 0. j k+1 zj (4. j which is positive from Step 2. j} (4.

j . jm ] and if . ji+1 . 1 ≤ i ≤ m . THE SIMPLEX ALGORITHM 41 . The analogous quantity in 3. ˜. . . . y i ≥ 0 . . . Hence at any degenerate basic feasible solution z k we can always ﬁnd I(z k ) ⊃ I(z k ) ¯ k ) has m elements and {Aj |j ∈ I(z k )} is a linearly independent set.A . . + ain xn + yi = bi . . ji−1 . . .A ]. . . jk < ˆ < jk+1 then D(z k+1 ) = [Aj1 .24) by the LP (4. ji−1 . Replace the LP (4. The reason for this is that I(z k ) is not unique.3. Let E be the . . ji−1 . .A . [1970]). .A . jm . . ji+1 . . j.4 is (∂f0 /∂uj )(xk ) − (λk ) (∂f /∂uj )(xk ). Next.. . .A . . . . . Remark 3: By eliminating any dependent equations in (4. We can apply ¯ such that I(z ¯ Phase II using I(z k ) instead of I(z k ).4 is striking. . . For details see (Canon.   1 M =      −γ j1 γ˜ j           1 γ˜ j −γ jm γ˜ j 1 . We now describe how to obtain an initial basic feasible solution. 1 ↑ ith column Then [D(z k+1 )]−1 = P M [D(z k )]−1 . For each j ∈ I(z k ) we can interpret the number cj − λ (z k )Aj to be the net increase in the objective value per unit increase in the jth component of z k . . . . . Remark 2: The similarity between Step 2 of Phase II and Step 2 of the algorithm in 3. . .A . ˆ. . . 1 ≤ j ≤ n . ˆ the column Aj . The basic variables at z k correspond to the variables wk and non-basic variables correspond to uk . matrix E = [Aj1 .A . so that we have to try various ¯ z ¯ k ) until we ﬁnd one for which θ > 0.A . .A . . .3. if ˆ Aj m = =1 γj Aj . . But then in Step 4 it may turn out that θ = 0 so that k+1 = z k . it is easy to check that E −1 = M [D(z k )]−1 where  1 1 . . Phase I: Step I. . In this way the non-degeneracy alternatives for I(z assumption can be eliminated. we can assume that b ≥ 0.32) subject to ail x1 + . by multiplying some of the equality constraints in (4. .A . . . so that these inverses can be easily computed. jm ]..3. j . ˆ. .  . .A j . ji+1 . et al.4.24) we can guarantee that the matrix A ¯ has rank n. . jk .A . . More precisely if D(z k ) = [Aj1 . .32) involving the variables x and y: m Maximize − i=1 yi (4. .A . xj ≥ 0 . This net increase is due to the direct increase cj minus the indirect decrease λ (z k )Aj due to the compensating changes in the basic variables necessary to maintain feasibility. jk+1 . . .A .24) by −1 if necessary. . . . 1 ≤ i ≤ m . . Then [D(z k+1 )]−1 = P E −1 where the matrix P permutes the columns of D(z k+1 ) such that E = D(z k+1 )P . . .

. however. By an input vector we mean any m-dimensional vector r = (r1 . different combinations of inputs can be used to produce the same combination of outputs. . . then it combines (transforms) the input vector (a1j xj . Exercise 3: Show that (4. .” dynamic Malthusian framework where the increase in labor is a function of the output. crude oil.4 LP Theory of a Firm in a Competitive Economy 4. of processes or activities.. . It is these services which are consumed in the transformation into outputs. since human labor can do the same job as some machines and machines can replace other kinds of machines. or computers. We now make three basic assumptions about the ﬁrm. (ii) Each activity combines the k inputs in ﬁxed proportions into the m outputs in ﬁxed proportions. by Exercise 3 below. . . i.32) starting with this solution. . this transformation can be conducted in different ways. . . is a basic feasible solution for (4.4. ﬁnally various kinds of labor services. it may be considered an output in a “closed. or factory buildings. . . . bkj ) so that if it is conducted at a level xj ≥ 0. . Inputs are usually classiﬁed into raw materials such as iron ore. yk ) with y ≥ 0. capital goods 3 such as machines of various kinds. (See the von Neumann model in (Nikaido [1968]). . p. . . . y ∗ ) m since the value of the objective function in (4. This substitutability among inputs is a fundamental concept in economics. .A . .1 Activity analysis of the ﬁrm. say n. bkj xj ) = xj B j . . If = 0. . intermediate products such as steel. We formalize it by specifying which transformation possibilities are available to the ﬁrm. Apply phase II to (4. each activity can be conducted at any non-negative intensity or level. . The ﬁrm’s outputs themselves may be raw materials (if it is a mining company) or intermediate products (if it is a steel mill) or capital goods (if it manufactures lathes) or ﬁnished goods (if it makes shirts or bakes cookies) which go directly to the consumer. Labor is not usually considered an output since slavery is not practiced. There are m kinds of inputs and k kinds of outputs. CHAPTER 4.24). . . amj xj ) = xj Aj into the output vector (b1j xj . LINEAR PROGRAMMING Step 2. a2j . b) is a basic feasible solution of (4. 141. Let . .) Within the ﬁrm. We think of a ﬁrm as a system which transforms input into outputs. Furthermore. . . .24) has no feasible solution. A be the m × n matrix [A1 . ofﬁce equipment. n ] and B be the k × n matrix B = [B 1 .32) lies between − i=1 bi and 0. (4. . .e. Step 3. . Note that (x0 . . Precisely. y∗ x∗ y∗ 4. and by an output vector we mean any k-dimensional vector y = (y1 . rm ) with r ≥ 0. y 0 ) = (0.32). . or textiles. amj ) and B j = (bij . . (i) The transformation of inputs into outputs is organized into a ﬁnite number. . the jth activity is characterized completely by two vectors Aj = (a1j . . chemicals. Go to Step 3. 3 .42 Go to step 2. or raw cotton.24) has a feasible solution iff y ∗ = 0. etc.B It is more accurate to think of the services of capital goods rather than these goods themselves as inputs. n ]. Phase II must terminate in an optimum based feasible solution (x∗ . If = 0. .

. Which of these possible transformations will actually take place depends upon their relative proﬁtability and availability of inputs. Then the manager of the ﬁrm. + xn An into the output vector x1 B 1 + . . x∗ . x∗ . By deﬁnition. The decision variables are the activity levels x1 . say. . . 1 ≤ j ≤ n. whereas the pi .4. . . . r ∗ are in equilibrium iff for all ﬁxed ∆ ∈ Rm . . certain kinds of labor. . . With these assumptions we know all the transformations technically possible as soon as we specify the matrices A and B.33) subject to y = Bx. + ain xn ≤ ri .4. which the ﬁrm ac∗ cepts as given. . . In the long run the supplies of the ﬁrst inputs are also variable and the ﬁrm can change these ∗ supplies from r1 . Whether the ﬁrm will actually change these inputs will depend upon whether it is proﬁtable to do so. . then it transforms the input vector x1 A1 + . .34) Proof: Let c = B p∗ . . (4. . r ∗ are in equilibrium if and only if q ∗ is an optimal solution of (4. q ∗ ) and a set of input supplies ∗ ∗ r ∗ = (r1 . . . qj are prices determined by the whole economy. . rm ) are in (long-term) equilibrium if the ﬁrm has no proﬁt incentive to change r ∗ under the prices (p∗ . In the short-term. . . faces the following decision problem: m Maximize p y − j= +1 q j rj (4. 1 ≤ j ≤ n. . . . . whereas the supply of the remaining inputs can be varied. pk ) of the outputs. . r ∗ by buying or selling these inputs at the market price q1 . Let us suppose that these ∗ inputs are 1. .34): Minimize (r ∗ ) q subject to A q ≥ B p∗ q≥0. . + 1 ≤ i ≤ m . p∗ . + ain xn ≤ ri . qm ) of the inputs is ﬁxed. and perhaps some raw materials. . (4. q ∗ . q ∗ . LP THEORY OF A FIRM IN A COMPETITIVE ECONOMY 43 (iii) If the ﬁrm conducts all the activities simultaneously with the jth activity at level xj ≥ 0. . . We say that the prices (p∗ . . + xn B n . . and they are available in the amounts r1 . . . . ∗ ai1 x1 + . . M (∆) ≤ M (0) where M (∆) is the maximum value of the LP (4.4. . r ∗+1 . ai1 x1 + . q.2 Short-term behavior. if he is maximizing the ﬁrm’s proﬁts. . Theorem 1: p∗ . . rm . . . We assume that the ﬁrm is operating in a competitive economy which means that the unit prices p = (p1 . We study this next. 2. 1 ≤ i ≤ . q ∗ ). 4. + 1 ≤ i ≤ m . and in turn this depends upon the prices p. . . Under realistic conditions (4.4. xj ≥ 0. q .33) has an optimal solution.35) . . ri ≥ 0 . n 1 4. . and q = (q1 . and the short-term input supplies r +1 .35): Maximize c x − (q ∗ ) ∆ subject to Ax ≤ r ∗ + ∆ . . .3 Long-term equilibrium behavior. . the ri are the ﬁxed shortterm supplies. xn . . . ∗ The coefﬁcients of B and A are the ﬁxed technical coefﬁcients of the ﬁrm. . x≥0. r ∗ . the ﬁrm cannot change the amount available to it of some of the inputs such as capital equipment. rm .

c x − (q ∗ ) ∆ ≤ q (r ∗ = ∆) − (q ∗ ) ∆ . (4.15) we see that if x= astj > 0 then m cj = i=1 ∗ qi aij . in equilibrium. M (0) = (r ∗ ) q ∗ . . cj is the revenue per unit level operation of the jth activity so that c x is the revenue when the n activities are operated at levels x. again from (4. and. q ∗ .16). But from (4. r ∗ has a very nice economic interpretation. cj = p∗ b1j + p∗ b2j + . at the optimum. i.. if an equilibrium the optimum ith input supply ri is greater than the optimum demand for the ith input. (4.34)) of (4. is i i=1 qi aij . Thus.38) then the output revenue is c x∗ and the input cost is (q ∗ ) Ax∗ . On the other hand if the jth activity is operated at level xj = 1...34).35) so that by the strong duality theorem. c x − (q ∗ ) ∆ ≤ (q ∗ ) (r ∗ + ∆) − (q ∗ ) ∆ = (q ∗ ) r ∗ ♦ (4. + p∗ bkj .34) becomes the dual of (4. at the optimum activity levels. r ∗ are in long-term equilibrium iff q ∗ is an optimum solution to the dual (namely (4. we can say even more. Hence p∗ . it uses an amount aij of the ith input.35). By weak duality if x is feasible for (4. r ∗ are in equilibrium iff c x − (q ∗ ) ∆ ≤ M (0) = (r ∗ ) q ∗ . if the revenue of an activity is less than its input cost. for q = q ∗ .38): Maximize c x subject to Ax ≤ r ∗ x≥0. then at the optimum it is ∗ operated at zero level. Recall that c = B p∗ .15). (4. q ∗ . (q ∗ ) (Ax∗ − r ∗ ) = 0 so that c x∗ = (q ∗ ) r ∗ . .36) whenever x is feasible for (4. the revenue of an activity operated at a positive level = input cost of that activity.38) This relation between p∗ . q ∗ .e. Also if m cj < i=1 ∗ qi aij . x∗ j .37) Remark 1: We have shown that (p∗ .e.. If the ith input is valued at m a∗ . (4.e. i. In fact. then the input cost of operating at xj = 1. in particular. if x∗ is the optimum activity levels for (4.e.35) and q is feasible for (4. Now bij is the amount of the ith output produced by operating the 1 2 k jth activity at a unit level xj = 1. Hence. i. total revenues = total cost of input supplies.39) i. Finally. LINEAR PROGRAMMING For ∆ = 0. From (4.44 CHAPTER 4. so that the input cost of operating the n activities at levels x is (A q ∗ ) = (q ∗ ) Ax. then = 0.

.e. . suppose that the managers of the various ﬁrms assume that the prices λ are not going to change for a long period of time. . B) characterizing a ﬁrm we can suppose that all the h commodities are possible inputs and all the h commodities are possible outputs. labor. . and j=1 ω(j) = ω. . . . λ∗ . and it is worth decreasing this parameter if the reduction in total cost per unit decrease is i greater than λ∗ .22) that it is always proﬁtable to increase the ith input by buying some additional amount at price qi if λ∗ > qi . . in other words it must be a free good. we can see from (4. r(i) ∈ Rh . are q1 . from our previous analysis we know that the manager of the ith ﬁrm will plan to buy input supplies r(i) ≥ 0. The proﬁt-maximizing behavior of the ﬁrm presented above is one of the two fundamental building blocks in the equilibrium theory of a competitive. . LP THEORY OF A FIRM IN A COMPETITIVE ECONOMY n ∗ ri > j=1 45 aij x∗ . . λ∗ ) is an optimum solution of the dual of (4. for an individual ﬁrm most of the inputs and most of the outputs will be zero. j ∗ then qi = 0. . the jth consumer owns the vector of commodiJ ties ω(j) ≥ 0. Of course. . . capitalist economy. Next. . capitalist economy. . where some of the coefﬁcients bi are design parameters. This interpretation has wide applicability.33).19). Remark 2: Returning to the short-term decision problem (4.. We observe the economy starting at time T . the equilibrium price of an input which is in excess supply must be zero. then we see (assuming i differentiability) that it is worth increasing b∗ if the unit cost of increasing this parameter is less i than λ∗ . Then.10) or (4. ∆ ) the optimum value of ∗ (4. . i. .4 Long-term equilibrium of a competitive. . Often engineering design problems can be formulated as LPs of the form (4. and carry out the i optimization problem. q . The design procedure is to ﬁx these parameters at some nominal value b∗ . . Then if (∂M/∂∆i )|∆=0 exists. . By adding zero rows to the matrices (A. . . and ﬁnished products. suppose that (λ∗ . We are including in ω(j) the amount of his labor services which consumer j is willing to sell. λ∗+1 . which means that the ownership of ω is divided among the various consumers j = 1. λh ) ≥ 0. ω is that portion of the outputs produced prior to T which have not been consumed up to T .4.4. . . Suppose the resulting optimal dual variables are λ∗ . which we mention brieﬂy. the sole purpose for making this change is that we no longer need to distinguish between prices of inputs and prices of outputs. Unfortunately we cannot present the details here. We think of the economy as a feedback process involving ﬁrms and consumers. intermediate and capital goods. such .33). . Let us denote by M (∆1 . . i 4. Let us suppose that there are a total of h commodities in the economy including raw materials. r ∗ + ∆ .4. . At this time there exists within the economy an inventory of the various commodities which we can represent by a vector ω = (ω1 . J. . . Now suppose that at time T the prevailing prices of the h commodities are λ = (λ1 . More precisely.33) when the amounts of the inputs in ﬁxed supply are r1 + ∆1 . ωh ) ≥ 0. We are assuming that this is a capitalist economy. We shall limit ourselves to a rough sketch. . and conversely it is proﬁtable to sell some i of the ith input at price qi if λ∗ < qi . Thus λ∗ can be interpreted as the ﬁrm’s internal valuation of i i the ith input or the ﬁrm’s imputed or shadow price of the ith input. . . . Suppose that the market m 1 prices of inputs 1. . . .

(4. the net supply offered for sale to consumers is S(λ). λ) . because if such an equilibrium price λE exists. 4.44) The most basic question of equilibrium theory is to determine conditions under which there exists a price vector λE such that the economy is in equilibrium. Unfortunately we must stop at this point since we cannot proceed further without introducing some more convex analysis and the ﬁxed point theorem. 2.43) which also satisﬁes J λ D(λ) = j=1 λ ω(j) . We know that r(i) and y(i) depend on λ. and Solow [1958]. .e.42) that is the value of the supply offered to consumers is equal to the value of the commodities (and labor) which they own. the ith manager can sell his planned output y(i) either as input supplies to other ﬁrms or to the consumers. . (4. Samuelson. LINEAR PROGRAMMING that (λ. (4. and he will plan to produce an optimum amount.41) we immediately conclude that J λ S(λ) = j=1 λ ω(j) . (4. . . so we write d(j. λ).46 CHAPTER 4. y(i. The value of the jth consumer’s possessions is λ ω(j). Here also d(j) will depend on λ. λ) ≥ 0 . The second point is that there is no reason to expect that S(λ) ≥ 0. The theory assumes that he will plan to buy a set of commodities d(j) = (d1 (j). dh (j)) ≥ 0 so as to maximize his satisfaction subject to the constraint λ d(j) = λ ω(j). . Chapter 13).40) Now the ith manager can buy r(i) from only two sources: outputs from other ﬁrms. . i. where I is the total number of ﬁrms.40). say y(i).38)) λ r(i. so that we explicitly write r(i. Now we come to the second building block of equilibrium theory. (4. r(i)) is in long term equilibrium. where J I i S(λ) = j=1 ω(j) + i=1 y(i. First of all. Similarly. λ) − i=1 r(i. λ) = λ y(i. . λ) . then at that price the production plans of all the ﬁrms and the buying plan of all the consumers can be realized. If we add up the buying plans of all the consumers we obtain the total demand J D(λ) = j=1 d(j.41) We note two important facts. λ). For a much more general mathematical treatment see (Nikaido [1968]. . I. For a simple treatment the reader is referred to (Dorfman. We also recall that (see (4.. from (4. S(λE ) = D(λE ). Thus. (4. Chapter V). Here i = 1. and the consumers who collectively own ω. λ).5 Miscellaneous Comments . 1 ≤ i ≤ I .

and. The constraints are given as usual by Ax ≤ b. . . However.45) is equivalent to (4.45) iff (x∗ .45) This is not a LP since f0 is not linear.46). in the sense that x∗ is optimal for (4. Exercise 3: Construct an example of the kind (4. 4. . (c2 ) x. It turns out however. .47): n (4. k objective functions (c1 ) x. Exercise 1: Show that (4.47). .5. x ≥ 0 .4. . the Simplex method (together with its variants) is an extremely powerful technique for solving LPs involving thousands of variables.46) below. .5. y ∗ ) = (x∗ .5. there are. an elementary modiﬁcation of the Simplex algorithm can be given to obtain a “local” optimal decision. Exercise 1 will also indicate how to do Exercise 2. 1 ≤ i ≤ k . the following exercise shows how to transform (4. LP is today the single most important optimization technique. See (Miller [1963]). To obtain a feeling for the scope of LP we refer the reader to the book by one of the originators of LP (Dantzig [1963]). piecewise-linear functions of the kind shown in Figure 4. In our LP framework this situation can be formulated as follows. Exercise 2: Obtain an equivalent LP for (4. where the ci are piecewise linear (but not concave). . There may be a number of plausible objective functions. MISCELLANEOUS COMMENTS 47 4. It is often the case in practical decision problems that the objective is not well-deﬁned. x ≤ 0 . x ≤ 0 y ≤ (ci ) x . where ci : R → R are concave. (4. In the next exercise. x ≥ 0. f0 (x∗ )) is optimal for (4. This is because many decision problems can be adequately formulated as LPs. (ck ) x. the interpretation of “equivalent” is purposely left ambiguous. say. and such that there is no equivalent LP. (ck ) x}.46) Maximize j=1 ci (xi ) (4.1 Some mathematical tricks. given the capabilities of modern computers. Maximize f0 (x) subject to Ax ≤ b.3.2 Scope of linear programming.45) into an equivalent LP. The above-given assumption of the concavity of the ci is crucial. so that we have the decision problem. However. . Maximize y subject to Ax ≤ b. that even if the ci are not concave. It is reasonable then to deﬁne a single objective function f0 (x) by f0 (x) = minimum {(c1 ) x.47) subject to Ax ≤ b.

. . .48 CHAPTER 4.3: A function of the form used in Exercise 2. LINEAR PROGRAMMING ci (xi ) . xi Figure 4.

1) where x ∈ Rn . From the discussion in 4. The last section is devoted to computational considerations. .1). is equivalent to (5. The next exercise shows that we could restrict ourselves to objective functions which are linear. Section 3 is devoted to the important special case of quadratic programming. x ∈ Rn . . i = 1.1): 49 . 1. x∗ ∈ Ω is said to be an optimal decision or optimal solution if f0 (x∗ ) ≥ f0 (x) for x ∈ Ω. . Exercise 1: Show that (5. . In Section 1 we present the general nonlinear programming problem (NP) and prove the Kuhn-Tucker theorem.1). fi : Rn → R. .Chapter 5 OPTIMIZATION OVER SETS DEFINED BY INEQUALITY CONSTRAINTS: NONLINEAR PROGRAMMING In many decision-making situations the assumption of linearity of the constraint inequalities in LP is quite restrictive. .1 Qualitative Theory of Nonlinear Programming 5. As in Chapter 4. Two applications are given. m. Section 2 deals with Duality theory for the case where appropriate convexity conditions are satisﬁed. The general NP is a decision problem of the form: Maximize f0 (x) subject to (x) ≤ 0 . with variables y ∈ R.2 it is clear that equality constraints and sign constraints on some of the components of x can all be transformed into the form (5. are differentiable functions. i = 0. however. m. we will not do this. 5.1 The problem and elementary results. .1.2). (5. The linearity of the objective function is not restrictive as shown in the ﬁrst exercise below. . x ∈ Rn is said to be a feasible solution if it satisﬁes the constraints of (5.1. and Ω ⊂ Rn is the subset of all feasible solutions.

. . x) is called the tangent cone of Ω at x. Let K(Ω. The following elementary result is more interesting in this light and should be compared with (2. and y − f0 (x) ≤ 0 . Proof: Let xk ∈ Ω. εmk > 0}∞ be such that xmk → x and (1/εmk )(xmk − x) → hm as k → ∞. . A vector h ∈ Rn is said to be an admissible direction for Ω at x if there exists a sequence xk . .18) in Chapter 2 and Exercise 1 of 4. x) and θ ≥ 0. The argument parallels very closely that developed in Exercise 1 of 4.1 and 4. in Ω and a sequence of numbers εk .2 and compare them with Figures 4. k = 1.50 CHAPTER 5. we see that 0 ∈ C(Ω. we are interested in obtaining conditions which any optimal decision must satisfy. if h ∈ C(Ω.1 and 5. x) we made no use of the particular functional description of Ω. x)}. Lemma 1: Suppose x∗ ∈ Ω is an optimum decision for (5.. (The set I(x) is called the set of active constraints at x. . . Exercise 2: (i) Show that C(Ω.1 and Exercise 1 of 4. Suppose k=1 that hm → h as m → ∞. Two more properties are stated below. x) = {h|h is an admissible direction for Ω at x}.3) . (ii) Let C(Ω. . (5. and let I(x) ⊂ {1. with εk > 0 for all k such that k lim x = x . εmkm }∞ such that m=1 xmkm → x and (1/εmkm )(xmkm − x) → h as m → ∞. . . .1). .1.2. 2.1). k = 1. x) so that the tangent cone is always nonempty. m} be such that fi (x) = 0 for ı ∈ I(x). 3. . Show that there exist subsequences {xmkm . . (Hint for (ii): For m = 1. 2. k→∞ k→∞ lim 1 (xk εk − x) = h . fi (x) < 0 for i ∈ I(x). i. Deﬁnition: Let x be a feasible solution. . x). . . Then f0x (x∗ )h ≤ 0 for all h ∈ C(Ω. let hm and {xmk . x) = {x + h|h ∈ C(Ω. . x) is a cone. (See Figures 5.2) Returning to problem (5.) Deﬁnition: (i) Let x ∈ Ω.e. NONLINEAR PROGRAMMING Maximize y subject to fi (x) ≤ 0.) In the deﬁnition of C(Ω. . x∗ ) .) If we take xk = x and εk = 1 for all k. (ii) Show that C(Ω. 1 ≤ i ≤ m. The basic idea is to linearize the functions fi in a neighborhood of an optimal decision x∗ . then θh ∈ C(Ω. εk > 0. be such that (5. x) is a closed subset of Rn . 2.2. . C(Ω. k = 1. .

we can see that 0≥ = k→∞ lim f0x (x∗ ) (xk −x∗ ) εk + f0x (x∗ )h. Taking limits as k → ∞.4) and (5.4) Note that in particular (5. QUALITATIVE THEORY OF NONLINEAR PROGRAMMING direction of increasing payoff π(k) = {x|f0 (x) = k} Q x∗ 51 {x|f3 (x) = 0} P {x|f2 (x) = 0} Ω R .5. 1 (xk εk k→∞ k→∞ lim − x∗ ) = h . so that 0 ≥ f0x (x∗ ) (x k −x∗ ) (5. (5.4) implies lim 1 εk |xk − x∗ | = |h| .5) k→∞ Since f0 is differentiable. (5.1. using (5. by Taylor’s theorem we have f0 (xk ) = f0 (x∗ + (xk − x∗ )) = f0 (x∗ ) + f0x (x∗ )(xk − x∗ ) + o(|xk − x∗ |) . {x|f1 (x) = 0} Figure 5. and x∗ is optimal.5). Since xk ∈ Ω. we have f0 (xk ) ≤ f0 (x∗ ). ♦ k→∞ lim o(|xk −x∗ |) |xk −x∗ | k→∞ lim |xk −x∗ | εk .1: Ω = P QR k ∗ lim x = x .6) εk + o(|xk −x∗ |) εk .

2: C(Ω. Lemma 2: Let x∗ ∈ Ω. fi (xk ) ≤ 0.4). fi (x∗ ) = 0. Since fi is differentiable. Following the proof of Lemma 1 we can conclude that 0 ≥ fix (x∗ )h. Exercise 3: Let x ∈ R2 . The basic problem that remains is to characterize the set C(Ω.52 CHAPTER 5. x∗ ) - - - - 0 - C(Ω. Show that 1 2 - - - . εk > 0. . NONLINEAR PROGRAMMING x∗ K(Ω. x∗ ) in terms of the derivatives of the functions fi . k = 1. Unfortunately. Then we can apply Farkas’ Lemma just as in Exercise 1 of 4. satisfy (5. in general the inclusion sign in (5. x∗ ) = (1.2. 2}. x2 ) = (x1 − 1)3 + x2 . so that fi (xk ) ≤ fi (x∗ ). x∗ ) Figure 5. (5. ♦ Lemma 2 gives us a partial characterization of C(Ω.7) cannot be reversed. x∗ ). by Taylor’s theorem we have fi (xk ) = fi (x∗ ) + fix (x∗ )(xk − x∗ ) + o(|xk − x∗ |) . The main reason for this is that the set {fix (x∗ )|i ∈ I(x∗ )} is not in general linearly independent. 2. Then I(x∗ ) = {1. Then C(Ω. x∗ ) ⊂ {h|fix (x∗ )h ≤ 0 for all i ∈ I(x∗ )} . and if i ∈ I(x∗ ). 0). and f2 (x1 . . .7) Proof: Let h ∈ Rn and xk ∈ Ω. x∗ ) is the tangent cone of Ω at x∗ . x2 ) = −x2 . Since xk ∈ Ω. Let (x∗ . f1 (x1 . .

5.1. QUALITATIVE THEORY OF NONLINEAR PROGRAMMING
C(Ω, x∗ ) = {h|fix (x∗ )h ≤ 0 , i = 1, 2, }. (Note that {f1x (x∗ ), f2x (x∗ )} is not a linearly independent set; see Lemma 4 below.)

53

5.1.2 Kuhn-Tucker Theorem.
Deﬁnition: Let x∗ ∈ Ω. We say that the constraint qualiﬁcation (CQ) is satisﬁed at x∗ if C(Ω, x) = {h|fix (x∗ )h ≤ 0 for all i ∈ I(x∗ )}, and we say that CQ is satisﬁed if CQ is satisﬁed at all x ∈ Ω. (Note that by Lemma 2 C(Ω, x) is always a subset of the right-hand side.) Compare the next result with Exercise 2 of 4.2. Theorem 1: (Kuhn and Tucker [1951]) Let x∗ be an optimum solution of (5.1), and suppose that CQ is satisﬁed at x∗ . Then there exist λ∗ ≥ 0, for i ∈ I(x∗ ), such that i f0x (x∗ ) =
i∈I(x∗ )

λ∗ fix (x∗ ) i

(5.8)

Proof: By Lemma 1 and the deﬁnition of CQ it follows that f0x (x∗ )h ≤ 0 whenever fix (x∗ )h ≤ 0 for all i ∈ I(x∗ ). By the Farkas’ Lemma of 4.2.1 it follows that there exist λ∗ ≥ 0 for i ∈ I(x∗ ) i such that (5.8) holds. ♦ In the original formulation of the decision problem we often have equality constraints of the form rj (x) = 0, which get replaced by rj (x) ≤ 0, −rj (x) ≤ 0 to give the form (5.1). It is convenient in application to separate the equality constraints from the rest. Theorem 1 can then be expressed as Theorem 2.

Theorem 2: Consider the problem (5.9). Maximize f0 (x) subject to fi (x) ≤ 0 , i = 1, . . . , m, rj (x) = 0 , j = 1, . . . , k .

(5.9)

Let x∗ be an optimum decision and suppose that CQ is satisﬁed at x∗ . Then there exist λ∗ ≥ 0, i = i 1, . . . , m, and µ∗ , j = 1, . . . , k such that j
m

f0x (x∗ ) =
i=1

λ∗ fix (x∗ ) + i

k j=1

µ∗ rjx (x∗ ) , j

(5.10)

and λ∗ = 0 whenever fi (x∗ ) < 0 . i Exercise 4: Prove Theorem 2. (5.11)

54

CHAPTER 5. NONLINEAR PROGRAMMING

An alternative form of Theorem 1 will prove useful for computational purposes (see Section 4). Theorem 3: Consider (5.9), and suppose that CQ is satisﬁed at an optimal solution x∗ . Deﬁne ψ : Rn → R by ψ(h) = max {−f0x (x∗ )h, f1 (x∗ ) + f1x (x∗ )h, . . . , fm (x∗ ) + fmx (x∗ )h} , and consider the decision problem Minimize ψ(h) subject to −ψ(h) − f0x (x∗ )h ≤ 0, −ψ(h) + fi (x∗ ) + fix (x∗ )h ≤ 0 , 1 ≤ i ≤ m −1 ≤ hi ≤ 1 , i = 1, . . . , n . Then h = 0 is an optimal solution of (5.12). Exercise 5: Prove Theorem 3. (Note that by Exercise 1 of 4.5, (5.12) can be transformed into a LP.) Remark: For problem (5.9) deﬁne the Lagrangian function L:
m k

(5.12)

(x1 , . . . , xn ; λ1 , . . . , λm ; µ1 , . . . , µk ) → f0 (x) −
i=1

λi fi (x) −
j=1

µj rj (x).

Then Theorem 2 is equivalent to the following statement: if CQ is satisﬁed and x∗ is optimal, then there exist λ∗ ≥ 0 and µ∗ such that Lx (x∗ , λ∗ , µ∗ ) = 0 and L(x∗ , λ∗ , µ∗ ) ≤ L(x∗ , λ, µ) for all λ ≥ 0, µ. There is a very important special case when the necessary conditions of Theorem 1 are also sufﬁcient. But ﬁrst we need some elementary properties of convex functions which are stated as an exercise. Some additional properties which we will use later are also collected here. Recall the deﬁnition of convex and concave functions in 4.2.3. Exercise 6: Let X ⊂ Rn be convex. Let h : X → R be a differentiable function. Then (i) h is convex iff h(y) ≥ h(x) + hx (x)(y − x) for all x, y, in X, (ii) h is concave iff h(y) ≤ h(x) + hx (x)(y − x) for all x, y in X, (iii) h is concave and convex iff h is afﬁne, i.e. h(x) ≡ α + b x for some ﬁxed α ∈ R, b ∈ Rn . Suppose that h is twice differentiable. Then (iv) h is convex iff hxx (x) is positive semideﬁnite for all x in X, (v) h is concave iff hxx (x) is negative semideﬁnite for all x in X, (vi) h is convex and concave iff hxx (x) ≡ 0. Theorem 4: (Sufﬁcient condition) In (5.1) suppose that f0 is concave and fi is convex for i = 1, . . . , m. Then (i) Ω is a convex subset of Rn , and (ii) if there exist x∗ ∈ Ω, λ∗ ≥ 0, i ∈ I(x∗ ), satisfying (5.8), then x∗ is an optimal solution of i (5.1). Proof: (i) Let y, z be in Ω so that fi (y) ≤ 0, fi (z) ≤ 0 for i = 1, . . . , m. Let 0 ≤ θ ≤ 1. Since fi is convex we have

5.1. QUALITATIVE THEORY OF NONLINEAR PROGRAMMING
fi (θy + (1 − θ)z) ≤ θfi (y) + (1 − θ)fi (z) ≤ 0 , 1 ≤ i ≤ m, so that (θy + (1 − θ)z) ∈ Ω, hence Ω is convex. (ii) Let x ∈ Ω be arbitrary. Since f0 is concave, by Exercise 6 we have f0 (x) ≤ f0 (x∗ ) + f0x (x∗ )(x − x∗ ) , so that by (5.8) f0 (x) ≤ f0 (x∗ ) +
i∈I(x∗ )

55

λ∗ fix (x∗ )(x − x∗ ) . i

(5.13)

Next, fi is convex so that again by Exercise 6, fi (x) ≥ fi (x∗ ) + fix (x∗ )(x − x∗ ) ; but fi (x) ≤ 0, and fi (x∗ ) = 0 for i ∈ I(x∗ ), so that fix (x∗ )(x − x∗ ) ≤ 0 for i ∈ I(x∗ ) . (5.14) Combining (5.14) with the fact that λ∗ ≥ 0, we conclude from (5.13) that f0 (x) ≤ f0 (x∗ ), so that i x∗ is optimal. ♦ Exercise 7: Under the hypothesis of Theorem 4, show that the subset Ω∗ of Ω, consisting of all the optimal solutions of (5.1), is a convex set. Exercise 8: A function h : X → R deﬁned on a convex set X ⊂ Rn is said to be strictly convex if h(θy + (1 − θ)z) < θh(y) + (1 − θ)h(z) whenever 0 < θ < 1 and y, z are in X with y = z. h is said to be strictly concave if −h is strictly convex. Under the hypothesis of Theorem 4, show that an optimal solution to (5.1) is unique (if it exists) if either f0 is strictly concave or if the fi , 1 ≤ i ≤ m, are strictly convex. (Hint: Show that in (5.13) we have strict inequality if x = x∗ .)

5.1.3 Sufﬁcient conditions for CQ.
As stated, it is usually impractical to verify if CQ is satisﬁed for a particular problem. In this subsection we give two conditions which guarantee CQ. These conditions can often be veriﬁed in practice. Recall that a function g : Rn → R is said to be afﬁne if g(x) ≡ α + b x for some ﬁxed α ∈ R and b ∈ Rn . We adopt the formulation (5.1) so that Ω = {x ∈ Rn |fi (x) ≤ 0 , 1 ≤ i ≤ m} . Lemma 3: Suppose x∗ ∈ Ω and suppose there exists h∗ ∈ Rn such that for each i ∈ I(x∗ ), either fix (x∗ )h∗ < 0, or fix (x∗ )h∗ = 0 and fi is afﬁne. Then CQ is satisﬁed at x∗ . Proof: Let h ∈ Rn be such that fix (x∗ )h ≤ 0 for i ∈ I(x∗ ). Let δ > 0. We will ﬁrst show that (h + δh∗ ) ∈ C(Ω, x∗ ). To this end let εk > 0, k = 1, 2, . . . , be a sequence converging to 0 and set xk = x∗ + εk (h + δh∗ ). Clearly xk converges to x∗ , and (1/εk )(xk − x∗ ) converges to (h + δh∗ ). Also for i ∈ I(x∗ ), if fix (x∗ )h < 0, then fi (xk ) = fi (x∗ ) + εk fix (x∗ )(h + δh∗ ) + o(εk |h + δh∗ |) ≤ δεk fix (x∗ )h∗ + o(εk |h + δh∗ |) < 0 for sufﬁciently large k , whereas for i ∈ I(x∗ ), if fi is afﬁne, then

NONLINEAR PROGRAMMING fi (xk ) = fi (x∗ ) + εk fix (x∗ )(h + δh∗ ) ≤ 0 for all k . so that {fix (x∗ . fi (xk ) = fi (g(uk ). uk ) = 0.16) If we compare (5. (Hint: Show fi (x i i x i that h∗ = x − x∗ satisﬁes the hypothesis of Lemma 3. either ˆ ∗ ) < 0 and f is convex. u∗ )|i ∈ Jδ } is linearly independent. and hence 0 = fiw (x∗ )gu (u∗ )(η + δη ∗ ) + fiu (x∗ )(η + δη ∗ ) . Thus we have also shown that xk ∈ Ω for sufﬁciently large k. Next we partition h. . or f (ˆ) ≤ 0 and f is afﬁne. 0 = fi (g(u). Finally. First of all. x∗ ). Thus. h∗ as h = (ξ. such that fi (w. i ∈ Jδ . Since δ > 0 can be arbitrarily small. (5. fix (x∗ )h∗ = 0} is a linearly independent set. We will show that (h + δh∗ ) ∈ C(Ω. ♦ Exercise 9: Suppose x∗ ∈ Ω and suppose there exists x ∈ Rn such that for each i ∈ I(x∗ ). an open set V ⊂ Rn containing x∗ = (w∗ . . and w = g(u) . u∗ ). 2 . so wk = g(uk ) converges to w∗ = g(u∗ ). whereas for i ∈ Jδ . i ∈ I(x∗ ). and set uk = u∗ + εk (η + δη ∗ ). uk − u∗ ) = (1/εk )(g(uk ) − g(u∗ ). x∗ ) is a closed set by Exercise 2. u) ∈ V iff u ∈ U. and so by deﬁnition (h + δh∗ ) ∈ C(Ω. But for i ∈ Jδ we have 0 = fix (x∗ )(h + δh∗ ) = fiw (x∗ )(ξ + δξ ∗ ) + fiu (x∗ )(η + δη ∗ ) . η ∗ ) corresponding to the partition of x = (w. and since C(Ω. it follows that h ∈ C(Ω. Let Jδ = {i|i ∈ I(x∗ ). wk = g(uk ). u). Since g is differentiable. . uk ). xk converges to x∗ . where U = {u ∈ Rn−p ||u − u∗ | < ρ}. and (w. h∗ = (ξ ∗ .16) and recall that {fiw (x∗ )|i ∈ Jδ } is a basis in Rp we can conclude that (ξ + δξ ∗ ) = gu (u∗ )(η + δη ∗ ) so that (1/εk )(xk − x∗ ) converges to (h + hδh∗ ). Let εk > 0. fix (x∗ )(h + δh∗ ) = 0}. so that fi (xk ) < 0 for sufﬁciently large k. εk (η + δη ∗ )). u) = 0. We note that uk converges to u∗ . By the Implicit Function Theorem.15) and (5. u) for u ∈ U so that 0 = fiw (x∗ )gu (u∗ ) + fiu (x∗ ). x∗ ). and ﬁnally xk = (sk . η + δη ∗ ). for i ∈ I(x∗ ) we have fi (x∗ ) < 0. Then CQ is satisﬁed at x∗ .) ˆ Lemma 4: Suppose x∗ ∈ Ω and suppose there exists h∗ ∈ Rn such that fix (x∗ )h∗ ≤ 0 for i ∈ I(x∗ ). and a differentiable function g : U → Rp . be any sequence converging to 0. for i ∈ Jδ . for i ∈ Jδ .56 CHAPTER 5. and {fix (x∗ )|i ∈ I(x∗ ). consist of p elements. Proof: Let h ∈ Rn be such that fix (x∗ )h ≤ 0 for all i ∈ I(x∗ ). Now (1/εk )(xk − x∗ ) = (1/εk )(wk − w∗ . (5. fi (xk ) = fi (x∗ ) + fix (x∗ )(xk − x∗ ) + o(|xk − x∗ |) fi (x∗ ) + εk fix (x∗ )(h + δh∗ ) + o(εk ) + o(|xk − x∗ |). Clearly Jδ ⊂ J = {i|i ∈ I(x∗ ). Let δ > 0.15) Also. . Then CQ is satisﬁed at x∗ . there exist ρ > 0. fi x(x∗ )h∗ = 0}. k = 1. it follows that (1/εk )(xk − x∗ ) converges to (gu (u∗ )(η + δη ∗ ). It remains to show that xk ∈ Ω for sufﬁciently large k. x∗ ). η).

and it has provided many unifying conceptual insights into economics and management science.2. and C(Ω.17) then M (ˆ = f0 (ˆ). ˆm ) is a given vector. rjx (x∗ )h∗ = 0 for 1 ≤ j ≤ k. Lemma 5: Suppose x∗ is feasible for (5. so that h ∈ C(Ω.17) as ˆ varies. X is a given convex subset of Rn and ˆ = (ˆ1 . 1 ≤ i ≤ m b x∈X . . Thus. and fix (x∗ )h∗ ≤ 0 for i ∈ I(x∗ ). λ ≥ 0. we will give some geometric insight. fi : Rn → R. fm ) : Rn → Rm . . are given convex functions. DUALITY THEORY 57 and since fi (x∗ ) = 0 whereas fix (x∗ )(h + δh∗ ) < 0.2.2 Duality Theory Duality theory is perhaps the most beautiful part of nonlinear programming.2. f (x) ≤ b}. . To ﬁnish the proof we note that δ > 0 can be made arbitrarily small. . x∗ ). .5. (5. . Hence.3.9). (h + δh∗ ) ∈ C(Ω. However. We can only present some of the basic results here.1 and the results in 4. . . Then CQ is satisﬁed at x∗ . x∗ ). fix (x∗ )h∗ = 0} {rjx (x∗ )|j = 1. x∗ ) is closed by Exercise 2. be ﬁxed. B = {b|Ω(b) = φ}. we can conclude that fi (xk ) < 0 for sufﬁciently large k. Exercise 10: Prove Lemma 5 5. ♦ The next lemma applies to the formulation (5. In 2.18) . (5. in terms of suggesting important computational algorithms.1 should be compared with Theorems 1 and 4 of 4. Consider problem (5. 5. 1 ≤ i ≤ m. f0 : Rn → R is a given concave function. . let f = (f1 . So we deﬁne b Ω(b) = {x|x ∈ X. xk ∈ Ω for sufﬁciently large k. and even so some of the proofs are relegated to the Appendix at the end of this Chapter since they depend on advanced material.9) and suppose there exists h∗ ∈ Rn such that the set {fix (x∗ )|i ∈ I(x∗ ). b b b For convenience.2. It may be useful to note in the following discussion that most of the results do not require differentiability of the various functions. k} is linearly independent. We wish to examine the behavior of the maximum value of (5. f (x) ≤ b} = sup{f0 (x)|x ∈ Ω(b)} .1 Basic results. . Maximize f0 (x) − λ (f (x) − ˆ b) subject to x ∈ X . Its proof is left as an exercise since it is very similar to the proof of Lemma 4. so that in particular if x∗ is an optimal solution of (5. The results in 2. and M : B→R {+∞} by M (b) = sup{f0 (x)|x ∈ X. . It has resulted in many applications within nonlinear programming. We need to consider b) x the following problem also.17) where x ∈ Rn . Let λ ∈ Rm .17) which we call the primal problem: Maximize f0 (x) subject to fi (x) ≤ ˆi . .3 we give some application of duality theory and in 2.2 we refer to some of the important generalizations.

b). However. (5. b)|x Problem (5.e. it is sometimes possible to include some of the constraints in X in such a way that the calculation of m(λ) by (5.. ˆ Deﬁnition: A pair (ˆ. b).) Exercise 1: Prove Lemma 1. we have f0 (x) ≤ M (ˆ ≤ m(λ) for x ∈ Ω(ˆ λ ≥ 0 . Lemma 2: (Weak duality) If x is feasible for (5.17) is usually equal to Rn and then. Remark 2: It is sometimes useful to know that Lemmas 1 and 2 below hold without any convexity conditions on f0 . b b) f0 (x) ≤ f0 (x) − λ (f (x) − ˆ for x ∈ Ω(ˆ λ ≥ 0 . b) b).2 below. b)|x Thus. So. Lemma 1 shows that the cost function of the dual problem is convex which is useful information since there are computation techniques which apply to convex cost functions but not to arbitrary nonlinear cost functions. X. Lemma 2 shows that the optimum value of the dual problem is always an upper bound for the optimum value of the primal. f. (Here R+ = {λ ∈ Rn |λ ≥ 0}.18) and the solution of the dual problem (5. ≤ sup {f0 (x) − λ (f (x) − b)|x b)} b) ≤ sup {f0 (x) − λ (f (x) − ˆ ∈ X} = m(λ) .58 and deﬁne CHAPTER 5. and λ ≤ 0 is said to satisfy the optimality conditions if x ˆ ˆ (5.19) is called the dual problem: Minimize m(λ) subject to λ ≥ 0 .20). f0 (x) ≤ M (ˆ ≤ m∗ ≤ m(λ) . NONLINEAR PROGRAMMING m(λ) = sub{f0 (x) − λ (f (x) − ˆ ∈ X} . λ) with x ∈ X.20) . Remark 1: The set X in (5. ♦ ˆ = m∗ in The basic problem of Duality Theory is to determine conditions under which M (b) (5. if we take the inﬁmum with respect to λ ≥ 0 in the right-hand b) inequality we get (5. and λ ≥ 0. of course.3. i. there is no reason to separate it out.20).19) become simple. we have λ (f (x) − ˆ ≤ 0. x ∈ Ω(ˆ and if λ ≥ 0. Hence f0 (x) ≤ sup {f0 (x)|x ∈ Ω(ˆ = M (ˆ b)} b) ˆ ∈ Ω(ˆ and since Ω(ˆ ⊂ X. n n Lemma 1: m : R+ → R {+∞} is a convex function. and since M (ˆ is independent of λ. b) Proof: Since f (x) − ˆ ≤ 0. then b).3. For example see the problems discussed in Sections 2.1 and 2.19) Let m∗ = inf {m(λ)|λ ≥ 0}.17). We ﬁrst give a simple sufﬁciency condition.

23) x so that x is optimal for the primal.17). and M (ˆ = m∗ . so that b fi (θx + (1 − θ)˜) ≤ θb + (1 − θ)˜ . ♦ We now proceed to a much more detailed investigation. λ) satisfy the optimality conditions. . b) ˆ Proof: Let x ∈ Ω(ˆ so that λ (f (x) − ˆ ≤ 0.18) with λ = λ. f0 (ˆ) = M (b) x ˆ so that from Weak Duality λ is optimal for the dual. and hence by deﬁnition f0 (ˆ) = M (ˆ Also ˆ x b). fi (ˆ) ≤ ˆi for i = 1. ˆ 59 (5. condition.. m . x ∈ Ω(˜ let 0 ≤ θ ≤ 1. i. Also. and fi (θx + (1 − θ)˜) ≤ θfi (x) + (1 − θ)fi (˜) x x since fi is convex. equivalently. ˆ ˆ m(λ) = f0 (ˆ) − λ (f (ˆ) − ˆ x x b) ˆ . ˆ x b (5. λ) satisfy the optimality ˆ x ˆ ˆ by virtue of (5. ˜ belong to B. B is convex. Proof: Let b.24) . Note that in this case x ∈ Ω(b) ˆ The next result is equivalent to Theorem 4(ii) of Section 1 if X = Rn . are differentiable. Then (θx + (1 − θ)˜) ∈ X b ˜ b). . x since X is convex.21) x is feasible for (5. b) ˆ f0 (x) ≤ f0 (x) − λ (f (x) − ˆ b) ˆ (f (x) − ˆ ∈ X} ≤ sup{f0 (x) − λ b)|x ˆ = f0 (ˆ) − λ (f (ˆ) − ˆ by (5. 0 ≤ i ≤ m. let x ∈ Ω(b). Lemma 3: B is a convex subset of Rm . since f0 is concave. λ is an optimal solution to the dual. then x is an optimal solution to x ˆ ˆ ˆ the primal.23) ˆ λ ≥ 0 is said to be an optimal price vector if there is x ∈ X such that (ˆ. .22) ˆ ˆ λi = 0 when fi (ˆ) < ˆi .2. Then b).22). x b x b) (5. λ (f (ˆ) − ˆ = 0. x x (5.e.5. DUALITY THEORY ˆ x is optimal solution of (5. and M : B → R {+∞} is a concave function. x hence (θx + (1 − θ)˜) ∈ Ω(θb + (1 − θ)˜ x b) and therefore. and fi . f0 (θx + (1 − θ)˜) ≥ θf0 (x) + (1 − θ)f0 (˜) . Theorem 1: (Sufﬁciency) If (ˆ. .21) x x b) = f0 (ˆ) by (5.

- ˆ b - - b . NONLINEAR PROGRAMMING Since (5. b) b| (In words. −∞}. b M (b) - ˆ b M is stable at ˆ b M (ˆ + λ (b − ˆ b) b) - .) M (b) M (b) M (ˆ b) b) . x ∈ Ω(˜ b) x ˜ b)} ≥ sup{f0 (x)|x ∈ Ω(b)} + (1 − θ) sup {f0 (˜)|˜ ∈ Ω(˜ x x b)} ˜ = θM (b) + (1 − θ)M (b). ♦ Deﬁnition: Let X ⊂ Rn and let g : X → R {∞.3: Illustration of supergradient of stability.24) holds for all x ∈ Ω(b) and x ∈ Ω(˜ it follows that ˜ b) M (θb + (1 − θ)ˆ ≥ sup {f0 (θx + (1 − θ)˜)|x ∈ Ω(b).∈M (ˆ b B . x ˆ (g(x) ≥ g(ˆ) + λ (x − x) for x ∈ X.) A more geometric way of thinking about subgradients is the following. b ˆ b M is not stable at ˆ b . Figure 5.3.60 CHAPTER 5. M is stable at ˆ if M does not increase inﬁnitely steeply in a neighborhood of ˆ See b b. Deﬁnition: The function M : B → R number K such that {∞} is said to be stable at ˆ ∈ B if there exists a real b M (b) ≤ M (ˆ + K|b − ˆ for b ∈ B .) x ˆ (See Figure 5-3. λ is a supergradient at ˆ b Figure 5. A vector λ ∈ Rn is said to be a supergradient (subgradient) of g at x ∈ X if ˆ g(x) ≤ g(ˆ) + λ (x − x) for x ∈ X. Deﬁne the subset A ⊂ R1+m by .

(M (b). −λm ) deﬁnes a non-vertical hyperplane b supporting A at (M (ˆ ˆ (ii) if (λ0 . λm ) is said to be the normal to a hyperplane supporting A at a point(ˆ. f (ˆ) = M (ˆ x ∈ X. . b). and M (ˆ < ∞. −λ1 . . and (1. Then x b) x (f0 (x).23). −λm ) deﬁnes a hyperplane supporting A at b). A lies below the hyperplane π = {(r. . See Figure 5. ˜ ∈ A. ˆ x b. (λm /λ0 )) is a supergradient of M at ˆ b. M (b) ≤ M (ˆ + λ (b − ˆ b) b) ≤ M (ˆ + |λ||b − ˆ . hence again by Exercise 3 ˆ M (ˆ − λ ˆ ≥ f0 (x) − λ f (x) . M (˜ and (˜. if the hyperplane is non-vertical then b). λ ≥ 0 and ˆ By Exercise 2. DUALITY THEORY A = {(r. . b)|b ∈ B. . λ) x ˆ Proof: By hypothesis. b b b. ˆ ˆ Lemma 5: Suppose that x is optimal for (5. . . We will prove only one part of the next crucial result. . i=1 (5.) The supporting (In words. . (M (ˆ ˆ then λ0 ≥ 0. But then λ (ˆ − f (ˆ)) = 0. 1 . This neologism contrasts with the epigraph of a function which is the set lying above the graph of the function. . b) ˆ b ˆ Since f0 (ˆ) = M (ˆ and λ (f (ˆ) − ˆ = 0. ˆ if r b) m m λ0 r + ˆ i=1 b λiˆi ≥ λ0 r + λi bi for all (r.4. . Show that (i) if λ = (λ1 . Lemma 4: (Gale [1967]) M is stable at ˆ iff M has a supergradient at ˆ Proof: (Sufﬁciency only) b b.17). λi ≥ 0 for 1 ≤ i ≤ m. b) b| ♦ The next two results give important alternative interpretations of supergradients. and f (ˆ) ≤ ˆ Let λ be a supergradient of M at ˆ x b). r b) ˆ ∈ B. ˜ b). b) ∈ A . Deﬁnition: A vector (λ0 . . Exercise 2: Show that if ˆ ∈ B. b b).5. and then (ˆ. b. ˆ then Let λ be a supergradient at b. and r ≤ M (b)} . λm ) is a Exercise 3: Assume that b b) supergradient of M at ˆ then λ ≥ 0. . . giving (5. . λ1 . b)|λ0 r+ λi bi = λ0 r + ˆ ˆ hyperplane is said to be non-vertical if λ0 = 0.25) λi bi }. b) ((λ1 /λ0 . 61 Thus A is the set lying ”below” the graph of M . optimal price vector. x M (ˆ − λ ˆ ≥ M (ˆ − λ f (ˆ) . futhermore. .4. we can rewrite the inequality above as x b). We call A the hypograph1 of M . Next let x ∈ X. f (x)) ∈ A. ˜ ≥ ˆ and r ≤ M (ˆ then ˜ ∈ B. . ˆ f (ˆ)) ∈ A and by Exercise 3. Since M is concave it follows immediately that A is convex (in fact these are equivalent statements). . . or see the Appendix at the end of this Chapter. . Then λ is a supergradient of M at ˆ iff λ is an ˆ b ˆ satisfy the optimality conditions.2. −λ1 . x b) From the Greek “hypo” meaning below or under. b) ˆ b b) ˆ x ˆ ˆ b so that λ (f (ˆ) − ˆ ≥ 0. The reader who is familiar with the Separation Theorem of convex sets should be able to construct a proof for the second part based on Figure 5.

. f (x) ≤ b. . ♦ ˆ ˆ Lemma 6: Suppose that ˆ ∈ B.23) . . Then λ is a supergradient of M at ˆ iff λ is an b b) b ˆ optimal solution of the dual (5. Let x ∈ Ω(b).23). Then λ ˆ f0 (x) ≤ f0 (x) λ (f (x) − b) ˆ = f0 (x) − λ (f (x) − ˆ + λ (b − ˆ b) ˆ b) ˆ (f (ˆ) − ˆ + λ (b − ˆ ˆ ≤ f0 (ˆ) − λ x x b) b) ˆ (b − ˆ = f0 (ˆ) + λ x b) = M (ˆ + λ (b − ˆ .4: Hypograph and supporting hyperplane. i. and (5. suppose x ∈ X. b) M (b) π ˆ M (ˆ b) (λ0 . b) Figure 5. A ˆ b b π is a non-vertical hyperplane supporting A at (M (ˆ ˆ ˆ b). ˆ ˆ (f (x) − b) ≤ 0 so that x ∈ X.e. x x b) b) so that (5. b) ˆ b) ˆ so that λ is a supergradient of M at ˆ b.21).21) by (5. NONLINEAR PROGRAMMING M (ˆ b) . by (5. λm ) . ˆ ˆ f0 (ˆ) + λ (f (ˆ) − ˆ ≥ f0 (x) − λ (f (x) − ˆ . ˆ be a supergradient of M at ˆ Let x ∈ X. (5.19) and m(λ) = M (ˆ b).21) holds. λ) satisfy the optimality conditions. It follows that (ˆ. b) ˆ b) Hence M (b) = sup{f0 (x)|x ∈ Ω(b)} ≤ M (ˆ + λ (b − ˆ . . x ˆ ˆ Conversely. . A b ˆ b No non-vertical hyperplane supporting A at (M (ˆ ˆ b). and M (ˆ < ∞.62 M (b) CHAPTER 5. λ ≥ 0 satisfy (5. By Exercises 2 and 3 Proof: Let λ b.22).

23). M (b) = sup{f0 (x)|x ∈ Ω(b)} ≤ M (ˆ + λ (b − ˆ . ♦ We can now summarize our results as follows.2. DUALITY THEORY ˆ M (ˆ − λ ˆ ≥ f0 (x) − λ f (x) b) ˆ b or ˆ M (ˆ ≥ f0 (x) − λ (f (x) − ˆ . (ii) is implied by Lemma 6.19).21). by Lemmas 4.22). the various conditions of Section 1. ˆ is any optimal solution for the dual. ♦ ˆ Corollary 1: Under the hypothesis of Theorem 2. then M is stable at ˆ b. Proof: (i) follows from Lemmas 4. ˆ M (ˆ ≥ f0 (x) − λ (f (x) − ˆ . 6 stability is equivalent to the existence of optimal dual variables. Thus the condition of stability of M at ˆ plays a similar role to the constraint qualiﬁcation. It is easy to see using convexity properties that.2. ˆ ˆ (ii) λ is optimal for the dual iff λ is a supergradient of M at ˆ b. . so that ˆ M (ˆ ≥ f0 (x) − λ (f (x) − ˆ + λ (f (x) − b) b) b) ˆ ˆ ˆb = f0 (x) − λ b + λ ˆ for x ∈ Ω(b) .6. then λ (f (x) − b) ≤ 0. whereas CQ is only a sufﬁcient condition. b However. In other words if CQ holds at x then M is stable at ˆ In particular. Here if X = R i we give one sufﬁcient condition which implies stability for the general case.23) are equivalent to the Kuhn-Tucker condition (5.2. in particular if there exists x ∈ X such that fi (x) < ˆi for b b 1 ≤ i ≤ m. then the optimality conditions (5. (5. Hence. b) b) so that ˆ ˆ M (ˆ ≥ sup{f0 (x) − λ (f (x) − ˆ ∈ X} = m(λ) . and (5. The “if” part of (iii) follows from Theorem 1.3. and m(λ) = M (ˆ b).3 imply stability. ˆ b. if λ is an optimal solution to the dual then (∂M + /∂bi )(ˆ ≤ λi ≤ (∂M − /∂bi )(ˆ b) ˆ b). Lemma 7: If ˆ is in the interior of B.5. whereas the “only if” part of (iii) follows from Lemma 5.21). Theorem 2: (Duality) Suppose ˆ ∈ B. are differentiable. Exercise 4: Prove Corollary 1. (5. if X = Rn and fi . n and the f are differentiable. ˆ ˆ (i) there exists an optimal solution λ for the dual.2 Interpretation and extensions. and M is stable at ˆ Then b b) b.8). and (5. 0 ≤ i ≤ m. then x is optimal for the primal iff (ˆ. M (ˆ < ∞.) 5. (Hint: See Theorem 5 of 4. b) b)|x ˆ ˆ By weak duality (Lemma 2) it follows that M (ˆ = m(λ) and λ is optimal for (5.22). λ) satisfy the (iii) if λ ˆ x ˆ optimality conditions of (5. b) ˆ b) 63 ˆ so that λ is a supergradient. and m(λ) = M (ˆ Then for any x ∈ X ˆ Conversely suppose λ b). b) ˆ ≥ 0. b) b) ˆ and if moreover f (x) ≤ b.

b). the ﬁrm faces the decision problem ˆ an optimal solution of (5. Next.26) . If for a price system λ. . Assuming the realistic condition ˆ ∈ B.17) is the short-term decision problem faced by the ﬁrm. M (ˆ < ∞ we can see from Theorem 2 and its Corollary 1 that there exists b b) an equilibrium price system iff (∂M + /∂bi )(ˆ < ∞. The basic idea involved is to consider supporting A at (M (ˆ ˆ by (non-vertical) surfaces π more genb).26): Maximize f0 (x) − F (f (x) − ˆ b) subject to x ∈ X . b as the vector of current resource supplies.64 CHAPTER 5. if the current ˆ resources can be bought or sold at prices λ = (λ1 . b) ˆ eral than hyperplanes. The primal problem (5. M (b) M (ˆ b) . . equivalently. there may be no hyperplane supporting A at (M (ˆ ˆ This is the reason why duality theory requires the often restrictive convexity hypothb). Instead of (5. and comparing with Figure 5. f0 (x) the corresponding revenue. and ˆ is the interior of B. M (ˆ < ∞ without loss of generality. ˆ then we can interpret λ as a system of equilibrium prices just as in 4.6. and ﬁnally f (x) the amount of these resources used up at activity levels x.2. It is possible to obtain the duality theorem under conditions slightly weaker than convexity but since these conditions are not easily veriﬁable we do not pursue this direction any further (see Luenberger [1968]).5: If M is not concave there may be no supporting hyperplane at (M (ˆ ˆ b). λm ) .5 it is evident that if M is not concave or. A ˆ b b Figure 5. esis on X and fi . if we interpret (∂M + /∂bi )(ˆ b) b) as the marginal revenue of the ith resource. Thus. if its hypograph A is not convex.18) we would then have more general problem of the form (5. X as constraints due to physical or long-term limitations.18).4. we can think of x as the vector of n activity levels. NONLINEAR PROGRAMMING The proof rests on the Separation Theorem for convex sets. and only depends on the fact that M is concave. A much more promising development has recently taken place. (5. see Figure 5.4. . (5. 1 ≤ i ≤ m. . Referring to Figure 5. These ideas are developed in (Gale [1967]).3 or Figure 5. b).18).17) also is an optimal solution for (5. we can say that equilibrium prices exist iff marginal productivities of every (variable) resource is ﬁnite. The various convexity conditions are generalizations of the economic hypothesis of non-increasing returns-to-scale. Much of duality theory can be given an economic interpretation similar to that in Section 4. For details see the b) b Appendix.

in analogy with (5. µk ) ≥ 0. For non-economic applications. b) ≥ φ(µ. b). and φ(µ. 5. presented in (Frank [1969]).22) and those optimal solutions of (5.3 Applications. If we let ψ(µ) =sup{f0 (x) − φ(µ. .27) instead of (5. . M (b) π ˆ (5. but quite interesting development along these lines is b. Parts (i) and (iii) of Theorem 2 make duality theory attractive for computation purposes.6) is the graph of the function b → M (ˆ − ˆ b) F (b − ˆ Usually F is chosen from a class of functions φ parameterized by µ = (µ1 . (Banerjee [1971]).6: The surface π supports A at (M (ˆ ˆ ˆ b). f (x) − ˆ ∈ X} . b).1 see (Geoffrion [1970a]) and for a mathematically more elegant treatment see (Rockafellar [1970]). (Greenberg and Pierskalla [1970]). . A ˆ b b Figure 5. such an interpretation to make sense we should have φ(µ. if we have an optimal dual solution λ then the optimal primal solutions are ˆ which also satisfy the feasibility condition (5.27) M (ˆ b) . For more details concerning the topics of 2.26): Maximize f0 (x) − φ(µ. of course. The economic interpretation of (5. b)|x then the dual problem is Minimize ψ(µ) subject to µ ≥ 0 .2. ˜ b) whenever b ≥ ˜ A relatively unnoticed. Then for each ﬁxed µ ≥ 0 we have (5. In particular ˆ from Theorem 2 (iii).19). .5. no such limitation on φ is necessary.18) for λ = λ . f (x) − ˆ b) subject to x ∈ X .27) would be that if the prevailing (non-uniform) price system is φ(µ. ·) then the resources f (x) − ˆ can be bought (or sold) for the amount φ(µ. f (x) − ˆ For b b).2. b) ≥ 0 for b ≥ 0. The following references are pertinent: (Gould [1969]). DUALITY THEORY 65 where F : Rm → R is chosen so that π (in Figure 5. Also see (Arrow and Hurwicz [1960]). Decentralized resource allocation.

18) has fewer constraints. mi (λ) + λ ˆ then the dual problem is Minimize m(λ) .g. NONLINEAR PROGRAMMING the “complementary slackness” condition (5. . the problem corresponding to (5. subject to x i k i If we let mi (λ) = sup{f0 (xi ) − λ f i (xi )|xi ∈ X i }. 1 ≤ i ≤ k. . we need to ﬁnd an optimal dual solution so that we can use Theorem 2(iii).28) For λ ∈ Rm . (5. Suppose that the objective function of the large system 1 k i is additive.66 CHAPTER 5.28) because. but perhaps more importantly the decision problems in (5. i. . + f k (xk ) ≤ ˆ where f i : Rni → Rm are convex functions and ˆ ∈ Rm b b is the vector of available common resources. say xi (λ). Consider the following algorithm.29) may be trivial if the dimensions of xi are small.29) involves fewer constraints. 1 ≤ i ≤ k. subject to λ ≥ 0 .. Furthermore. f i (xi ) ≤ ˆ . for each λ ≥ 0 there is a unique optimal solution of (5. ﬁrst of all. and m(λ) = i=1 (5. . The sub-system has individual constraints of the form xi ∈ X i where xi is a convex set. it is the form f0 (x1 ) + . 1 ≤ i ≤ k . The system is made up of k sub-systems (divisions).e.17) since (5. a multi-divisional ﬁrm).23). Assuming that (5. if k is very large it may be practically impossible to solve (5. 1 ≤ i ≤ k.29) has an optimal solution for every λ ≥ 0. xk are coupled together.29) are decentralized whereas in (5. the sub-systems share some resources in common and this limitation is expressed as f 1 (x1 ) + . 1≤i≤k . (5. Thus we have the decision problem (5. and the decision variable of the ith sub-system is a vector xi ∈ Rni . . and also suppose that (5.28): k Maximize i=1 k i=1 i f0 (xi ) subject to xi ∈ X i . For simplicity suppose that the i f0 . This is useful because generally speaking (5.18) is easier to solve than (5. which decomposes into k separate problems: i Maximize f0 (xi ) − λ f i (xi ) i ∈X .28) whereas (5.29). λ ≥ 0. .28) has an optimal solution and the stability condition is satisﬁed. .29) may be much easier to solve than (5. + f0 (xk ) where f0 : Rni → R are concave functions.29) b. b (5. are strictly concave. Then by Exercise 8 of Section 1. Consider a decision problem in a large system (e. .30) Note that (5. . in fact.28) all the decision variables x1 .19) is k i Maximize f0 (xi ) − λ f i (xi ) − λ ( i=1 f i (xi ) − ˆ b) subject to xi ∈ X i .

si (t) = concentration of BOD of efﬂuent discharge in mg/liter. Hence. and for other decentralization schemes for solving (5. . The discussion in this section is mainly based on (Kendrick.” Therefore. (5. It can be shown that if the step sizes dp are chosen properly. it is enough to study the problem over a 24-hour period. Select λ0 ≥ 0 arbitrary. In this example we are concerned with ﬁnding the optimal balance between costs of waste treatment and costs of high BOD in the stream. t = 1. k Compute ep = i=1 f i (xi (λp )) − ˆ If ep ≥ 0. Step 2. Set λp=1 according to λp+1 = i if ep ≥ 0 λp i i p p ep if ep < 0 λi − d i i where dp > 0 is chosen a priori. . DUALITY THEORY 67 Step 1. namely the biochemical oxygen demand (BOD) which they place on the dissolved oxygen (DO) in the stream. We divide this period into T intervals. .28) see (Geoffrion [1970b]). . During interval t and in area i let zi (t) = concentration of BOD measured in mg/liter.32): zi (t + 1) − zi (t) = −αi zi (t) + ψi−1 zi−1 (t) vi − ψi zi (t) vi + si (t)mi (t) vi . N. For an informal discussion of schemes of pollution control which derive their effectiveness from duality theory see (Solow [1971]). Set p = 0. the quality of the stream improves with the amount of DO and decreases with increasing BOD. . and go to Step 2. . Control of water quality in a stream. We ﬁrst derive the equations which govern the evolution in time of BOD and DO in the n areas of the streams. . The pollutants consist of various materials.5. Step 3.31) s qi (t + 1) − qi (t) = βi (qi − qi (t)) + ψi−1 qi−1 (t) − ψi qi (t) vi vi +αi zi (t) − ηi vi . it is important to treat the efﬂuents before they enter the stream in order to reduce the BOD to concentration levels which can be safely absorbed by the DO in the stream. T .28).7 is a schematic diagram of a part of a stream into which n sources (industries and municipalities) discharge polluting efﬂuents. .31) and (5. The principle of conservation of mass gives us equations (5. It is a well-advertized fact that if the DO drops below a certain concentration. xk (λp )). xp will converge to the optimum solution of (5.) Figure 5. . Set p = p + 1 and return to Step 3. Solve (5. . . and mi (t) = amount of efﬂuent discharge in liters. indeed. .32) . Since the DO in the stream is used to breakdown chemically the pollutants into harmless substances. then life in the stream is seriously threatened. . See (Dorfman and Jacoby [1970]. xp is feasible for (5. . the stream can “die.. T and i = 1. et al.28) and can easily be seen to be b.29) for λ = λp and obtain the optimal solution xp = (x1 (λp ). (5. The ﬂuctuations of BOD and DO will be cyclical with a period of 24 hours. optimal. . but for simplicity of exposition we assume that their impact on the quality of the stream is measured in terms of a single quantity. [1971]). For more detail see (Arrow and Hurwicz [1960]). qi (t) = concentration of DO measured in mg/liter. . t = 1.2.

. Therefore. . The increase in DO is due to various natural oxygen-producing biochemical reactions in the stream and the increase is proportional to (q s − qi ) where q s is the saturation level of DO in the stream. βi is the rate of generation of DO.31) is replaced by zi (t + 1) − zi (t) = −αi zi (t) + ψi zi−1 vi − ψi zi (t) vi + (1−πi (t))si (t)mi (t) vi . . . Also z0 (t). zi−1 qi−1 . Then we face the ¯ . ψi .7: Schematic of stream with efﬂuent discharges. In period t the ith facility treats mi (t) liters of efﬂuent with a BOD concentration si (t) mg/liter of which the facility removes a fraction πi (t). We further assume that f is convex. qi (1). . agricultural. Let q be the minimum acceptable DO concentration and let z be the maximum permissible BOD concentration.. recreational). This decay occurs by combination of BOD and DO. q0 (t) which are the concentrations immediately upstream from area 1 are assumed known. the cost in period t will be fi (πi (t). and the disutility caused by a decrease in the water quality varies with the user. . (5. Hence. ηi . αi is the rate of decay of BOD per interval. They may vary with the time interval t. instead of attempting to quantify these costs let us suppose that some minimum water quality standards are set. zN qN (1 (1 − πi )si − πi+1 )si+1 (1 − πN )sN si si−a si si+1 sN Figure 5. The costs associated with increased amounts of BOD and reduced amounts of DO are much more difﬁcult to quantify since the stream is used by many institutions for a variety of purposes (e.. Finally. Here. The cost of waste treatment can be readily identiﬁed. The vi .. ψi = volume of water which ﬂows from area i to are i + 1 in each period measured in liters.g. i = 1. vi = volume of water in area i measured in liters. q s are parameters of the stream and are assumed known. Now suppose that the waste treatment facility in area i removes in interval t a fraction πi (t) of the concentration si (t) of BOD. ηi is the DO requirement in the bottom sludge. mi (t)) where the function must be monotonically increasing in all of its arguments.68 CHAPTER 5. municipal. . αi . si (t). N are assumed known. the initial concentrations zi (1). industrial. Finally. NONLINEAR PROGRAMMING direction of ﬂow 0 z0 q0 (1 − π1 )si given 1 z1 q1 1 − πi−1 i−1 i zi qi i+1 zi+1 qi+1 N N +1 . Then (5...33) We now turn to the costs associated with waste treatment and pollution.

.36) and (5. . . N . . . t = 1. the resulting amount of waste treatment is carried out at the minimum expenditure of resources (i. T.37) ∗ such that {πi (t)} is also an optimal solution of (5. N. t = 1. Note that the coefﬁcients of the matrix must be non-negative because an increase in any component of r cannot decrease the BOD levels and cannot increase the DO levels. . . and −qi (t) ≤ −q . i = 1.33) for w and obtain w = b + Ar . . si (t). mi (t)) − λ∗ (b + Ar − w) subject to 0 ≤ πi (t) ≤ 1. If we let p∗ = A λ∗ ≥ 0. Suppose that all the treatment facilities are in the control of a single public agency.32) and (5. . .36) subject to b + Ar ≤ w . . i = 1. . . si (t). i = 1. i = 1. . zi (t) ≤ z . . i = 1. T . will be an optimal solution of (5. Furthermore. t = 1. of the problem: Maximize − i t fi (πi (t). . ¯ 0 ≤ πi (t) ≤ 1 . . T. To see this let wi (t) = (zi (t).35) we can rewrite (5.32). . . N . But if ¯ there is no such centralized agency. It should be clear from the duality theory that the answer is in the afﬁrmative. . . mi (t)) (5. furthermore. .34) as follows: Maximize − i t fi (πi (t). N . . . (5. . . . (5. wN (t)). (5. mi (t)) (5. z ) and it does this at a minimum cost it will ¯ solve the NP (5. ¯ ¯ ∗ ≥ 0. furthermore. N . Then we can solve (5. . t = 1. . Let the minimum cost be m(q. then the resulting water quality will be acceptable and. Then assuming that the agency is required to maintain the standards (q. . . DUALITY THEORY following NP: N T 69 Maximize − i=1 t=1 fi (πi (t). and an optimal solution By the duality theorem there exists a 2N T -dimensional vector λ ∗ πi (t). T . .34)). i = 1. . . . . T. .34) subject to (5.2. then the individual polluters may not (and usually do not) have any incentive to cooperate among themselves to achieve these standards. N . t = 1. . T. and let w = (w(1).e. On the other hand.. . The question we now pose is whether there exist tax rates such that if each individual polluter minimizes its own total cost (i. . . zi (t) ≤ z on every polluter since the ¯ pollution levels in the ith area depend upon the pollution levels on all the other areas lying upstream. the optimal values of (5. it may be economically and politically acceptable to tax individual polluters in proportion to the amount of pollutants discharged by the individual. . . .37) are equal. −qi (t)) . . let w(t) = (w1 (t). it does not make sense to enforce legally a minimum standard qi (t) ≥ q. . . t = 1. .5. .. . cost of waste treatment + tax on remaining pollutants). where the 2N T -dimensional vector w has its components equal to −q or z in the obvious manner. ¯ 0 ≤ πi (t) ≤ 1 .36) and.34) and arrive at an optimal solution. and r is the NT-dimensional vector with components (1 − πi (t))si (t)mi (t).35) where the matrix A and the vector b depend upon the known parameters and initial conditions. . Using (5. .e.33). . and we write the components of p∗ as p∗ (t) to match i . w(t)). si (t). . . . z ).

c ∈ Rn .3. (λ∗ ) (Ax∗ − b) = 0 . µ∗ ∈ Rn .3 Quadratic Programming An important special case of NP is the quadratic programming (QP) problem: Maximize c x − 1 x P x 2 subject to Ax ≤ b. . x ≥ 0 y ≥ 0. t = 1. We noted earlier that the quality standard (q. ¯ i i then the change in the minimum cost necessary to achieve the new standard will be approximately λq∗ (t)∆qi (t) + λz∗ (t)∆zi (t). (5. Then we must apply Phase II to the LP: m n Maximize − i=1 zi − j=1 ξj .43) Suppose we try to solve (5. . and (5. i = 1. (5.2). N .40) we can see that x∗ is optimal for (5. µ x = 0 .41). T . . µ∗ ) to (5. i Before we leave this example let us note that the optimum dual variable or shadow price λ∗ plays an important role in a larger framework. Theorem 1: A vector x∗ ∈ Rn is optimal for (5. λ∗ ≥ 0. p∗ (t) is optimum tax per mg of BOD in area i during period t. since P is positive semi-deﬁnite it follows from Exercise 6 of Section 1. si (t). x∗ ≥ 0 c − P x∗ = A λ∗ − µ∗ . z ) ¯ was somewhat arbitrary.2. so that the sufﬁciency of these conditions follows from Theorem 4 of 1. µ∗ ≥ 0 . b ∈ Rm are ﬁxed. (5. . This estimate can now serve as a basis in making a beneﬁts/cost i i analysis of the proposed new standard. (5. µ ≥ 0 .2 that f0 : x → c x − 1/2 x P x is a concave function. A is a ﬁxed m × n matrix and P = P is a ﬁxed positive semi-deﬁnite matrix. hence the necessity of these conditions follows from Theorem 2 of 1. λ∗ . such that Ax∗ ≤ b. (5.42) (5. mi (t)) − p∗ (t)(1 − πi (t))si (t)mi (t) i 0 ≤ πi (t) ≤ 1 . ♦ From (5. .38) Thus. x ≥ 0 .39) iff there exist λ∗ ∈ Rm .39) iff there is a solution (x∗ . On the other hand.2. .λ y = 0 .43): Ax + Im Y = b −P x − A λ + In µ = −c .70 CHAPTER 5. CQ is satisﬁed. 5.37) is equivalent to the set of N T problems: Maximize − fi (πi (t). NONLINEAR PROGRAMMING with the components (1 − πi (t))si (t)mi (t) of r we can see that (5. λ ≥ 0. If the corresponding components of λ∗ are λq∗ (t) and λz∗ (t).3. y ∗ .42).40) Proof: By Lemma 3 of 1.41) and (5. . .41) (5. Now suppose it is proposed to change the standard in the ith area during period t to q + ∆qi (t) and z + ∆zi (t). (µ∗ ) x∗ = 0 .39) where x ∈ Rn is the decision variable .42) by Phase I of the Simplex algorithm (see 4.

−1 ≤ hj ≤ 1 . (5.39). (5.44) is 0. (5.43) we see that at most (n + m) components of (ˆ. The algorithm will stop in a ﬁnite number of steps at an ˆ optimal basic feasible solution (ˆ. and (5. . Cullum. If it not possible to remove the zi and ξj from the basis. µ) are non-zero.43).41). that b ≥ 0 and −c ≥ 0. COMPUTATIONAL METHOD subject to Ax + Im y +z =b −P x − A λ + In µ + ξ = −c x ≥ 0. m . Furthermore.46) . (We have assumed. (5. then there is an optimal basic feasible solution of (5. λ.42). x x −ψ(ˆ)(h) + fi (ˆ)fix h ≤ 0 . ξ) of (5. y . . ♦ This lemma suggests that we can apply the Simplex algorithm of 4. The above algorithm is due to Wolfe [1959]. Let Ω ⊂ Rn denote the set of feasible solutions. . 159 ff). f1 (ˆ) + f1x (ˆ)h. 0 ≤ i ≤ m.41).41).41).3. (5.44).41).1 will x ˆ ˆ ˆ also prove this lemma.42) have a solution then the maximum value in (5.43). if a variable yi is currently in the basis. y . (5.3. If z = 0 or ξ ˆ ˆ solution to (5.42). µ be a solution of (5. λ. For a proof of this result as well as for a generalization of the algorithm which permits positive semi-deﬁnite P see (Cannon. If z = 0 and ξ = 0 then (ˆ. do not consider µj as a candidate for entry into the basis.4. p. For x ∈ Ω deﬁne the function ψ(ˆ) : Rn → R by ˆ x ψ(ˆ)(h) = max{−f0x (ˆ)h. µ. λ ≥ 0. y ≥ 0. µ ≥ 0.42). . and Polak [1970].42). and (5. z ≥ 0. fi : Rn → R.43).5. y . λ. 5. x x 1 ≤ i ≤ m . λ. y . .45) where x ∈ Rn . in order to obtain a solution of (5. stop. and there is no feasible solution of (5. ξ ≥ 0.42). ξ = −c. Lemma 1: If (5.43) have a solution. from (5. We have the following result. and (5.44).2 to solve (5. starting with the basic feasible solution z = b. ˆ Proof: Let x. and (5.) If (5.44) starting with a basic feasible solution z = b. Then x. (5.41). µ) solve x ˆ ˆ ˆ ˆ ˆ ˆ x ˆ ˆ ˆ ˆ = 0. Step 2 of the Simplex algorithm must be modiﬁed as follows to satisfy (5. then there is no (5. µ. without loss of generality.42) and (5. . . z = 0.44) which is also a solution f (5. However. y . fm (ˆ) + fmx (ˆ)h}. .39). Theorem 2: Suppose P is positive deﬁnite. (5. But then a repetition of the proof of Lemma 1 of 4.43): If a variable xj is currently in the basis. and (5.45). ξ = −c. are differentiable. The behavior of the algorithm is summarized below. λ. x x x x x x Consider the problem: Minimize ψ(ˆ)(h) x subject to − ψ(ˆ)(h) − f0x (ˆ)h ≤ 0 . do not consider λi as a candidate for entry into the basis.41) and (5.44). Maximize f0 (x) subject to fi (x) ≤ 0.4 Computational Method We return to the general NP (5. 71 (5.43) and x is an optimal solution of (5. i = 1. z . 1 ≤ j ≤ n .43). ξ = 0 is ˆ ˆ ˆ ˆ ˆ ˆ ˆ ˆ ˆ an optimal solution of (5. (5.42).

subject to (xk + µh(xk )) ∈ Ω. For this reason h(xk ) is called a (desirable) feasible direction. .) . . generated by the algorithm. (See Figure 5. set k = 0. Maximize f0 (xk + µh(xk )) . otherwise go to Step ˆ 3. . Call h(ˆ) an optimum solution of (5. .46) for x = xk and obtain h0 (xk ). Set xk+1 = xk + µ(xk )h(xk ). which is dense in Ω(x0 ). Theorem 1: Suppose that the set Ω(x0 ) = {x|x ∈ Ω. Step 1. Compute an optimum solution µ(xk ) to the one-dimensional problem. . .8. x1 . (Note that by Exercise 1 of 4. stop.46) can be solved as an LP. For a proof of this result and for more efﬁcient algorithms the reader is referred to (Polak [1971]). f3 (xk ) xk h(xk ) f2 (xk ) f1 = 0 Ω f3 = 0 Figure 5. Step 4. Step 2. and has a non-empty interior.) The following algorithm is due to Topkis and Veinott [1967]. Remark: If h0 (xk ) < 0 in Step 2. The performance of the algorithm is summarized below. xk .8: h(xk ) is a feasible direction.46) and let h0 (ˆ) = ψ(ˆ)(h(ˆ)) be the minimum value atx x x x tained. . NONLINEAR PROGRAMMING f0 (xk )f0 (x) = F0 (x∗ ) > f0 (xk ) f1 (xk ) f0 (x) = f0 (xk ) f2 = 0 . µ ≥ 0 . and go to Step 2. f0 (x) ≥ f0 (x0 )} is compact.72 CHAPTER 5. Find x0 ∈ Ω. Then the Kuhn-Tucker conditions are satisﬁed at x∗ .1 (5.5. and go to Step 4. Solve (5. Let x∗ be any limit point of the sequence x0 . Step 3. 1 ≤ i ≤ m. and fi (xk )+ fix (xK )h(xk ) < 0. If h0 (xk ) = 0. . then the direction h(xk ) satisﬁes f0x (xk )h(xk ) > 0. set k = k + 1 and return to Step 2. h(xk ).

and θ such that m (5. and the fact that (λ0 .7 of Section 2 are based on the following extremely important theorem (see Rockafeller [1970]). Then there exists λ ∈ Rn . λm ) = 0. b| In R1+m consider the sets F = {(r. . λm ) = 0. . r ≤ M (b)} . . . there exist (λ0 .48) m b λiˆi ≤ θ. G are disjoint. Separation theorem for convex sets. b) (5. G are convex.50) . so that there exist (λ0 . there exists ε > 0 such that b b ∈ B whenever |b − ˆ < ε . .47) implies that F ∩ G = φ. b) b| In R1+m consider the sets F = {(r. b)|b ∈ B.49) we can see that m i=1 b λiˆi ≥ θ.49) can hold m only if λ0 > 0. b)} It is easy to check that F. F. b) ∈ F . Let F. r > K|b − ˆ .49) λ0 r + i=1 b λiˆi ≥ θ . b)|b ∈ Rm . . b|} G = {(r. λ0 ♦ Proof of Lemma 7: Since ˆ is in the interior of B.5 Appendix The proofs of Lemmas 4. Evidently. Hence. and (5. . and θ ∈ R such that λ g ≤ θ for all g ∈ G λ f ≥ θ for all f ∈ F . i=1 λ0 r + From the deﬁnition of F . whereas from (5. i=1 so that i=1 b λiˆi = θ. G are convex and F ∩ G = φ.48) we get m 1 λ0 [θ m M (b) − M (ˆ ≤ b) − i=1 λ i bi ] = i=1 (− λi )(bi − ˆ b). . λ = 0.48) λi bi ≥ θ for (r. r ≤ M (b) − M (ˆ . for r > M (ˆ . Proof of Lemma 4: Since M is stable at ˆ there exists K such that b M (b) − M (ˆ ≤ K|b − ˆ for all b ∈ B . . ˆ > M (ˆ b)|r b} G = {(r.5. and θ such that m (5. it can be veriﬁed that (5. b) ∈ G . . Also from (5. b)|b ∈ B. But then from (5. G be convex subsets of Rn such that the relative interiors of F. .47) λ0 r + i=1 m λi bi ≤ θ for (r. . APPENDIX 73 5. λm ) = 0.5. (5.

51) From (5.50).50) and (5. .74 m CHAPTER 5. NONLINEAR PROGRAMMING λ0 r + i=1 b λiˆi ≤ θ . (5. b) ∈ G .(5.49). .51) imply λ0 > 0. (− i=1 λi )(bi − ˆi ) . . for (r.51) we get λ0 M (ˆ + b) so that (5. and the fact that (λ0 . λm ) = 0 we can see that (5. From (5.52) implies M (b) ≤ (ˆ + b) m m i=1 b λiˆi = θ . . b λ0 ♦ .

See Figure 6. 0. assumed proportional to rate of fuel ejection. M ).1 Examples The trajectory of a vertical sounding rocket is controlled by adjusting the rate of fuel ejection which generates the thrust force. ρ(x1 ) is a friction coefﬁcient depending on atmospheric 2 density which is a function of x1 . x3 (0)) = (0. it is desired that the rocket be at a position as high above the ground as possible. The “dot” denotes differentiation with respect to t. the rate of fuel ejection. At a prescribed ﬁnal time tf . Speciﬁcally suppose that the equations of motion are given by (6. At time 0 we assume that (x1 (0). that is. the rocket is on the ground. The decision variable at ˙ time t is u(t). x3 (t) is the weight of the rocket (= weight of remaining fuel) at time t. ˙ CT x3 (t) u(t) (6.1) where x1 (t) is the height of the rocket from the ground at time t. 6. namely: inertia = x3 x1 = x3 x2 .1). gravitational force = gx3 with g assumed constant. with initial fuel of weight M . These equations can be derived from the force equations under the assumption that there are four forces acting on the rocket. x1 (t) = x2 (t) ˙ x2 (t) = − xCD ρ(x1 (t))x2 (t) − g + ˙ 2 3 (t) x3 (t) = −u(t) . In the ﬁrst section we give two examples. Thus.Chapter 6 SEQUENTIAL DECISION PROBLEMS: DISCRETE-TIME OPTIMAL CONTROL In this chapter we apply the results of the last two chapters to situations where decisions have to be made sequentially over time. x2 (t) is the (vertical) speed at time t.1. and in Section 2 we derive the main result. the 75 . x2 (0). drag ¨ ˙ force = CD ρ(x1 )x2 where CD is constant. and thrust force CT x3 . A very important class of problems where such situations arise is in the control of dynamical systems. at rest.

76

CHAPTER 6. DISCRETE-TIME OPTIMAL CONTROL

decision problem can be formalized as (6.2). Maximize x1 (tf ) subject to x(t) = f (x(t), u(t)), 0 ≤ t ≤ tf ˙ x(0) = (0, 0, M ) u(t) ≥ 0, x3 (t) ≥ 0, 0 ≤ t ≤ tf ,

(6.2)

where x = (x1 , x2 , x3 ) , f : R3+1 → R3 is the right-hand side of (6.1). The constraint inequalities u(t) ≥ 0 and x3 (t) ≥ 0 are obvious physical constraints.

x3 x1 = inertia ¨

CD ϕ(x1 )x2 = drag 2

gx3 = gravitational force ˙ CR x3 = thrust Figure 6.1: Forces acting on the rocket. The decision problem (6.2) differs from those considered so far in that the decision variables, which are functions u : [0, tf ] → R, cannot be represented as vectors in a ﬁnite-dimensional space. We shall treat such problems in great generality in the succeeding chapters. For the moment we assume that for computational or practical reasons it is necessary to approximate or restrict the permissible function u(·) to be constant over the intervals [0, t1 ), [t1 , t2 ), . . . , [tN −1 , tf ), where t1 , t2 , . . . , tN −1 are ﬁxed a priori. But then if we let u(i) be the constant value of u(·) over [ti , ti+1 ), we can reformulate (6.2) as (6.3): Maximize x1 (tN )(tN = tf ) subject to x(ti+1 ) = g(i, x(ti ), u(i)), i = 0, 1, . . . , N − 1 x(t0 ) = x(0) = (0, 0, M ) u(i) ≥ 0, x3 (ti ) ≥ 0, i = 0, 1, . . . , N .

(6.3)

In (6.3) g(i, x(t1 ), u(i)) is the state of the rocket at time ti+1 when it is in state x(ti ) at time ti and u(t) ≡ u(i) for ti ≤ t < ti+1 . As another example consider a simple inventory problem where time enters discretely in a natural fashion. The Squeezme Toothpaste Company wants to plan its production and inventory schedule for the coming month. It is assumed that the demand on the ith day, 0 ≤ i ≤ 30, is d1 (i) for

6.2. MAIN RESULT

77

their orange brand and d2 (i) for their green brand. To meet unexpected demand it is necessary that the inventory stock of either brand should not fall below s > 0. If we let s(i) = (s1 (i), s2 (i)) denote the stock at the beginning of the ith day, and m(i) = (m1 (i), m2 (i)) denote the amounts manufactured on the ith day, then clearly s(i + 1) + s(i) + m(i) − d(i) , where d(i) = (d1 (i), d2 (i)) . Suppose that the initial stock is s, and the cost of storing inventory s ˆ for one day is c(s) whereas the cost of manufacturing amount m is b(m). The the cost-minimization decision problem can be formalized as (6.4):
30

Maximize
i=0

(c(s(i)) + b(m(i))) (6.4)

subject to s(i + 1) = s(i) + m(i) − d(i), 0 ≤ i ≤ 29 s(0) = s ˆ s(i) ≥ (s, s) , m(i) ≥ 0, 0 ≤ i ≤ 30 .

Before we formulate the general problem let us note that (6.3) and (6.4) are in the form of nonlinear programming problems. The reason for treating these problems separately is because of their practical importance, and because the conditions of optimality take on a special form.

6.2 Main Result
The general problem we consider is of the form (6.5).
N −1

Maximize
i=0

f0 (i, x(i), u(i))

subject to dynamics : x(i + 1) − x(i) = f (i, x(i), u(i)), i = 0, . . . , N − 1 , initial condition: q0 (x(0) ≤ 0, g0 (x(0)) = 0 , ﬁnal condition: qN (x(N )) ≤ 0, gN (x(N )) = 0 , state-space constraint: qi (x(i)) ≤ 0, i = 1, . . . , N − 1 , control constraint: hi (u(i)) ≤ 0, i = 0, . . . , N − 1 .

(6.5)

Here x(i) ∈ Rn , u(i) ∈ Rp , f0 (i, ·, ·) : Rn+p → R, f (i, ·, ·) : Rn+p → Rn , qi : Rn → Rmi , gi : Rn → R i , hi : Rp → Rsi are given differentiable functions. We follow the control theory terminology, and refer to x(i) as the state of the system at time i, and u(i) as the control or input at time i. We use the formulation mentioned in the Remark following Theorem 3 of V.1.2, and construct the Lagrangian function L by L(x(0), . . . , x(N ); u(0), . . . , u(N − 1); p(1), . . . , p(N ); λ0 , . . . , λN ; α0 , αN ; γ 0 , . . . , γ N −1 )

78
N −1 N −1

CHAPTER 6. DISCRETE-TIME OPTIMAL CONTROL
=
i=0 N i i=0 0

f0 (i, x(i), u(i)) −
i=0

(p(i + 1)) (x(i + 1) − x(i) − f (i, x(i), u(i)))+
N −1 N

(λ ) qi (x(i)) + (α ) g0 (x(0)) + (α ) gN (x(N )) +
i=0

(γ i ) hi (u(i))

.

Suppose that CQ is satisﬁed for (6.5), and x∗ (0), . . . , x∗ (N ); u∗ (0), . . . , u∗ (N − 1), is an optimal solution. Then by Theorem 2 of 5.1.2, there exist p∗ (i) in Rn for 1 ≤ i ≤ N, λi∗ ≥ 0 in Rmi for 0 ≤ i ≤ N, αi∗ in R i for i = 0, N, and γ i∗ ≥ 0 in Rsi for 0 ≤ i ≤ N − 1, such that (A) the derivative of L evaluated at these points vanishes, and (B) λi∗ qi (x∗ (i)) = 0 for 0 ≤ i ≤ N , γ i∗ hi (u∗ (i)) = 0 for 0 ≤ i ≤ N − 1 . We explore condition (A) by taking various partial derivatives. Differentiating L with respect to x(0) gives f0x (0, x∗ (0), u∗ (0)) − {−(p∗ (1)) − (p∗ (1)) [fx (0, x∗ (0), u∗ (0))] +(λ0∗ ) [q0x (x∗ (0))] + (α0∗ ) [g0x (x∗ (0))]} = 0 , or p∗ (0) − p∗ (1) = [fx (0, x∗ (0), u∗ (x))] p∗ (1) +[f0x (0, x∗ (0), u∗ (0))] − [q0x (x∗ (0))] λ0∗ , where we have deﬁned p∗ (0) = [g0x (x∗ (x))] α0∗ . Differentiating L with respect to x(i), 1 ≤ i ≤ N − 1, and re-arranging terms gives p∗ (i) − p∗ (i + 1) = [fx (i, x∗ (i), u∗ (i))] p∗ (i + 1) +[f0x (i, x∗ (i), u∗ (i))] − [qix (x∗ (i))] λi∗ . Differentiating L with respect to x(N ) gives, p∗ (N ) = −[gN x (x∗ (N ))] αN∗ − [qN x (x∗ (N ))] λN∗ . It is convenient to replace αN∗ by −αN∗ so that the equation above becomes (6.9) p∗ (N ) = [gN x (x∗ (N ))] αN∗ − [qN x (x∗ (N ))] λN∗ . Differentiating L with respect to u(i), 0 ≤ i ≤ N − 1 gives [f0u (i, x∗ (i), u∗ (i))] + [fu (i, x∗ (i), u∗ (i))] p∗ (i + l) − [hiu (u∗ (i))] γ i∗ = 0 . We summarize our results in a convenient form in Table 6.1 Remark 1: Considerable elegance and mnemonic simpliﬁcation is achieved if we deﬁne the Hamiltonian function H by (6.10) (6.9) (6.8) (6.7)

(6.6)

. such that f0 (i. . N − 1 p∗ (i) − p∗ (i + 1) = [fx (i. x∗ (N ). . . . . . . u∗ (i))] + [fu (i. . N − 1 hi (u∗ (i)) ≤ 0 [f0u (i. λ0∗ . N − 1 x(i + 1) − x(i) = f (i. γ N −1∗ . . u∗ (i)] − [qix (x∗ (i)] γ i∗ transversality conditions: p∗ (0) = [g0x (x∗ (0))] α0∗ λ0∗ ≥ 0. x(i). u(i)) Table 6. . . . . (λi∗ ) qi (x∗ (i)) = 0 γ i∗ ≥ 0 (γ i∗ ) hi (u∗ (i) = 0 79 . αN∗ . . p∗ (i1 ) = [hiu (u∗ (i))] γ i∗ adjoint equations: i = 0. p∗ (N ) = [gN x (x∗ (N ))] αN∗ − [qN x (x∗ (N ))] λN∗ (λN∗ ) qN (x∗ (N )) = 0 λi∗ ≥ 0. N − 1 qi (x∗ (i)) ≤ 0 control constraint: i = 0. . α0∗ . x∗ (i). gN (x∗ (N )) = 0 state space constraint: i = 1. . g0 (x∗ (0)) = 0 ﬁnal conditions: qN (x∗ (N )) ≤ 0. u∗ (0). . . .2. . γ 0∗ .6. MAIN RESULT Suppose x∗ (0). . u∗ (i)] p∗ (i + 1) +[f0x (i. x(i). . . u∗ (N − 1) maximizes N −1 i=0 then there exist p∗ (N ).1: initial condition: q0 (x∗ (0)) ≤ 0. λN∗ . u(i)) subject to the constraints below dynamics: i = 0. x∗ (i)u∗ (i))] . . . . (λ0∗ ) q0 (x∗ (0)) = 0 λN∗ ≥ 0. . . x∗ (i). x∗ (i). . . .

1 are also sufﬁcient. Suppose q0 ≡ 0. x∗ (i). u) .80 CHAPTER 6. and the adjoint equations (6. (i). Remark 3: If the f0 (i.13) (6. then CQ is satisﬁed. p∗ (i + 1))] − [qix (x∗ (i))] λi∗ .5) are linear. x∗ (i). . (6. x∗ (i). u∗ (i). u.7). u) + p f (i. p) = f0 (i. x∗ (i). which has for it adjoint the system r(i) − r(i + 1) = [fx (i.10) becomes [hiu (u∗ (i))] γ i∗ = [Hu (i. x∗ (i. DISCRETE-TIME OPTIMAL CONTROL H(i. x. but note that these 2n boundary conditions are mixed.6). u∗ (i).12). p∗ (N ) = [gN x (x(N ))] αN∗ which means that p∗ (0) and p∗ (N ) are respectively orthogonal or transversal to the initial and ﬁnal surfaces. which describe surfaces in Rn . x∗ .5). The dynamic equations then become x∗ (i + 1) − x∗ (i) = [Hp (i.13) that u∗ (i) is an optimal solution of Maximize H(i. p∗ (i + 1))] . (6. Remark 2: If we linearize the dynamic equations about the optimal solution we obtain δx(i + 1) − δx(i) = [fx (i. u∗ (i).12) Since the homogeneous part of the linear difference equations (6.8) the adjoint equations. u∗ (i))] r(i + 1) . p∗ (i + 1))] . 0≤i≤N −1. x. 0 ≤ i ≤ N − 1 . (6. ·. For this reason the result is sometimes called the maximum principle. 0≤i≤N −1 . p∗ (i + 1)). Thus..e. (6. u∗ (i))]δx(i) + [fu (i. whose homogeneous part is z(i + 1) − z(i) = [fx (i. x∗ (i). Furthermore. we have a total of 2n boundary conditions for the 2n-dimensional system of difference equations (6. whereas (6. (6.7). Remark 4: The conditions (6.8) is (6.13). subject to hi (u) ≤ 0 . gN (x(N )) = 0.11) p∗ (i) − p∗ (i + 1) = [Hx (i. u. and the necessary conditions of Table 6. some of them refer to the initial time 0 and the rest refer to the ﬁnal time. Conditions (6. u∗ (i).6). x∗ (i). i. we note that in this case the initial and ﬁnal conditions specify ( 0 + n ) conditions whereas the transversality conditions specify (n − 0 ) + (n − n ) conditions. x. and the p∗ (i) are called adjoint variables. in this case we see from (6. we call (6.9) are called transversality conditions for the following reason.8) become (6. u∗ (i))]z(i) . x∗ (i).6) and (6. (6. ·) are concave and the remaining function in (6. Furthermore.9) become respectively p∗ (0) = [g0x (x∗ (0))] α0∗ . qN ≡ 0. so that the initial and ﬁnal conditions read g0 (x(0)) = 0. u∗ (i))]δu(i) .

. j=1 subject to x(i + 1) − x(i) = Ax(i) + Bu(i). ˆ ˆ u(i) ∈ Rp . ˆ and P = P is positive deﬁnite.6. show that the optimal solution is unique and can be obtained by solving a 2n-dimensional linear difference equation with mixed boundary conditions. x(N ) are ﬁxed. x(N ) = x(N ) . |u(i))j | ≤ 1. 0 ≤ i ≤ N − 1 can be transformed into a linear programming problem. x(0) is ﬁxed. 1 2 N −1 N −1 81 x(i) Qx(i) + where x(i) ∈ Rn . Q = Q is positive semi-deﬁnite. 0 ≤ i ≤ N − 1 x(0) = x(0). A and B are constant matrices. A and B are as ˆ ˆ in Exercise 1. 1 ≤ j ≤ p. 0 ≤ i ≤ N − 1 .   N −1 i=0 Minimize  P |(u(i))j | . Exercise 2: Show that the minimal fuel problem. Maximize 1 u(i) P u(i) 2 i=0 i=0 subject to x(i + 1) − x(i) = Ax(i) + Bu(i).2. MAIN RESULT Exercise 1: For the regulator problem. Here x(0). 0 ≤ i ≤ N − 1 x(0) = x(0). ˆ u(i) ∈ Rp .

DISCRETE-TIME OPTIMAL CONTROL .82 CHAPTER 6.

U denotes the set of all admissible controls.Chapter 7 SEQUENTIAL DECISION PROBLEMS: CONTINUOUS-TIME OPTIMAL CONTROL OF LINEAR SYSTEMS We will investigate decision problems similar to those studied in the last chapter with one (mathematically) crucial difference. and to be piecewise continuous.1): x(t) = A(t)x(t) + B(t)u(t).1 The Linear Optimal Control Problem We consider a dynamical system governed by the linear differential equation (7. ˙ (7. 7. x0 ∈ Rn be ﬁxed and let tf ≥ t0 be a ﬁxed time. ˙ where x(t) ∈ Rn and u(t) ∈ Rp are respectively the state and control of the system at time t. u(t)) . In Section 2 we study more general boundary conditions.1) Here A(·) and B(·) are n × n. The control u(·) is constrained to take values in a ﬁxed set Ω ⊂ Rp . x(t). We are concerned with the 83 .and n × p-matrix valued functions of time. we assume that they are piecewise continuous functions. Deﬁnition: A piecewise continuous function u : [t0 . The general nonlinear case is deferred to the next chapter. t ≥ t0 . A choice of control has to be made at each instant of time t where t varies continuously over a ﬁnite interval. In Section 1 we present the general linear problem and study the case where the initial and ﬁnal conditions are particularly simple. The evolution in time of the state of the systems to be controlled is governed by a differential equation of the form: x(t) = f (t. Let c ∈ Rn . ∞) → Ω will be called an admissible control. To understand the main ideas and techniques of analysis it will prove proﬁtable to study the linear case ﬁrst.

and the boundary condition Φ(t. and the control u(·) is applied.. t1 . u) denote the state of (7. ∞) → Ω. if a time t1 it is in state z. x0 ). and c x∗ ≥ c x for all x ∈ K . We call K the reachable set.1. and let x∗ ∈ K. control constraint: u(·) ∈ U . We say that c is the outward normal to a hyperplane supporting K at x∗ if c = 0. The next result is well-known. Lemma 2: Suppose c = 0. z) is convex even if Ω is not convex (see Neustadt [1963]). τ )B(τ )u(τ )dτ .) Proof: Clearly (i) is implied by (ii) because if x∗ (tf ) is in the interior of K there is δ > 0 such that (x∗ (tf ) + δc) ∈ K. t0 . z. u)|u ∈ U} . The next result gives a geometric characterization of the optimal solutions of (2). Let u∗ (·) ∈ U and let x∗ (t) = φ(t. show that U is a convex set. t1 . u) = Φ(t2 . (It is a deep result that K(t2 . Deﬁnition: Let Φ(t. be the transition-matrix function of the homogeneous part of (7. Thus.) Deﬁnition: Let K ⊂ Rn . t0 ≤ t ≤ tf . (See Figure 7. ﬁnal condition: x(tf ) ∈ Rn . CHAPTER 7. t0 ≤ τ ≤ t ≤ tf . z. (See Desoer [1970]. K(t2 . t1 . tf ] → Rp . t1 . t) ≡ In . i. and (ii) c is the outward normal to a hyperplane supporting K at x∗ . subject to dynamics: x(t) = A(t)x(t) + B(t)u(t) .2). t0 . t1 . τ ). z. provided we include in U any measurable function u : [t0 . Then u∗ is an optimal solution of (2) iff (i) x∗ (tf ) is on the boundary of K = K(tf . and any t0 ≤ t1 ≤ t2 ≤ tf let φ(t2 .1). Exercise 1: (i) Assuming that Ω is convex. t1 )z + t2 t1 Φ(t2 . u∗ ). (7. for any z ∈ Rn .) Lemma 1: φ(t2 . t1 . x0 . z) = {φ(t2 . z) is a convex set. τ ) = A(t)Φ(t. Φ satisﬁes the differential equation ∂Φ ∂t (t. ˙ initial condition: x(t0 ) = x0 . CONTINUOUS-TIME LINEAR OPTIMAL CONTROL Maximize c x(tf ). (ii) Assuming that U is convex show that K(t2 .84 decision problem (7. but then . z) is the set of states reachable at time t2 starting at time t1 in state z and using admissible controls. τ ) .1) at time t2 . (ii) Let K(t2 .2) Deﬁnition: (i) For any piecewise continuous function u(·) : [t0 . t1 .e.

6) . THE LINEAR OPTIMAL CONTROL PROBLEM x3 c 85 x∗ (tf ) c x2 K π ∗ = {x|c x = c x∗ (tf )} x1 Figure 7. ♦ The result above characterizes the optimal control u∗ in terms of the ﬁnal state x∗ (tf ).4): adjoint equation: p∗ (t) = −A (t)p∗ (t) . from the deﬁnition of K it follows immediately that u∗ is optimal iff c x∗ (tf ) ≥ c x for all x∈K. τ )B(τ )u(τ )dτ (7.1. τ )B(τ )u(τ )dτ ] .4) (7. t0 ≤ t ≤ tf . τ )B(τ )u∗ (τ )dτ ] t ≥ (p∗ (tf )) [Φ(tf . τ )B(τ )u (τ )dτ t ≥ t0f (p∗ (tf )) Φ(tf . c (x∗ (tf ) + δc) = c x∗ (tf ) + δ|c|2 > c x∗ (tf ) . t0 )x0 + t0f Φ(tf .1: c is the outward normal to π ∗ supporting K at x∗ (tf ) . ˙ ﬁnal condition: p∗ (tf ) = c . except possibly for a ﬁnite set. x0 . t0 ≤ t ≤ tf .6). Let p∗ (t) be the solution of (7.3) and (7.3) (7. tf ∗ ∗ t0 (p (tf )) Φ(tf . u∗ ). tf ]. t (7. The beauty and utility of the theory lies in the following result which translates this characterization directly in terms of u∗ .7. Then u∗ (·) is optimal iff (p∗ (t)) B(t)u∗ (t) = sup{(p∗ (t)) B(t)v|v ∈ Ω} . Theorem 1: Let u∗ (·) ∈ U and let x∗ (t) = φ(t.5) which is equivalent to (7. Finally. t0 . Proof: u∗ (·) is optimal iff for every u(·) ∈ U (p∗ (tf )) [Φ(tf . t0 )x0 + t0f Φ(tf . for all t ∈ [t0 .

it follows that there exists δ > 0 such that (p∗ (t)) B(t)u∗ (t) < (p∗ (t)) B(t)v. u∗ (t). x0 ) and p∗ (t) is the normal to a hyperplane supporting K(t. This condition is known as the maximum principle. t0 .7). via the adjoint differential equation.6) is equivalent to (7. Remark 1: The geometric meaning of (7. t) so that (7. then (7. p)|u ∈ Ω}. for |t − t∗ | < δ . x∗ (t1 )).9) Exercise 2: Prove Corollary 1.7) and the sufﬁciency of (7. x∗ (t). we see that if u∗ (·) is optimal. t∗ ∈ D. t0 . u. x∗ (t). then x∗ (t) is on the boundary of K(t. x0 ) at x∗ (tf ). and v ∈ Ω such that (p∗ (t∗ )) B(t∗ )u∗ (t∗ ) < (p∗ (t∗ )) B(t∗ )v . then there exists t∗ ∈ [t0 . (7. This normal is obtained by transporting backwards in time.7) we see that u∗ (·) cannot be optimal. t0 . tf ].86 CHAPTER 7. Deﬁne u(·) ∈ U by ˜ u(t) = ˜ Then (7.10) . x.5) can be rewritten as H(t. and since t∗ is a point of continuity of B(·) and u∗ (·). Corollary 1: For t0 ≤ t1 ≤ t2 ≤ tf .e. the outward normal p∗ (tf ) at time tf . t ∈ [t0 .8) v |t − t∗ | < δ. We shall show that if u∗ (·) is optimal then (7. B(t)˜(t)dt > u tf ∗ t0 (p (t)) B(t)u∗ (t)dt . if c = p∗ (tf ) is the outward normal to a hyperplane supporting K(tf . Remark 2: If we deﬁne the Hamiltonian function H by H(t. p∗ (t)) = M (t. (p∗ (t2 ))x∗ (t2 ) ≥ (p∗ (t2 )) x for all x ∈ K(t2 . p) = sup{H(t. tf ] u∗ (t) otherwise ..8) implies that tf ∗ t0 (p (t)) (7. To prove the necessity let D be the ﬁnite set of points where the function B(·) or u∗ (·) is discontinuous. (7. Taking t1 = t0 in (7. x0 ) at x∗ (t). Indeed if this is not the case. t1 .9). i. p) = p (A(t)x + B(t)u) . and we deﬁne M by M (t. x. tf ∗ t0 (p (τ )) B(τ )u∗ (τ )dτ ≥ tf ∗ t0 (p (τ )) B(τ )u(τ )dτ. p∗ (t)) . u.5) is satisﬁed for t ∈ D.5) is immediate. x.9) is the following. giving a contradiction. (7. ♦ But then from (7.2. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL Now by properties of the adjoint equation we know that p∗ (t)) = (p∗ (tf )) Φ(tf . The situation is illustrated in Figure 7.

β]. p∗ (t)) is a Lipschitz function of t. t0 . Exercise 4: Suppose that Ω is bounded and closed. x∗ (t). Show that u∗ (·) satisﬁes the maximum principle (7. Show that there is an optimal control u∗ (·) and t0 ≤ t1 ≤ t2 ≤ .11) In (7. and then translate these conditions in terms of the control. (7. show that m(t) is constant. so that B(t) is an n × 1 matrix. then f0x (x∗ (tf )(x∗ (tf ) − x) ≤ for all x ∈ K(tf . .11) G0 and Gf are ﬁxed matrices of dimensions 0 xn and f × n respectively. For convenience let T 0 = {z ∈ Rn |G0 z = b0 } . + γn exp(δn (t)) for some γi .) 7. (Hint: ﬁrst show that (p∗ (t)) B = γ1 exp(δ1 t) + .2).. Let f0 : Rn → R be a differentiable function and suppose that the objective function in (7. : [ . T f = {z ∈ Rn |Gf z = bf } . . t0 . The notion of the previous section is retained. δi in R. { ] → ⊗ and u(·)piecewise continuous. Suppose u∗ (·) is an optimal control. Deﬁnition: Let p ∈ Rn . ˙ initial condition: G0 x(t0 ) = b0 .e.) The next two exercises show how we can obtain important qualitative properties of an optimal control. Also show that this condition is sufﬁcient for optimality if f0 is concave. Let z ∗ ∈ T 0 .2) is f0 (x(tf )) instead of c x(tf ). t0 ≤ t ≤ tf . Maximize c x(tf ) subject to dynamics: x(t) = A(t)x(t) + B(t)u(t). Show that there exists an optimal control u∗ (·) such that u∗ (t) belongs to the boundary of Ω for all t.1. (Hint: Show that (dm/dt) ≡ 0. while b0 ∈ R 0 .7. i.10) where p∗ (·) is the solution of the adjoint equation (7.1 to show that if u∗ (·) is optimal. bf ∈ R f are ﬁxed vectors. That is.2. Suppose that A(t) ≡ A and B(t) ≡ B are constant matrices and A has n real eigenvalues. we ﬁrst characterize optimality in terms of the state at the ﬁnal time. We say that p is orthogonal to T 0 at z ∗ and we write p ⊥ T 0 (z ∗ ) if . (ii) If A(t). x0 ).) Exercise 6: Assume that K(tf . We will analyze the problem in the same way as before. 0 ≤ i ≤ n. . B(t) are constant. Exercise 5: Suppose Ω = [α.2 More General Boundary Conditions We consider the following generalization of (7. ﬁnal condition: Gf x(tf ) = bf . MORE GENERAL BOUNDARY CONDITIONS 87 Exercise 3: (i) Show that m(t) = M (t. x0 ) is convex (see remark in Exercise 1 above). (Hint: Use Lemma 1 of 5. . control constraint: u(·) ∈ U . ti+1 ). ≤ tn ≤ tf such that u∗ (t) ≡ α or β on [ti .3) with the ﬁnal condition p∗ (tf ) = f0 (x∗ (tf )) .

x0 ) p∗ (tf ) = c x∗ (tf ) CHAPTER 7. t0 . t0 .9) for t1 = t0 . CONTINUOUS-TIME LINEAR OPTIMAL CONTROL Figure 7. t0 .88 Rn Rn p∗ (t2 ) p∗ (t 1) Rn Rn x0 0) = x∗ (t1 ) K(t1 . t1 t2 tf t t0 .2: Illustration of (7. x0 ) x∗ (t = K(t0 . x0 ) x∗ (t2 ) K(t2 . x0 ) K(tf . t0 .

z. S 1 is convex since T f is convex. But then by the separation theorem for convex sets (see 5. Similarly if z ∗ ∈ T f . Since Ω is convex by hypothesis it follows by Exercise 1 of Section 1 that S 2 is convex. Exercise 1: X(tf ) = {Φ(tf . MORE GENERAL BOUNDARY CONDITIONS p (z − z ∗ ) = 0 for all z ∈ T 0 . ˆ [Φ(tf .18) . Proof: Clearly u∗ (·) is optimal iff c x∗ (tf ) ≥ c x for all x ∈ X(tf ) ∩ T f . r > c x∗ (tf ). x)|r = c x .13) are satisﬁed.19) can hold only if p0 ≥ 0. p ⊥ T f (z ∗ ) if p (z − z ∗ ) = 0 for all z ∈ T f . t0 )z + w|z ∈ T 0 . 89 Lemma 1: Let x∗ (t0 ) ∈ T 0 and u∗ (·) ∈ U . such that p0 r 1 + p x1 ≥ p0 r 2 + p x2 for all (r i . x ∈ X(tf )} .5) there exists p0 ∈ R. x)|r > c x∗ (tf ).12) (7. Secondly. u∗ ).15).14) (ii) Conversely if there exist p0 > 0. p ˆ p ˆ p ⊥ T f (x∗ (tf )) . not ˆ ˆ both zero. such that (ˆ0 c + p) x∗ (tf ) ≥ (ˆ0 c + p) x for all x ∈ X(tf ) . w ∈ K(tf . ˆ ˆ ˆ ˆ which is the same as (7. there exist p0 ∈ R. p ∈ Rn . Let x∗ (t) = φ(t.20) (7.12). t0 . (i) Suppose the Ω is convex.19) can hold only if p0 c x∗ (tf ) + p x∗ (tf ) ≥ p0 c x + p x for all x ∈ X(tf ) . and suppose that x∗ (tf ) ∈ T f . (7. ˆ ˆ ˆ ˆ In particular (7. xi ) ∈ S i . p ˆ (7. S 2 by S 1 = {(r. u(·) ∈ U}. x∗ (t0 ). then u∗ (·) is ˆ ˆ optimal and (7.7. 2. ˆ ˆ ˆ ˆ (7.13) (7. p0 ≥ 0 and p ∈ Rn . If u∗ (·) is optimal.16) (7. x ∈ T f } .14) is also satisﬁed. t0 )] (ˆ0 c + p) ⊥ T 0 (x∗ (t0 )) .18) we get (7. and p such that (7. Deﬁnition: Let X(tf ) = {Φ(tf .18) implies that p0 r + p x∗ (tf ) ≥ p0 c x + p x for all x ∈ X(tf ). (i) Suppose that u∗ (·) is optimal. In R1+m deﬁne sets S 1 .17) First of all S 1 ∩ S 2 = φ because otherwise there exists x ∈ X(tf ) ∩ T f such that c x > c x∗ (tf ) contradicting optimality of u∗ (·) by (7. not both ˆ ˆ ˆ zero.2. u)|z ∈ T 0 .19) Letting r → ∞ we conclude that (7. i = 1. Also from (7. S 2 = {(r. On the other hand letting r → ˆ c x∗ (tf ) we see that (7. t0 . 0)}.15) (7. t0 .12) and (7.

12). Then in part (i) we must have p0 > 0.13).12).15) u∗ (·) is optimal.12) we get 0 ≥ (ˆ0 c + p) Φ(tf . (ii) Now suppose that p0 > 0 and p are such that (7. we note that part (i) is not too useful if p0 = 0. In higher dimensions the reasons may be more ˆ f is “tangent” to X(t ) we may be forced to set p = 0 (see complicated. ˆ ˆ˜ so that from (7. we illustrate a 2-dimensional situation where T 0 = {x0 }. (7. (7. {Φ(tf . ˆ ˆ ˜ Then from (7. 0) is convex even if Ω is not (see Neustadt [1963]). ˆ We now translate the conditions obtained in Lemma 1 in terms of the control u∗ . ˆ ˆ ˆ ˆ which can hold only if p1 c x∗ (tf ) + p x ≥ p0 c x∗ (tf ) + p x∗ (tf ) for all x ∈ T f .12) always implies (7. because by the deﬁnition of X(tf ) and Exercise 1.90 CHAPTER 7. Remark 3: In (i) the convexity of Ω is only used to guarantee that K(tf . In particular. t0 .21) But {x − x∗ (tf )|x ∈ T f } = {z|Gf z = 0} is a subspace of Rn . 0) is convex.12).21) can hold only if p (x − x∗ (tf )) = 0 for all x ∈ T f . Exercise 2: Suppose there exists z in the interior of X(tf ) such that z ∈ T f .13).14) holds. ˆ ˆ ˆ p p (7. t0 . ˆ and (7. in part (ii) of the Lemma we may assume p0 = 1. Intuitively p0 = 0 means that it is so difﬁcult to satisfy ˆ the initial and ﬁnal boundary conditions in (7. But in Figure ˆ 7. ˆ Remark 2: it would be natural to conjecture that in part (i) p0 may be chosen > 0. but basically if T ˆ0 f Exercise 2 below). Finally.14) hold for any vector c whatsoever. u∗ ) ∈ T f is optimal for any c. x ∈ T f .14). It follows that the control u∗ (·) ∈ U for which x∗ (tf ) = φ(tf . ˆ ˆ ˜ but then by (7. Clearly then for some c (in particular for the c in Figure 7.11) that optimization becomes a secondary matter. t0 )(z − x∗ (t0 )) + x∗ (tf )} ∈ X(tf ) for all z ∈ T 0 . t0 . . since then (7.13). t0 )(z − x∗ (t0 )) for all z ∈ T 0 . and (7.13) are satisﬁed. p = (ˆ/ˆ0 ) will also satisfy (7.13) we conclude that p x∗ (tf ) = p x . Let x ∈ X(tf ) ∩ T f .3 below.3) we are forced to set p0 = 0. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL p0 r + p x ≥ p0 c x∗ (tf ) + p x∗ (tf ) for all r > c x∗ (tf ).12) we get p0 c x∗ (tf ) ≥ p0 c x . ˆ ˆ ˆ ˆ or p (x − x∗ (tf )) ≥ 0 for all x ∈ T f ˆ (7. x0 . p ˆ which can hold only if (7. so that (7. T f is the vertical line. But it is known that K(tf . and T f ∩ X(tf ) consists of just one vector. ˆ which is the same as (7. ♦ Remark 1: If it is possible to choose p0 > 0 then p0 = 1. Finally (7.14). so that from (7.

u. p) = p (A(t)x + B(t)u). (7. p)|v ∈ Ω}. (7. (i) Suppose that Ω is convex. and (7. x.22) with the ﬁnal condition p∗ (tf ) = p∗ c + p = p0 c + p .11).13). (7.13) and (7.22). x∗ (t).7.14) are satisﬁed. (ii) Suppose p∗ > 0 and (7.13) are respectively equivalent to (7. (7. [Here H(t.26) (i) Suppose u∗ (·) is optimal and Ω is convex. such that (7. ˆ ˆ ˆ 0 Then (7. MORE GENERAL BOUNDARY CONDITIONS 91 Theorem 1: Let x∗ (t0 ) ∈ T 0 and u∗ (·) ∈ U .24). Let x∗ (t) = φ(t. then (7. v. then there exist a number p∗ ≥ 0.26): (p∗ (tf )) x∗ (tf ) ≥ (p∗ (tf )) x for all x ∈ K(tf .12). x.] Proof: A repetition of a part of the argument in the proof of Theorem 1 of Section 1 show that if p∗ satisﬁes (7. (7.22). Then by Lemma 1 there exist p ≥ 0. If u∗ (·) is optimal for (7.12). t0 .24). x∗ (t). tf ] except possibly for a ﬁnite set.14) and (7. M (t. u∗ (t). (7.23) and (7.25). p∗ (t)) . not ˆ ˆ both zero. x∗ (t0 ). Next if x ∈ X(tf ) we have 0 (ˆ0 c + p) x = (p∗ (tf )) x p ˆ = (p∗ (tf )) (Φ(tf . not both identically zero. 0 and the maximum principle H(t. x. Let p∗ = p0 and let p∗ (·) be the solution ˆ 0 of (7. x∗ (t0 )) ⊂ X(tf ).26) is implied by (7.24) becomes equivalent to (7.24).23) ﬁnal condition: (p∗ (tf ) − p∗ c)⊥T f (x∗ (tf )) .24) (7. t0 . t0 ≤ t ≤ tf ˙ (7.25) holds for all t ∈ [t0 . so that (7. t0 . Let p0 = p∗ and p = ˆ ˆ 0 0 p∗ (tf ) − p∗ c. .25) is equivalent to (7. p ∈ Rn .23). tf ] → Rn .2. (7. and a 0 function p∗ : [t0 .22). p) = sup{H(t. 0 Then u∗ (·) is optimal. t0 )z + w) . u∗ ) and suppose that x∗ (tf ) ∈ T f .22) initial condition: p∗ (t0 )⊥T 0 (x∗ (t0 )) (7. (ii) Conversely suppose there exist p∗ > 0 and p∗ (·) satisfying (7.23). and (7. x∗ (t0 )) . p∗ (t)) = M (t. whereas since K(tf .26) are satisﬁed. (7. satisfying adjoint equation: p∗ (t) = −A (t)p∗ (t) .

t0 . x (t ) = X(t ) ∗ K(tf . x0 ) = X(tf ) f f Tf Figure 7.3: Situation where p0 = 0 ˆ x0 = T 0 c t .92 CHAPTER 7. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL Tf .

Thus (ˆ0 c + p) x∗ (tf ) ≥ (ˆ0 c + p) x for all x ∈ X(tf ) . Show that Theorem 1 also holds for this case where. ♦ Exercise 3: Suppose that the control constraint set is Ω(t) which varies continuously with t. t0 )x∗ (t0 )) .23) the ﬁrst term on the right vanishes. v. x. t0 .25). t0 )x∗ (t0 )) ∈ K(tf . it follows from (7. and we require that u(t) ∈ Ω(t) for all t.26) that the second term is bounded by (p∗ (tf )) x∗ (tf ).2. Exercise 4: How would you use Exercise 3 to solve Example 3 of Chapter 1? . 93 But by (7. p ˆ p ˆ and so u∗ (·) is optimal by Lemma 1. x. t0 )x∗ (t0 )) = (p∗ (t0 )) (z − x∗ (t0 )) +(p∗ (tf )) (w + Φ(tf . p) =sup{H(t. in (7.7. Hence (ˆ0 c + p) x = (p∗ (f )) Φ(tf . x∗ (t0 )). 0). MORE GENERAL BOUNDARY CONDITIONS for some z ∈ T 0 and some w ∈ K(tf . and since (w+φ(tf . t0 )(z − x∗ (t0 )) p ˆ +(p∗ (tf )) (w + φ(tf . p)|v ∈ Ω(t)}. M (t. t0 .

94 CHAPTER 7. CONTINUOUS-TIME LINEAR OPTIMAL CONTROL .

1 Preliminary results based on differential equation theory. (t).1) where x(t) ∈ is the state and u(t) ∈ is the control.Chapter 8 SEQUENTIAL DECISION PROBLEMS: CONTINUOUS-TIME OPTIMAL CONTROL OF NONLINEAR SYSTEMS We now present a sweeping generalization of the problem studied in the last chapter. et al.2 is presented in Section 1. Finally. ˙ Rn Rp u∗ (·) (8. Section 3 deals with the minimum-time problem and Section 4 considers the important special case of linear systems with quadratic cost. in Section 5 we discuss the so-called singular case and also analyze Example 4 of Chapter 1. An alternative form of the objective function is discussed in Section 2.1). u(t)) . which is a direct generalization of Theorem 1 of 7.1 Main Results 8. We can then estimate the difference between x(·) and x∗ (·) by the solution to a linear differential equation as shown in Lemma 1 below. tf ] × Rn × Rp → Rn satisﬁes the following conditions: 95 .. Unfortunately we are forced to omit the proofs of the results since they require a level of mathematical sophistication beyond the scope of these Notes. Unfortunately when f is nonlinear such a characterization is not available. t0 ≤ t ≤ tf . Suppose is an optimal control ∗ (·) is the corresponding trajectory. This comparison was possible because we had an explicitly characterization of x(·) in terms of u(·).1. (For complete proofs see (Lee and Markus [1967] or Pontryagin. it is possible to convey the main ideas of the proofs at an intuitive level and we shall do so. [1962]. Instead we shall settle for a comparison between the trajectory x∗ (·) and trajectories x(·) obtained by perturbing the control u∗ (·) and the initial condition x∗ (t0 ). In the case of linear systems we obtained the necessary and x conditions for optimality by comparing x∗ (·) with trajectories x(·) corresponding to other admissible controls u(·). But ﬁrst we need to impose some regularity conditions on the differential equation (8. We assume throughout that the function f : [t0 . However.) The principal result. x. We are interested in the optimal control of a system whose dynamics are governed by the nonlinear differential equation x(t) = f (t. 8.

tf ]. z. x(t). ˙ satisfying the initial condition x(t1 ) = z . The following result is proved in every standard treatise on differential equations. (t). tf ] → Ω. z. . . . u ∈ Rp with |u| ≤ α . i = 1. . 1. . the n × n matrix-valued function Φ deﬁned by Φ(t2 . t1 . Moreover. for every t1 ∈ [t0 . Now let Ω ⊂ Rp be a ﬁxed set and let U be set of all piecewise continuous functions u(·) : [t0 . t1 ≤ t ≤ tf . u1 . u(·)) is the solution of the linear homogeneous differential equation ∂Φ ∂t (t. z. i D. t1 .96 CHAPTER 8. of the differential equation x(t) = f (t. tm . tm < tf . t1 . tf ]. the function φ(t2 . . t1 ≤ t ≤ tf . m. . 3. t1 . . Theorem 1: For every z ∈ Rn . the functions f. fx . u(t))]Φ(t. x. t0 < t1 < t2 < . . . there exists a unique solution x(t) = φ(t. z. ∂x and the initial condition Φ(t1 . tf ]×Rn × Rp . . m (recall that D is the set of ≥ 0. t1 . there exist ﬁnite number β and γ such that |f (t. ·) : Rn xRp → Rn is continuously differentiable in the remaining variables (x. and every piecewise continuous function u(·) : [t0 . . . x ∈ Rn . . u(t)) . . u)| ≤ β + γ|x| for all t ∈ [t0 . . . 0 Deﬁnition: π = (t1 . i = 1. for every ﬁnite α. tf ] → Rp . ·. u(·)) . Let x∗ ∈ Rn be a ﬁxed initial condition. . i = 1. Furthermore. u(·)) : Rn → Rn is differentiable. . CONINUOUS-TIME OPTIMAL CONTROL 1. tf ]. u(·)) = ∂φ ∂z (t2 . (·)) = [ ∂f (t. and 4. except for a ﬁnite subset D ⊂ [t0 . . u). 2. for each ﬁxed t ∈ [t0 . z. . . tf ] and ﬁxed u(·). for ﬁxed t1 ≤ t2 in [t0 . tf ]. x. t1 . fu are continuous on [t0 . . ui ∈ Ω. t1 . and ti ∈ D ∗ discontinuity points of f ). 2. m. u(·)). u. . . Let u∗ (·) ∈ U be ﬁxed and let D∗ be the set of discontinuity points of u∗ (·). and 3. ·. . f (t. m is a nonnegative integer. . um ) is said to be a perturbation data for u∗ (·) if 1. z. t1 ≤ t ≤ tf . u(·)) = In . m.

.ε) (·) ∈ U corresponding to π is deﬁned by u(π. x∗ (t1 ).8. t ∈ [tm . t1 ]. t ∈ [ti . t1 ) 1 . tf ] . u∗ (t1 ))] i . t0 . let Q(t) = {h(π. ti ] . . x∗ ). tj )[f (tj . 0)} . t0 )ξ = Φ(t. u∗ (tj ))] Φ(t. tf ] let K(t. t0 .0) (t)|πis a perturbation data for u∗ (·). x∗ (t1 ). uj ) − f (tj . u(π. t ∈ [t0 . .ε) (t)| = 0 for t ∈ [t0 . The proof of the lemma is a straightforward exercise in estimating differences of solutions to differential equations. t0 )ξ + j=1 j (See Figure 8. Lemma 1: lim |xε (t) − x∗ (t) − εh(π. x∗ (t1 ). starting at time t0 in state z. and using controls u(·) ∈ U . m u∗ (t) otherwise . x∗ (tj ). Deﬁnition: Any vector ξ ∈ Rn is said to be a perturbation for x∗ .0) (·)is the linearized perturbation corresponding to(π. uj ) − f (tj . t0 . where h(π. x(ξ. and it is omitted (see for example (Lee and Markus [1967])). x∗ (tj ). Deﬁnition: For each t ∈ [t0 . . The following lemma gives an estimate of x∗ (t) − xε (t). u∗ (·)). . Then for 0 ≤ ε ≤ ε(π). u1 ) − f (t1 .ξ) ) belongs to the set K(t. tf ] for all i. i = 1. t2 ) j = Φ(t. 0 More precisely we have the following result which we leave as an exercise.ε) . x(ξ. 0 ε→0 and ε→0 lim 1 ε (x(ξ. t ∈ [t1 .ε) − x∗ ) = ξ . ti ] [tj −ε j . ξ). Remark: By Lemma 1 (x∗ (t)+εh(π. u(·))|u(·) ∈ U} be the set of states reachable at time t. u∗ tj ))] .1. t0 .ε) deﬁned for 0 ε > 0 is said to be a perturbed initial condition if lim x(ξ. = Φ(t.ε) (·)). t0 . ti+1 ) . and h(π. z. z) = {φ(t.ε) (t) = ui for all t ∈ [ti − ε i . MAIN RESULTS 97 Let ε(π) > 0 be such that for 0 ≤ ε ≤ ε(π) we have [ti − ε i . Let Φ(t2 .ε) (t) = Φ(t.1.the perturbed control u(π. Deﬁnition: For z ∈ Rn . t ∈ [t0 . t0 )ξ + Φ(t. t1 ) = 0 Φ(t2 . t1 )[f (t1 .ε) (·) is given by ε→0 h(π. and a function x(ξ. x∗ . 0 Now let x∗ (t) = φ(t.) We call h(π. t0 )ξ + j=1 m Φ(t. tj ] = φ for i = j.ε) ) up to an error of order o(ε). x∗ (tj ). t1 . tf ]. In particular for ξ = 0. t0 . ti ] ⊂ [t0 .ξ) (·) the linearized (trajectory) perturbation corresponding to (π. tm )[f (tj . x∗ (tj ). the set x∗ (t) + Q(t) can serve as an approximation to the set K(t. and [ti −ε i . u∗ (·)) and let xε (t) = φ(t.ε) = x∗ .

t0 ≤ t ≤ tf . ˙ initial condition: x(t0 ) = x∗ . CONINUOUS-TIME OPTIMAL CONTROL u1 u(πε) (·) u∗ (·) ε | 1 u2 t1 | | u3 ε 2 ε | | 3 t0 x t2 t3 tf x∗ (·) | | xε (·) x( ξ.98 u CHAPTER 8. u∗ (·)). Theorem 2: Consider the optimal control problem (8. Exercise 1: (Recall the deﬁnition of the tangent cone in 5. 0 We can now prove a generalization of Theorem 1 of 7. i.. u∗ (t))] p∗ (t).1. Let p∗ (t). t0 ≤ t ≤ tf . be the 0 corresponding trajectory. ˙ ∂x (8. (8. t0 ≤ t ≤ tf . t0 .e. u(t)) . be the solution of (8.4) and (8. t0 ≤ t ≤ tf . x∗ . ε) εhπξ t1 | | | | t2 t3 tf Figure 8.) Show that Q(t) ⊂ C(K(t.2) (8.5): adjoint equation: p∗ (t) = −[ ∂f (t. tf ] → Ω and u(·) piecewise continuous . Let u∗ (·) ∈ U be an optimal control and let x∗ (t) = φ(t. x∗ (t)) .1. u : [t0 . t0 .3): Maximize ψ(x(tf )) subject to dynamics: x(t) = f (t.4) . x∗ ). x(t). control constraint: u(·) ∈ U .1: Illustration for Lemma 1.3) where ψ : Rn → R is differentiable and f satisﬁes the conditions listed earlier. x∗ (t).1. 0 ﬁnal condition: x(tf ) ∈ Rn .

Then for all ε > 0 sufﬁciently small.1. x.5) (8. x∗ (t0 )) up to an error of order o(ε). t0 . x.2 More general boundary conditions. 0 and so by Lemma 1 of 5.1. t0 . In Theorem 2 the initial condition is ﬁxed and the ﬁnal condition is free. x∗ (t). v. t∗ ). u. MAIN RESULTS ﬁnal condition: p∗ (tf ) = Then u∗ (·) satisﬁes the maximum principle H(t. Deﬁnition: Let C(t) denote the closure of Q(t). p) = sup{H(t. Exercise 3: Show that (i) C(t) is a convex cone. 1. x(t∗ ). p) = p f (t. x∗ (tf )) .7) Now suppose that (8. (8.9) Now from (8. Proof: Since u∗ (·) is optimal we must have ψ(x∗ (tf )) ≥ ψ(z) for all z ∈ K(tf . t∗ )h(π.2) ψx (x∗ (tf ))h ≤ 0 for all h ∈ Q(tf ) . The problem involving more general boundary conditions is much more complicated and requires more reﬁned analysis. Lemma 1 needs to be extended to Lemma 2 below. x. t1 )Q(t1 ) ⊂ Q(t2 ) . Then there exists v ∈ Ω such that (8. Also h(π.6) does not hold from some t∗ ∈ D ∗ ∪ D.1. Φ(t2 . t0 .9) is equivalent to p∗ (tf ) h(π.4) we can see that p∗ (t∗ ) = p∗ (tf ) Φ(tf .0) (t∗ ) so that (8. Remark: From Lemma 1 we know that if h ∈ C(t) then (x∗ (t) + εh) belongs to K(t. then λh ∈ Q(t). . if h ∈ Q(t) and λ ≥ 0. u.7).0) (tf ) = Φ(tf .8) 8. p)|v ∈ Ω}]. Lemma 2: Let h belong to the interior of the cone C(t). p∗ (t∗ ) [f (t∗ . [Here H(t. then (8. x∗ ) . But ﬁrst we need some simple properties of the sets Q(t) which we leave as exercises. i. Φ(t2 . Lemma 2.e. asserts further that if h is in the interior of C(t) then in fact (x∗ (t) + εh) ∈ K(t.8) is equivalent to p∗ (t∗ ) h(π. 0 and in particular from (8. x(t∗ ). If we consider the perturbation data π = (t∗ . x∗ (t0 )) for ε > 0 sufﬁciently small. tf ] except possibly for a ﬁnite set. v). Instead we offer a plausibility argument. p∗ (t)) = M (t..0) (t∗ ) > 0 . M (t. ♦ (8. u∗ (t∗ ))] > 0 . (ii) for t0 ≤ t1 ≤ t2 ≤ tf .6) for all t ∈ [t0 . 99 (8. The proof of the lemma depends upon some deep topological results and is omitted. x∗ ). (ii) for t0 ≤ t1 ≤ t2 ≤ tf . x. x∗ (t). t1 )C(t1 ) ⊂ C(t2 ) . u∗ (t).0) (tf ) > 0 which contradicts (8. v) − f (t∗ . Exercise 2: Show that (i) Q(t) is a cone. t0 . Speciﬁcally. ). p∗ (t)) ψ(x∗ (tf )) .1 ψ(x∗ (tf ))h ≤ 0 for all h ∈ C(K(tf .8. below.

100

CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
(x∗ (t) + εh) ∈ K(t, t0 , x∗ ) . 0 (8.10)

Plausibility argument. (8.10) is equivalent to εh ∈ K(t, t0 , x∗ (t0 )) − {x∗ (t)} , where we have moved the origin to x∗ (t). The situation is depicted in Figure 8.2. ˆ K(ε) ˆ C(ε) o(ε) K(t1 , t0 , x∗ ) − {x∗ (t)} εh 0

(8.11)

h

δε

C(t)

Figure 8.2: Illustration for Lemma 2. ˆ Let C(ε) be the cross-section of C(t) by a plane orthogonal to h and passing through εh. Let ˆ K(ε) be the cross-section of K(t, t0 , x∗ ) − {x∗ (t0 )} by the same plane. We note the following: 0 ˆ ˆ (i) by Lemma 1 the distance between C(ε) and K(ε) is of the order o(ε); ˆ (ii) since h is in the interior of C(t), the minimum distance between εh and C(ε) is δε where δ > 0 is independent of ε. ˆ Hence for ε > 0 sufﬁciently small εh must be trapped inside the set K(ε). (This would constitute a proof except that for the argument to work we need to show that there ˆ are no “holes” in K(ε) through which εh can “escape.” The complications in a rigorous proof arise precisely from this drawback in our plausibility argument.) ♦ ∗ ) in a neighborhood of x∗ (t) when we Lemmas 1 and 2 give us a characterization of K(t, t0 , x0 perturb the control u∗ (·) leaving the initial condition ﬁxed. Lemma 3 extends Lemma 2 to the case when we also allow the initial condition to vary over a ﬁxed surface in a neighborhood of x∗ . 0 0 Let g0 : Rn → R 0 be a differentiable function such that the 0 × n matrix gx (x) has rank 0 n 0 0 0 ∗ 0 0 for all x. Let b ∈ R be ﬁxed and let T = {x|g (x) − b }. Suppose that x0 ∈ T and let 0 (x∗ ) = {ξ|g 0 (x∗ )ξ = 0}. Thus, T 0 (x∗ ) + {x∗ } is the plane through x∗ tangent to the surface T x 0 0 0 0 0 T 0 . The proof of Lemma 3 is similar to that of Lemma 2 and is omitted also. Lemma 3: Let h belong to the interior of the cone {C(t)+Φ(t, t0 )T 0 (x∗ )}. For ε ≥ 0 let h(ε) ∈ Rn 0 1 be such that lim h(ε) = 0, and lim ( )h(ε) = h. Then for ε > 0 sufﬁciently small there exists ε→0 ε x0 (ε) ∈ T 0 such that (x∗ (t) + h(ε)) ∈ K(t, t0 , x0 (ε)) .

8.1. MAIN RESULTS

101

We can now prove the main result of this chapter. We keep all the notation introduced above. f Further, let gf : Rn → R f be a differentiable function such that gx (x) has rank f for all x. f ∈ Rn be ﬁxed and let T f = {x|g f (x) − bf }. Finally, if x∗ (t ) ∈ T f let T f (x∗ (t )) = Let b f f f {ξ|gx (x∗ (tf ))ξ = 0}. Theorem 3: Consider the optimal control problem (8.12): Maximize ψ(x(tf )) subject to dynamics: x(t) = f (t, x(t), u(t)) , t0 ≤ t ≤ tf , ˙ initial conditions: g0 (x(t0 )) = b0 , ﬁnal conditions: gf (x(tf )) = bf , control constraint: u(·) ∈ U , i.e., u : [t0 , tf ] → Ω and u(·) piecewise continuous .

(8.12)

Let u∗ (·) ∈ U , let x∗ ∈ T 0 and let x∗ (t) = φ(t, t0 , x∗ , u∗ (·)) be the corresponding trajectory. 0 0 Suppose that x∗ (tf ) ∈ T f , and suppose that (u∗ (·), x∗ ) is optimal. Then there exist a number 0 p∗ ≥ 0, and a function p∗ : [t0 , tf ] → Rn , not both identically zero, satisfying 0 adjoint equation: p∗ (t) = −[ ∂f (t, x∗ (t), u∗ (t))] p∗ (t), t0 ≤ t ≤ tf , ˙ ∂x initial condition: p∗ (t0 )⊥T 0 (x∗ ) , 0 ﬁnal condition: (p∗ (tf ) − p∗ ψ(x∗ (tf )))⊥T f (x∗ (tf )) . 0 Furthermore, the maximum principle H(t, x∗ (t), u∗ (t), p∗ (t)) = M (t, x∗ (t), p∗ (t)) (8.16) (8.13) (8.14) (8.15)

holds for all t ∈ [t0 , tf ] except possibly for a ﬁnite set. [Here H(t, x, p, u) = p f (t, x, u, ), M (t, x, p) = sup{H(t, x, v, p)|v ∈ Ω}]. Proof: We break the proof up into a series of steps. Step 1. By repeating the argument presented in the proof of Theorem 2 we can see that (8.15) is equivalent to p∗ (tf ) h ≤ 0 for all h ∈ C(tf ) . (8.17)

Step 2. Deﬁne two convex sets S1 , S2 in R1+m as follows: S1 = {(y, h)|y > 0, h ∈ T f (x∗ (tf ))}, S2 = {(y, h)|y = ψx (x∗ (tf ))h, h ∈ {C(tf ) + Φ(tf , t0 )T 0 (x∗ )}} . 0 We claim that the optimality of (u∗ (·), x∗ ) implies that S1 ∩ Relative Interior (S2 ) = φ. Suppose 0 this is not the case. Then there exists h ∈ T f (x∗ (tf )) such that ψx (x∗ (tf ))h > 0 , (8.18)

102

CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
h ∈ Interior{C(tf ) + Φ(tf , t0 )T 0 (x∗ )} . 0 (8.19)

f f Now by assumption gx (x∗ (tf ) has maximum rank. Since gx (x∗ (tf ))h = 0 it follows that the Implicit Function Theorem that for ε > 0 sufﬁciently small there exists h(ε) ∈ Rn such that

gf (x∗ (tf ) + h(ε)) = bf ,

(8.20)

and, moreover, h(ε) → 0, (1/ε)h(ε) → h as ε → 0. From (8.18) and Lemma 3 it follows that for ε > 0 sufﬁciently small there exists x0 (ε) ∈ T 0 and uε (·) ∈ U such that x∗ (tf ) + h(ε) = φ(tf , t0 , x0 (ε), uε (·)) . Hence we can conclude from (8.20) that the pair (x0 (ε), uε (·)) satisﬁes the initial and ﬁnal conditions, and the corresponding value of the objective function is ψ(x∗ (tf ) + h(ε)) = ψ(x∗ (tf )) + ψx (x∗ (tf ))h(ε) + o(|h(ε)|) , and since h(ε) = εh + o(ε) we get ψ(x∗ (tf ) + h(ε)) = ψ(x∗ (tf )) + ε)ψx (x∗ (tf ))h + o(ε) ; but then from (8.18) ψ(x∗ (tf ) + h(ε)) > ψ(x∗ (tf )) for ε > 0 sufﬁciently small, thereby contradicting the optimality of (u∗ (·), x∗ ). 0 Step 3. By the separation theorem for convex sets there exist p0 ∈ R, p1 ∈ Rn , not both zero, such ˆ ˆ that p0 y 1 + p1 h1 ≥ p0 y 2 + p1 h2 for all (y i , hi ) ∈ S1 , i = 1, 2 . ˆ ˆ ˆ ˆ (8.21)

Arguing in exactly the same fashion as in the proof of Lemma 1 of 7.2 we can conclude that (8.21) is equivalent to the following conditions: p0 ≥ 0 , ˆ p1 ⊥T f (x∗ (tf )) , ˆ Φ(tf , t0 ) (ˆ0 ψ(x∗ (tf )) + p1 )⊥T 0 (x∗ ) , p ˆ 0 and (ˆ0 ψx (x∗ (tf )) + p1 )h ≤ 0 for all h ∈ C(tf ) . p ˆ (8.24)

(8.22)

(8.23)

If we let p∗ = p0 and p∗ (tf ) = p0 ψ(x∗ (tf )) + p1 then (8.22), (8.23), and (8.24) translate respecˆ0 ˆ ˆ ˆ tively into (8.15), (8.14), and (8.17). ♦

and M (t. x. p∗ (t)) ˜ ˜ ˜ holds for all t ∈ [t0 . tf ] → R1+m . and control constraints are the same as before. let x∗ ∈ T o and let x∗ (t) = φ(t. p∗ ) : [t0 . Theorem 1: Consider the optimal control problem (8. We proceed to show how such objective functions can be treated as a special case of the problems of the last section. t0 . (Hint: For the ﬁnal part show that (d/dt) M (t. Finally. then there exists a function p∗ = (p∗ . x) ∈ ˜ R1+m as follows: · f0 (t. p∗ (t). if f0 and f do not explicitly depend on t. u(t)) x0 (t) ˙ ˜ ˜ .25) The dynamics of the state. u(t))dt .2. [Here H(t. the maximum principle ˜ ˜ H(t. 0 ﬁnal condition: p∗ (tf )⊥T f (x∗ (tf )) . but rather as an integral of the form tf t0 f0 (t. x(t). u∗ (t))] p∗ (t) .26): tf Maximize t0 f0 (t. control constraint: u(·) ∈ U . x∗ (t). gf (x) = bf are augmented g0 (˜) = ˜ x x0 g0 (x) = ˜0 = b 0 b0 and gf (˜) = gf (x) = bf . x(t). x subject to the augmented dynamics and constraints which is of the form treated in Theorem 3 of Section 1. ﬁnal conditions: gf (x(tf )) = bf .8. Evidently then the problem of maximizing (8. x∗ (t). x. INTEGRAL OBJECTIVE FUNCTION 103 8. x(t). u(t)) = x(t) ˙ f (t. and with p∗ (t) ≡ constant and p∗ (t) ≥ 0. u∗ (·)). u(t)) The initial and ﬁnal conditions which are of the form g0 (x) = b0 . x∗ (t). p) = sup{H(t. and suppose that x∗ (tf ) ∈ T f . p∗ (t)) ≡ constant. x∗ . tf ] except possibly for a ﬁnite set. To this end we deﬁned the augmented system with state variable x = (x0 . u) + p f (t. If 0 0 (u∗ (·). Let u∗ (·) ∈ U. u∗ (t)) = M (t.] ˜ ˜ ˜ (t. x= ˜ = f (t. satisfying 0 0 (augmented) adjoint equation: p∗ (t) = −[ ∂ f (t. p. not identically ˜ 0 0 zero.2 Integral Objective Function In many control problems the objective function is not given as a function ψ(x(tf )) of the ﬁnal state. x. the boundary conditions. u(t))dt (8. v)|v ∈ Ω}. x. u). p∗ (t)) ≡ 0. x(t). x∗ (t).25) is equivalent to the ˜ x problem of maximizing ψ(˜(tf )) = x0 (tf ) .26) subject to dynamics: x(t) = f (t. then M ˜ ˜ Exercise 1: Prove Theorem 1. x∗ (t). p. x. x∗ ) is optimal. Futhermore. and we get the following result. ˜ ˜ ∂x ˜ initial condition: p∗ (t0 )⊥T 0 (x∗ ) . (8. ˙ initial conditions: g0 (x(t0 )) = b0 . x(t). u(t)). t0 ≤ t ≤ tf . u) = ˜ ˜˜ ˜ ˜ p0 f0 (t. x(t). x.) ˜ · ˜ . u) = p f (t.

ﬁnal-time constraint: tf ∈ (t0 . The equation for t is dt(s) ds = α(s) . t0 ≤ t ≤ tf . 0 ≤ s ≤ 1 z(0) = x0 .1 Main result. tf ] into a ﬁxed-time interval [0. 1] is the functional inverse of s(t). where s(·) : [t0 . v(s) = u(t(s)) . 0 ≤ s ≤ 1 . consider the optimal control problem (8. Now if x(·) is the solution of x(t) = f (t.29) we can obtain the solution x(·) of (8. tf Maximize t0 f0 (t. Here α(s) is a new control variable constrained by α(s) ∈ (0.27) We analyze (8. ˙ . tf ] → [0. u(t))dt subject to dynamics: x(t) = f (t. ∞) .29) Conversely from the solution z(·) of (8. In the problem considered up to now the ﬁnal time tf is assumed to be ﬁxed. then it is easy to see that z(·) is the solution of dz ds (s) (8. (t). u(t)). t0 ≤ t ≤ tf . in fact. with initial condition t(0) = t0 . v(s)) .28) by x(t) = z(s(t)) . More generally. t0 ≤ t ≤ tf .27). u(t)) .3 Variable Final Time 8. ∞). (8. s(·) is the solution of the differential equation s(t) = 1/α(s(t)). x. s(t0 ) = 0. . x(t). 0 ≤ s ≤ 1 . 1]. control constraint: u(·) ∈ U . z(s). x(t0 ) = x0 ˙ and if we deﬁne z(s) = x(t(s)).3. This change of time-scale is achieved by regarding t as a new state variable and selecting a new time variable s which ranges over [0. One such case is the minimum-time problem where we want to transfer the state of the system from a given initial state to a speciﬁed ﬁnal state in minimum time. ˙ initial condition: g0 (x(t0 )) = b0 .27) by converting the variable time interval [t0 .28) = α(s)f (s. (8.104 CHAPTER 8. In many important cases the ﬁnal time is itself a decision variable. CONINUOUS-TIME OPTIMAL CONTROL 8. 1]. x(t). ﬁnal condition: gf (x(t)f )) = bf .

let x∗ ∈ T 0 . and α∗ (·) by ∗ z0 = x∗ 0 ∗ (s) = u∗ (t + s(t∗ − t )) v . x∗ . z) ∈ R1+m . f ∗ where s∗ (·) is functional inverse of t∗ (·).30). Deﬁne z0 . t0 . 0 u∗ (t) = v ∗ (s∗ (t)) . control constraint: (v(s). t∗ ∈ (t0 . 0≤s≤1. Deﬁne x∗ . 0 ·∗ ˜ (8. x∗ .30) such that the correspond∗ ing trajectory (t∗ (·). x∗ (t). t∗ ] → R1+m . u∗ (·) ∈ U . v(s))α(s)ds (8. z0 ) is optimal for (8. let t∗ ∈ (0.27). t0 ≤ t ≤ t∗ . and suppose that x∗ (t∗ ) ∈ T f . z0 ) is optimal for (8. ∗ (ii) Let z0 ∈ T 0 . α(·) piecewise continuous. u∗ (·)) be the 0 0 f corresponding trajectory.27). Theorem 1: Let u∗ (·) ∈ U. α∗ (·)) be an admissible control for (8. α(s)). ˙ initial constraint: g0 (z(0)) = b0 . where the state vector (t. t∗ ) is optimal 0 0 f f for (8. Suppose that ((v ∗ (·). α∗ (·)). f ∗ Then ((v ∗ (·).27). 0≤s≤1. Suppose that x∗ (t∗ ) ∈ T f . α∗ (·)). and the control (α. f Exercise 1: Prove Lemma 1. u∗ (t))] p∗ (t) .30) is established in the following result. v ∗ (·).30).32) . p∗ ) : [t0 . z0 . t0 ≤ t ≤ tf . t∗ ) is optimal for (8. ∞).30). u∗ (·)). and let 0 f x∗ (t) = φ(t.30) Lemma 1: (i) Let x∗ ∈ T 0 . The relation between problems (8. Then (u∗ (·). and suppose that (u∗ (·). t(s)) = (f (t(s). x∗ . and t∗ by 0 f ∗ x∗ = z0 . ∞) and let x∗ (t) = φ(t. If (u∗ (·). t(1) ∈ R . v(s))α(s). satisfying 0 0 (augmented) adjoint equation: p (t) = −[ ∂ f (t. t(0) = t0 . t∗ ) is optimal for 0 f f ∗ (8. VARIABLE FINAL TIME 105 With these ideas in mind it is natural to consider the ﬁxed-ﬁnal-time optimal control problem (8. and ˜ 0 f with p∗ (t) ≡ constant and p∗ (t) ≥ 0. f t∗ = t∗ (1) . z(s). 0 f0 (t(s). z(s).3.8. α(s)) ∈ Ω × (0. v) ∈ R1+p : 1 Maximize subject to ˙ dynamics: (z(s). t0 . ˜ ˜ ∂x ˜ initial condition: p∗ (t0 )⊥T 0 (x∗ ) . ﬁnal constraint: gf (z(1)) = bf . x∗ .31) (8. and let (v ∗ (·). 0 0 f α∗ (s) = (t∗ − t0 ) .27) and (8. then there exists a function p∗ = (p∗ . not identically zero. z ∗ (·)) satisﬁes the ﬁnal conditions of (8.30). ∞) for 0 ≤ s ≤ 1 and v(·). u∗ (·) ∈ U.

then from (8. ˜ ˜ holds for all t ∈ [t0 . tf ] except possibly for a ﬁnite set. n+1 Furthermore.36). v ∗ (s))] λ∗ (s) n+1 0 +[ ∂f (t∗ (s). x∗ (t). 1] → R1+n+1 . f f ˜ f f ˆ Finally. 0 ≤ s ≤ 1 . and with λ0 0 (8. p∗ (t) = λ∗ (s∗ (t)). ∞)} n+1 (8. satisfying identically zero. λ∗ ) : [0. Because if p∗ ≡ 0.35)   ˙ 0 λ∗ (t) 0  {[ ∂f0 (t∗ (s). so that in particular f f z ∗ (1) = x∗ (t∗ ). v ∗ (s))] λ∗ (s) ˙  λ∗ (t)   0 ∂z    = −  +[ ∂f (t∗ (s).34) (8.39) holds for all s ∈ [0. β ∈ (0. z ∗ (s). λ∗ ≡ constant. u∗ (t)) = M (t. x∗ (t). z ∗ (s). p∗ (t)) . λ∗ (1) = 0 . z ∗ (s). there exists a function λ∗ = (λ∗ . w)β 0 +λ∗ (s) f (t∗ (s). λ∗ . λ∗ ) ≡ 0 and ˜ ˜ 0 ˜ then from (8. z0 = x∗ . v ∗ (s))α∗ (s) + λ∗ (s)α∗ (s) n+1 = sup{[λ∗ (s)f0 (t∗ (s). v ∗ (s))] λ∗ (s)}α∗ (s) ∂t  ∗ initial condition: λ∗ (0)⊥T 0 (z0 )        (8.40) we have (λ∗ . p∗ (t). z ∗ (s). then M (t. but from (8. p∗ ) : [t0 . Let s∗ (t) = (t − t0 )/(t∗ − t0 ). not 0 n+1 ∗ (s) ≡ constant and λ∗ (s) ≥ 0. z ∗ (s). t∗ (s) = t0 + s(t∗ − t0 ). ˜ ∗ Proof: By Lemma 1. t∗ must be such that f ˆ H(t∗ . f f (8. z ∗ (s). w)β + λ∗ (s)β]|w ∈ Ω. x∗ (t). v ∗ (s))] λ∗ (s)}α∗ (s) adjoint equation:    ∂z   ∂f0 ∗  λ∗ (t)  ˙  {[ ∂t (t (s).38) ﬁnal condition: λ∗ (1)⊥T f (z ∗ (1)) . λ∗ (1) = 0 so that we would have λ∗ ≡ 0 n+1 n+1 .37) (8. Furthermore. u∗ (t∗ )) = 0 . v ∗ (s))α∗ (s) 0 +λ∗ (s) f (t∗ (s). t∗ ] → R1+n by ˜ 0 f f f p∗ (t) = λ∗ (s∗ (t)).106 CHAPTER 8.38).33) Also the maximum principle ˜ ˜ H(t. and deﬁne p∗ = (p∗ . p∗ (t∗ ). 1] except possibly for a ﬁnite set. p∗ (t)) ≡ 0. x∗ (t∗ ). if f0 and f do not explicitly depend on t.36) (8. The resulting trajectory is z ∗ (s) = x∗ (t0 + s(t∗ − t0 )). f ˜ By Theorem 1 of Section 2. z ∗ (s). p∗ is not identically zero.30). z ∗ (s). CONINUOUS-TIME OPTIMAL CONTROL ﬁnal condition: p∗ (t∗ )⊥T f (x∗ (t∗ )) . 0 0 f (8. t0 ≤ t ≤ t∗ . the maximum principle λ∗ (s)f0 (t∗ (s).40) First of all. t0 ≤ t ≤ t∗ . v ∗ (s) = u∗ (t0 + s(t∗ − t0 )) and α∗ (s) = (t∗ − t0 ) for 0 ≤ s ≤ 1 0 f f constitute an optimal solution for (8.

w)]|w ∈ Ω}.39) is equivalent to λ∗ (s)f0 (t∗ (s). ﬁnal condition: x(tf ) = xf .46) (8.34) and (8. 0 (8. z ∗ (s). (8. x(t). control constraint: u(·) ∈ U .3.42) (8. and. v ∗ (s)) + λ∗ (s) f (t∗ (s). p∗ (t)) holds for all t ∈ [t0 . x∗ (t).43) In (8.31). Applying Theorem 1 to this problem gives Theorem 2. x∗ (t). Then there exists a function p∗ : [t0 . t0 ≤ t ≤ t∗ . Theorem 2: Let t∗ ∈ (t0 . w) + λ∗ (s) f (t∗ (s). p∗ (tf )) ≥ 0 f and if f does not depend explicitly on t then M (t. z ∗ (s). We consider the following special case of (8. u∗ (t))] p∗ (t).41) and the fact that λ∗ (1) = 0.38) respectively imply (8.8. ˙ f ∂x initial condition: p∗ (t0 ) ∈ Rn . z ∗ (s). ♦ 8. x∗ (t).43). t0 ≤ t ≤ tf ˙ initial condition: x(t0 ) = x0 . Let x∗ (·) be the corresponding f f trajectory. n+1 ˜ Finally. ∞) . t∗ ] except possibly for a ﬁnite set. z ∗ (s). xf are ﬁxed.27): tf Maximize t0 (−1)dt subject to dynamics: x(t) = f (t.35) follows from (8. u∗ (t)) = M (t. so that the optimal control problem consists of ﬁnding a control which transfers the system from state x0 at time t0 to state xf in minimum time. Next.3.37) ˜ and (8. (8. t∗ ] → Ω be optimal. v ∗ (s)) 0 = Sup {[λ∗ (s)f0 (t∗ (s). on the other hand (8. x∗ (t). p∗ . M (t∗ .44) . (t)) ≡ constant . f Also the maximum principle H(t. x0 .45) (8. satisfying f adjoint equation: p∗ (t) = −[ ∂f (t.33). (8. ∞) and let u∗ : [t0 .2 Minimum-time problems . x∗ (tf ). VARIABLE FINAL TIME 107 which is a contradiction. z ∗ (s). p∗ (t).32) and (8. u(t)). f are not explicitly dependent on t. x∗ (t).35) and the fact that M (t. v ∗ (s)) + λ∗ (s) = 0 n+1 and λ∗ (s)f0 (t∗ (s).42) is equivalent to (8. v ∗ (s)) 0 +λ∗ (s) f (t∗ (s). It is trivial to verify that p∗ (·) satisﬁes (8. f Finally. z ∗ (s). not identically zero. ﬁnal condition: p∗ (t∗ ) ∈ Rn . t∗ ] → Rn . ﬁnal-time constraint: tf ∈ (t0 . the last assertion of the Theorem follows from (8.41) Evidently (8. p∗ (t)) ≡ ˜ constant if f0 .

1 2 2 2 1 ∗ α p1 (0) 1 + eαt (− α p∗ (0) + p∗ (0)) .47) where α = (σ/m) > 0 and b = (1/m) > 0.45). CONINUOUS-TIME OPTIMAL CONTROL Exercise 2: Prove Theorem 2. v) = (p∗ (t) − αp∗ (t))x∗ (t) + bp∗ (t)v 1 2 2 2 = eαt (p∗ (0) − αp∗ (0))x∗ (t) + pb∗ (t)v . p∗ (t). 1]. x = 0 in minimum time. u = applied force. x(0) = x02 we wish to ﬁnd an admissible control which brings the ˙ particle to the state x = 0. By Theorem 2 there exists a non-zero solution p∗ (·) of p∗ (t) ˙1 p∗ (t) ˙2 =− 0 0 1 −α p∗ (t) 1 p∗ (t) 2 (8.48) is p∗ (t) 1 p∗ (t) 2 or = e−α(t−τ ) − e−α(t−τ ) ) .47) is 1 0 1 α (1 Φ(t.48) such that (8. x2 = x we rewrite the particle dynamics as ˙ x1 (t) ˙ x2 (t) ˙ = 0 1 0 −α x1 (t) x2 (t) + 0 b u(t) . x ˙ where m = mass. τ ) = so that the solution of (8.44).49) . p∗ (t) ≡ p∗ (0) . We now study a simple example illustrating Theorem 2.108 CHAPTER 8. and x = position of the particle. (8. The control constraint set is Ω = [−1. Example 1: The motion of a particle is described by m¨(t) + σ x(t) = u(t) .46) hold. ˙ Solution: Taking x1 = x. For simplicity we suppose that x ∈ R. σ = coefﬁcient of friction. 1 2 (8. Starting with an initial condition x(0) = x01 . Suppose that u∗ (·) is optimal and x∗ (·) is the corresponding trajectory. 1 1 αt α (1 − e ) 0 eαt p∗ (0) 1 p∗ (0) 2 . (8. u ∈ R and u(t) constrained by |u(t)| ≤ 1. Now the transition matrix function of the homogeneous part of (8. 1 1 and p∗ (t) = 2 The Hamiltonian H is given by H(x∗ (t). and (8.

2 ∗ −1 if p∗ (t) < 0. Hence u∗ (·) we can behave in one of two ways: 1 either u∗ (t) ≡ +1 and p∗ (t) ≡ 2 or u∗ (t) ≡ −1 and p∗ (t) ≡ 2 1 ∗ α p1 (0) 1 ∗ α p1 (0) ˆ ˆ +1 for t < t and p∗ (t) > 0 for t < t. Case 3. First of all since p∗ (t) ≡ 1 can have three qualitatively different forms. Case 1. −p∗ (0) + αp∗ (0) = 0 : In this case p∗ (t) ≡ (1/α)p∗ (0). from (8. 2 ˆ and p∗ (t) < 0 for t > t. 2 109 (8.47) does not depend on t explicitly we must also have eαt (p∗ (0) − αp∗ (0))x∗ (t) + bp∗ (t)u∗ (t) ≡ constant. 2 Case 2. ˆ ˆ +1 for t > t and p2 u∗ (t) = or u∗ (t) ≡ +1 and p∗ (t) > 0 for all t. VARIABLE FINAL TIME so that from the maximum principle we can immediately conclude that   +1 if p∗ (t) > 0. 1 2 2 2 p∗ (0).3. −p∗ (0) + αp∗ (0) < 0 : Evidently u∗ (·) can behave in one of two ways: 1 2 either u∗ (t) = or u∗ (t) ≡ −1 and p∗ (t) < 0 for all t.49) and (8. −p∗ (0) + αp∗ (0) > 0: Evidently then. 2 ∗ (t) > 0 for t > t. p∗ (·) 1 2 (8.50) Furthermore. since the right-hand side of (8. u (t) = 2  ? if p∗ (t) = 0 . Also since p∗ (t) ≡ 0.8.50) u∗ (·) can behave in one of two ways: either ˆ ˆ −1 for t < t and p∗ (t) < 0 for t < t. .51) We now proceed to analyze the consequences of (8. we must 1 2 2 1 have in this case p∗ (0) = 0. <0.49) we see that p∗ (t) must be a strictly 1 2 2 monotonically increasing function so that from (8.50). ˆ −1 for t > t 2 >0.

f f There are at least two ways of solving the two-point boundary value problem (8. Integrating (8. An alternative is to guess at the value of p∗ (0) and then integrate (8.52) and (8. CONINUOUS-TIME OPTIMAL CONTROL Thus.54) is not satisﬁed then modify p∗ (0) and repeat. Suppose we choose p∗ (0) such that −p∗ (0) = αp∗ (0) = 0 and p∗ (0) > 0.53) (8. This gives b ξ1 (t) = α (−t + eαt −1 α ) . and (8. and then t∗ is the minimum time. . 1 1 2 Thus the search for the optimal control reduces to ﬁnding p∗ (0). ξ2 (t) = b α (1 − eαt ) . The optimal control is given by u∗ (t) = sgn p∗ (t) 2 1 1 = sgn [ α p∗ (0) + eαt (− α p∗ (0) + p∗ (0))] .52) for some t∗ > 0.53). which is the curve OB. If (8.54) backward in time give us a trajectory ξ(t) where u ˙ ˙ ξ1 (t) = −ξ2 (t) ˙ ξ2 (t) = αξ2 (t) − b .52). p∗ (0) such that the solution of the 1 2 differential equation x = x2 ˙ 1 1 x2 = −αx2 + b sgn[ α p∗ (0) + eαt (− α p∗ (0) + p∗ (0))] . x20 = x20 also satisﬁes the ﬁnal condition x1 (t∗ ) = 0.53) is satisﬁed.3. On the other hand.54) is satisﬁed.54).53) forward in time and check if (8.54) backward in time and check of (8. x2 (t∗ ) = 0 . (8. Let us follow this procedure. the optimal control u∗ is always equal to +1 or -1 and it can switch at most once between these two values.110 CHAPTER 8.52) and (8. then u∗ (t) ≡ −1 1 2 2 and we get b ξ1 (t) = − α (−t + eαt −1 α ) b .54) (8. One way is to guess at the value of p∗ (0) and then integrate (8. ξ2 (t) = − α (1 − eαt ) . f f (8. ˙ 1 1 2 with initial condition x1 (0) = x10 .52) and (8. The latter approach is more advantageous because we know that any trajectory obtained by this procedure is optimal for initial conditions which lie on the trajectory. if p∗ (0) is such that −p∗ (0) + αp∗ (0) = 0 and p∗ (0) < 0. Then we must have 1 2 2 ∗ (t) ≡ 1. with ξ1 (0) − ξ2 (0) = 0 . which is the curve OA in Figure 8.

(8. Next suppose p∗ (0) is such that −p∗ (0) + αp∗ (0) > 0. Then [(1/α)p∗ (0) + 1 2 2 1 ˆ + p∗ (0))] will have a negative value for t ∈ (0.3.52).54). VARIABLE FINAL TIME u∗ ≡ −1 B C D u∗ ≡ 1 O E 111 ξ1 ξ2 u∗ ≡ 1 A F u∗ ≡ −1 Figure 8.3: Backward integration of (8.52) and (8. We see then that the optimal control u∗ (·) has the following characterizing properties: u∗ (t) = 1 if x∗ (t) is above BOA or on OA −1 if x∗ (t) is below BOA or on OB . This give us the curve OCD.4: Optimal trajectories of Example 1. if we integrate (8. Finally if p∗ (0) is such that −p∗ (0) + 1 ˆ ˆ αp∗ (0) < 0. b for t > t with ξ1 (0) = 0. and we get the 2 2 curve OEF .8. t) and a positive value for t ∈ 2 ˆ (t. Hence. and p∗ (0) < 0. ∞). ξ2 (0) = 0. then u∗ (t) = 1 for t < t and u∗ (t) = −1 for t > t. .54) backwards in time we get trajectory ξ(t) where eαt (−(1/α)p∗ (0) 1 ˙ ξ(t) = −ξ2 (t) ˙ ξ2 (t) = αξ2 (t)+ ˆ −b for t < t ˆ . Hence we can synthesize the optimal control in feedback from: u∗ (t) = ψ(x∗ (t)) where the B u∗ ≡ −1 x2 u∗ ≡ 1 u∗ ≡ −1 O x1 u∗ ≡ 1 A Figure 8. and p∗ (0) < 0.

We apply Theorem 1 of Section 2. 0 ≤ t ≤ T . control constraint: u(t) ∈ Rp . T is a ﬁxed ﬁnal time.112 CHAPTER 8. ˙ initial condition: x(0) = x0 . (8. x2 ) is above BOA or on OA −1 if (x1 . x2 ) is below BOA or on OB . 8. 2 0 If p∗ > 0. The Hamiltonian function is H(t.4 Linear System. and x0 ∈ Rn . Speciﬁcally. T ] → Rn . CONINUOUS-TIME OPTIMAL CONTROL function ψ : R2 → {1. p∗ (t). consider the optimal control problem (8. this will imply 0 u∗ (t) = whereas if p∗ = 0. x2 ) = 1 if (x1 . positive semi-deﬁnite matrix whereas Q(t) is a p × p symmetric.55) subject to dynamics: x(t) = A(t)x(t) + B(t)u(t). x∗ (t).56) (8. bf ∈ R f are given vectors. v) = − 1 p∗ [x∗ (t) P (t)x∗ (t) + v Q(t)v] ˜ 2 0 +p∗ (t) [A(t)x∗ (t) + B(t)v] so that the optimal control u∗ (t) must maximize − 1 p∗ v Q(t)v + p∗ (t) B(t)v for v ∈ Rp .58) cannot have a maximum. In (8.55): T Minimize 0 1 [x (t)P (t)x(t) + u (t)Q(t)u(t)]dt 2 (8. so that we must search for a number p∗ ≥ 0 and a function 0 ∗ : [0. ﬁnal condition: Gf x(t) = bf .58) (t)p∗ (t) . then we must have 0 p∗ (t) B(t) ≡ 0 because otherwise (8. Quadratic Cost An important class of problems which arise in practice is the case when the dynamics are linear and the objective function is quadratic.60) 1 −1 p∗ Q (t)B 0 (8. positive deﬁnite matrix.4) ψ(x1 . such that p p∗ (t) = −p∗ (−P (t)x∗ (t)) − A (t)p∗ (t) . not both zero. u(·) piecewise continuous. (8. −1} is given by (see Figure 8. Gf is a given f × n matrix. ˙ 0 and p∗ (t)⊥T f (x∗ (t)) = {ξ|Gf ξ = 0} .59) .57) (8.56) we assume that P (t) is an n × n symmetric.

0 ≤ τ ≤ T . control constraint: s(t) ∈ [0.56) 0 0 we can see that p∗ (t) = (Φ(T. under the controllability assumption.) Let Φ(t. THE SINGULAR CASE 113 We make the following assumption about the system dynamics. t)) p∗ (T ) and hence from (8. The optimal trajectory and the optimal control is 0 obtained by solving the following two-point boundary value problem: x∗ (t) = A(t)x∗ (t) + B(t)Q−1 (t)B (t)p∗ (t) ˙ p(t) = P (t)x∗ (t) − A (t)p∗ (t) ˙ x∗ (0) = x0 . t)B(t) = 0 . Now if p∗ > 0 it is trivial that p∗ (t) = (1.61) Next we claim that if the system is controllable then p∗ = 0.60) (p∗ (t)) Φ(T. 1]. τ )B(τ ) = 0 .8. (p∗ (t)/p∗ )) will satisfy all the necessary ˆ 0 0 conditions so that we can assume that p∗ = 1. because if p∗ = 0 then from (8. then we must have p∗ (t) ≡ 0 which is a ˜ 0 contradiction. p∗ (t). but then from (8. 0 ≤ t ≤ T initial constraint: k(0) = k0 . In such cases the maximum principle does not help in selecting the optimal value of the control.61) we get p∗ (T ) = 0. Assumption: The control system x(t) = A(t)x(t) + B(t)u(t) is controllable over the interval ˙ [0.59). The problem can be summarized as follows: T Maximize subject to ˙ dynamics: k(t) = s(t)f (k(t)) − µk(t) . x∗ (t). For further details regarding the solution of this boundary value problem and for related topics see (See and Markus [1967]).5 The Singular Case In applying the necessary conditions derived in this chapter it sometimes happens that H(t. Then the controllability assumption is equivalent to the statement that for any ξ ∈ Rn ξ Φ(t. We illustrate this by analyzing Example 4 of Chapter 1. p∗ (T )⊥T f (x∗ (T )) . 8. We are faced with the so-called singular case (because we are in trouble–not because the situation is rare). Hence if p∗ = 0. 0 ≤ t ≤ T . implies ξ = 0 . τ ) be the transition matrix function of the homogeneous linear differential equation x(t) = ˙ A(t)x(t). 0 c(t)dt = T 0 (1 − s(t))f (k(t))dt . v) is independent of v for values of t lying in a non-zero interval. (See (Desoer [1970]) for a deﬁnition of controllability and for the properties we use below.5. Thus. Gf x∗ (T ) = bf . (8. T ]. ﬁnal constraint: k(t) ∈ R . s(·) piecewise continuous. and hence the optimal control is 0 given by (8. p∗ > 0.

p∗ ) we 0 0 0 0 can assume without losing generality that p∗ = 1. p)−. (8.62) k→0 (8.63) is mainly for technical convenience and can be dispensed with without difﬁculty. s) = (1 − s)f (k∗ (t)) + p∗ (t)[sf (k∗ (t)) − µk∗ (t)] is maximized over s ∈ [0. fkk (K) < 0 for all k . which immediately implies that   1 if p∗ (t) > 1 ∗ s (t) = 0 if p∗ (t) < 1  ? if p∗ (t) = 1 We analyze separately the three cases above. Here kG . Assumption (8. if p∗ = 0 then from (8. t)−planes in Figure 8.63) Assumption (8. p∗ (t).64) simpliﬁes to 0 p∗ (t) = −1(1 − s∗ (t))fk (k∗ (t)) − p∗ (t)[s∗ (t)fk (k∗ (t)) − µ] .66) (8. Case 1. Such solutions exist and are unique by virtue of the assumptions (8.63).) < < < > .5. t)− and (p. (See Figure 8. p∗ ) by (1/p∗ )(p∗ . Now suppose that s∗ : [0. and a function p∗ : [0.6. CONINUOUS-TIME OPTIMAL CONTROL We make the following assumptions regarding the production function f : fk (k) > 0. there exist a number p∗ ≥ 0. T ] → [0. lim fk (k) = ∞ . we note from (8. s∗ (t) = 1 : Then the dynamic equations become ˙ k∗ (t) = f (k∗ (t)) − µk∗ (t) .62) that kG < kM . (8. First of all. 1] at s∗ (t).68) is depicted in the (k. 0 ≤ t ≤ T .62) says that the marginal product of capital is positive and this marginal product decreases with increasing capital. kH are the solutions of fk (kG ) − µ = 0 and f (kM ) − µk = 0. p∗ (t) > 1. k∗ (t). T ] → R. such that 0 p∗ (t) = −p∗ (1 − s∗ (t))fk (k∗ (t)) − p∗ (t)[s∗ (t)fk (k∗ (t)) − µ] ˙ 0 with the ﬁnal condition p∗ (T ) = 0 . ˙ The maximum principle says that H(t.62) and (8. not both identically zero.67) The behavior of the solutions of (8.64) and (8.65) (8. and fk (k) − µ > 0 according as k > kG whereas f (k) − µk < 0 according as k > kM . so that (8. Futhermore. (k. Hence we must have p∗ > 0 and then by replacing (p∗ . be the corresponding trajectory of the capital-to-labor ratio. ˙ (8.65) we must also 0 have p∗ (t) ≡ 0.68) (8. Then by Theorem 1 of Section 2.64) and the maximum principle holds.114 CHAPTER 8. p∗ (t) = −p∗ (t)[fk (k∗ (t)) − µ] . 1] is an optimal savings policy and let k∗ (t).

5. s∗ (t) = 0: Then the dynamic equations are ˙ k∗ (t) = −µk∗ (t) .70) . and hence.) Evidently if p∗ (t) = 1 only for a ﬁnite set of times t then we do not have to worry about this case. But then we have p∗ (t) = 0 for t ∈ I so that from (8. so −fk (k∗ (t)) + µ = 0 for t ∈ I .69) (8. p∗ (t) < 1. We face the singular case only if p∗ (t) = 1 for t ∈ I. ˙ giving rise to the behavior illustrated in Figure 8. THE SINGULAR CASE 115 p l fk > µ fk < µ k kM f > µk kG p kM f < µk k t l t Figure 8. p∗ (t) = −fk (k∗ (t)) + µp∗ (t) .8. where I is a non-zero interval.7. (8. Case 3. s∗ (t) =?: (Possibly singular case. or k∗ (t) = kG for t ∈ I . ˙ In turn then we must have k∗ (t) = 0 for t ∈ I so that s∗ (t)f (kG ) − µKG = 0 for t ∈ I . Case 2. p∗ (t) = 1. kG s∗ (t) = µ f (kG ) for t ∈ I .66) ˙ we get −(1 − s∗ (t))fk (k∗ (t)) − [s∗ (t)fk (k∗ (t)) − µ] = 0 for t ∈ I .5: Illustration for Case 1.

For t < t1 we either have p∗ (t) > 1. s∗ (t) = 0 if k > kG . We then have three possibilities depending on the value of k∗ (t2 ): (Bi) k∗ (t2 ) < kG : then p∗ (t2 ) < 0 so that p∗ (t) > 1 for t < t2 and we are in Case 1 so that ˙ ∗ (t) = 1 for t < t .9. p∗ (t) < 1 so that we are in Case 2. T ) such that p∗ (t3 ) = 1.69) and (8.8. In particular we must have k < k . t2 ). or (B) there exists t2 ∈ (0. T ] and then s∗ (t) = 0. t2 ) so that p∗ (t) = 1.65) we know that for t close to T.116 f CHAPTER 8.70). (8. This contradicts the deﬁnition of t2 so that this possibility cannot arise. s 2 0 G (Bii) k∗ (t2 ) > kG : then p∗ (2 ) > 0 but then p∗ (t2 + ε) > 1 for ε > 0 sufﬁciently small and since ˙ p∗ (T ) = 0 there must exist t3 ∈ (t2 . as in Figure 8. The capital-to-labor ratio kG is called the golden mean and the singular solution is called the golden path. k kG kM Figure 8. Thus in the singular case the optimal solution is characterized by (8. s∗ (t) > 1 if k0 < kG .63). for 0 ≤ t ≤ T . CONINUOUS-TIME OPTIMAL CONTROL µk line of slope µ f (k) . We can now assemble separate cases to obtain the optimal control. k∗ (t) = k0 e−µt . First of all. The reason for this term is contained in the following exercise. . or we have p∗ (t) < 1. and s∗ (t) = µ(kG /f (kG )) for t ∈ (t1 .62).6: Illustration for assumptions (8. Show that kG is the unique sustainable capital-to-labor ratio which maximizes ˆ sf (k) ˆ sustainable consumption (1 − s)f (k). (Biii) k∗ (t2 ) − kG : then we can have a singular arc in some interval (t1 . 1] such that ˆ ˆ − µk = 0. The various possibilities are illustrated in Figure 8. T ) such that p∗ (t2 ) = 1 and p∗ (t) < 1 for t2 < t ≤ T . ˆ Exercise 1: A capital-to-labor ratio k is said to be sustainable if there exists s ∈ [0. from the ﬁnal condition (8. k∗ (t) = kG . We face two possibilities: Either (A) p∗ (t) < 1 for all t < [0.

[1967]). Several important generalizations of the maximum principle have appeared. Among the many useful techniques which have been proposed see (Lasdon. (McReynolds [1966]). et al. For applications of the maximum principle to optimal economic growth see (Shell [1967]). et al. et al. (Kelley [1962]). but mathematically difﬁcult. [1971]).. There is no single source of computational methods for optimal control problems. and (Balakrishnan and Neustadt [1964]).. whereas for a discussion of the singular case consult (Kelley.6 Bibliographical Remarks The results presented in this chapter appeared in English in full detail for the ﬁrst time in 1962 in the book by Pontryagin. That book contains many extensions and many examples and it is still an important source. cited earlier. For a less rigorous treatment of state-space constraints see (Jacobson. also consult (Jacobson and Mayne [1970]).8. [1968]). and (Polak [1971]). the derivation of the maximum principle given in the book by Lee and Markus is more satisfactory. et al. BIBLIOGRAPHICAL REMARKS p k 117 l k kG p t l t Figure 8. However. treatment see (Neustadt [1969]). 8.7: Illustration for Case 2. For an applications-oriented treatment of this subject the reader is referred to (Athans and Falb [1966]) and (Bryson and Ho [1969]).6. . For a uniﬁed. On the one hand these include extensions to inﬁnite-dimensional state spaces and on the other hand they allow for constraints on the state more general than merely initial and ﬁnal constraints.

8: Case 3. k kG p kG t 1 t Figure 8. The singular case. CONINUOUS-TIME OPTIMAL CONTROL p k 1 .118 CHAPTER 8. .

9: The optimal solution of example. BIBLIOGRAPHICAL REMARKS 119 p∗ 1 t s∗ 1 t k∗ T T p∗ 1 t s∗ 1 t2 T t k∗ t2 T t Case (A) p∗ T p∗ Case (Bi) t2 T t . t t2 T s∗ 1 s∗ µkG f (kG ) t k∗ kG k0 t k∗ . t Figure 8. . t1 . t t1 t2 T .8. t Case (Biii) . . .6. .

CONINUOUS-TIME OPTIMAL CONTROL .120 CHAPTER 8.

·) : X × U → R. Dynamic programming (DP is a technique which compares the optimal decision with all the other decisions. In (9.1). or ﬁnite-dimensional vector spaces (as in the previous chapters). . 1.. x0 ∈ X is ﬁxed. . This global comparison. X and U may be ﬁnite sets. therefore. for notational convenience we neglect ﬁnal conditions and state-space constraints.e. u(i)) + Φ(x(N )) (9. . 1. The Ωi are ﬁxed subsets of U . However.1) subject to dynamics: x(i + 1) = f (i. Finally f0 (i. The second section deals with the continuous-time problem. x(i). N − 1 . .Chapter 9 Dynamic programing SEQUENTIAL DECISION PROBLEMS: DYNAMIC PROGRAMMING FORMULATION The sequential decision problems discussed in the last three Chapters were analyzed by variational methods. x(i). ·. control constraint: u(i) ∈ Ωi . or even inﬁnitedimensional spaces. i = 0. initial condition: x(0) = x0 . ·) : X × U → X are ﬁxed functions. the necessary conditions for optimality were obtained by comparing the optimal decision with decisions in a small neighborhood of the optimum. besides the fact that it give sufﬁciency conditions. i = 0. N − 1 . 9. 121 . In the ﬁrst section we develop the main recursion equation of DP for discrete-time problems. . The main advantage of DP. .1 Discrete-time DP We consider a problem formulation similar to that of Chapter VI. i. N −1 Maximize i=0 f0 (i. is that DP permits very general problem formulations which do not require differentiability or convexity conditions or even the restriction to a ﬁnite-dimensional state space. ·. u(i)) . The only disadvantage (which unfortunately often rules out its use) of DP is that it can easily give rise to enormous computational requirements. Φ : X → R. leads to optimality conditions which are sufﬁcient. . f (i. . Some general remarks and bibliographical references are collected in the ﬁnal section. the state x(i) and the control u(i) belong to arbitrary sets X and U respectively.

ˆ The value of the objective function corresponding to this control for the problem (9. . x(N ). . . . . We begin with an elementary but crucial observation. u(i)).122 CHAPTER 9. (9. . .1) but with different initial states and initial times. k + 1. initial condition: x(k) = x. . ˆ and the corresponding trajectory. . u(N − 1). u(N − 1) with ˜ ˜ u(i) ˜ u∗ (i) . i = k. x(i). u(i)) + Φ(ˆ(N )) ˆ ˆ x i= i=k N −1 f0 (i. x(i). for each x ∈ X and k between ) and N − 1. . u(i)) + Φ(˜(n)) ˜ ˜ x i=k −1 = > i=k f0 (i.3) f0 (i. i = k. . u(i)) + Φ(ˆ(N )) ˆ ˆ x i= N −1 (9. control constraint: u(i) ∈ Ωi . i = + 1. . . x∗ (k + 1). . . . . x∗ (i). DYNAMIC PROGRAMING The main idea underlying DP involves embedding the optimal control problem (9. k ≤ ≤ N − 1. . N − 1. u(i)) + Φ(x(N )) . . u∗ (i)) + N −1 f0 (i. . . N . . . . Proof: Suppose not. . i = . . x(i). . Lemma 1: Suppose u∗ (k). Then there exists a control u( ).2)k. k + 1. in which the system starts in state x0 at time 0. . u∗ (N − 1) is an optimal control for (9. and control constraint as in (9.1). i = k. with corresponding ˆ ˆ ˆ trajectory x( ) = x∗ ( ). . . x∗ (i).2)k. objective function. . . . .x is N −1 f0 (i. u∗ (N − 1) is an optimal control for (9. such that ˆ ˆ ˆ N −1 f0 (i. x( + 1). x∗ (N ) be the corresponding optimal trajectory. i = k. we will sometimes use the index (9. . N − 1 . − 1 u(i) .2) . starting in state x at time k. .2)k. ·. .x to distinguish between different problems. x(i) . . x(i). x(i). Then for any . . . . and let x∗ (k) = x. .2) subject to dynamics: x(i + 1) = f (i. u( + 1). consider the following problem: N −1 Maximize i=k f0 (i. u∗ (i)) + Φ(x∗ (N )) . . N − 1 . . x(N ) where ˜ ˜ x(i) = ˜ x∗ (i) . u∗ ( ). . is x(k). into a family of optimal control problems with the same dynamics. . . More precisely. Since the initial time k and initial state x are the only parameters in the problem above. u (i)) + Φ(x (N )) .x . x (i). . ∗ ∗ ∗ > i= But then consider the control u(k). .x∗ ( ) .

for any k.4). x( )) ≤ f0 ( . . ·) by (V (N. .x . . u(i)) + Φ(x(N )) . Theorem 1: Deﬁne V (N. x) = Φ(x). x.5) is equal to f0 (k. x( ). k ≤ ≤ N − 1 . . x. .2)k. . x) satisﬁes the backward recursion equation V (k. u( )) + V ( + 1. contradicting the hypothesis. f (k. u(i)) + Φ(x(N )) = f0 (k. 0 ≤ k ≤ N − 1 . u∗ (k)) + V (k + 1. u(k)) i=k N +{ i=k+1 f0 (i. u. f ( . V (k. We have N −1 i=k N −1 f0 (i. and let x∗ (k) = x. xu∗ (k)) + V (k + 1. x. x. and all x ∈ X. u∗ (N − 1) cannot be optimal for 9. f (k. so that u∗ (k). x(i). x(N ). (k)) + V (k + 1. . which is equivalent to (9. x(i).1. u) + V (k1 . u) + V (k + 1. x( ). i.2)k.2)k. u))|u ∈ Ωk } . u∗ (i)) + Φ(x∗ (N )) (9. for all u(k) ∈ Ωk . u(i)) + Φ(x(N )) ≤ f0 (k.9. x∗ ( )). Then ψ(k. (end theorem) From now on we assume that an optimal solution to (9. u∗ (N − 1) be an optimal control for (9. (9.x and let x(k) = x. . ≥ i=k By Lemma 1 the left-hand side of (9. f (k.x exists for all 0 ≤ k ≤ N − 1. u(N − 1) is optimal for (9. . . DISCRETE-TIME DP 123 by (9. x)) + V (k + 1. f (k. x(i). (k. N − 1. . . let ψ(k. We call V the (maximum) value function. Let V (k. . . . . u( )). .x . . .3). u(N − 1) be any control for the problem (9. .2)k. f (k. k. . ))|u ∈ Ωk }.. .(end theorem) Corollary 1: Let u(k). 1. x. Then V ( .x . . x(N ) be the corresponding trajectory. . u∗ (N − 1) deﬁned by u∗ ( ) = ψ( . . ·). x. . x∗ . and equality holds for all k ≤ ≤ N − 1 if and only if the control is optimal for (9. u∗ (k))) ≥ f0 (k. x) = Max{f0 . x. . u(k)) + V (k + 1. . x)) = Max{f0 (k. x. u.2)k.e. . u(k))} .x(k+1) . with equality if and only if u(k + 1). . by the deﬁnition of V we have N −1 f0 (i. ψ(k. x. . x. x the control u∗ (k). where . . ·) : X → Ωk be such that f0 (k. k = 0.4) Proof: Let x ∈ X. . On the other hand. . k ≤ ≤ N − 1. f (x. . x) be the maximum value of (9. x. x. . . Combining these two facts we get f0 (k. u(k))) .5) f0 (i. x∗ (i).2)k . x∗ (N ) be the corresponding trajectory be x(k) = x. u∗ (k)) . Corollary 2: For k = 0.2)k+1. f (k. . x. let u∗ (k). N − 1 is an optimal feedback control. . ψ(k. .

9. For n = 3 this number is 80. As before. Then we have to compute and store 10 × 20 values of V .2): Maximize 0 f f0 (t. u(τ ))dτ (9. u ∈ Rp . Note that this feedback control is optimum for all initial conditions. t0 ≤ t ≤ tf ˙ initial condition: x(0) = x0 . and in evaluating the maximum in (9. u(t))dt + Φ(x(tf )) subject to dynamics: x(t) = f (t. However. x) = Φ(x) . But now suppose X = Rn and we approximate each dimension of x by 20 values.1. x) = Max{ t t+∆ f0 (τ. The instructor assigns a number Ui > 0 for each unit of item i.000.4) analytically.8) . we have to compute and store 10x(20)n values of V . Ω ⊂ Rp . control constraint: u : [t0 . ψ( . x(t). x(t). is optimal for (α)k. tf ] → Ω and u(·) piecewise continuous. and V (tf . x(t + ∆))|u : [t. x(τ ). and for n = 5 it is 32. x∗ (k) = x . For instance.000.4).7) +V (t + ∆.6). These numbers represent the relative utility of that item during the hike. x ∈ Rn . (9. Each unit of item i weighs wi pounds. The recursion equation (9. There are N possible items to choose from. This “curse of dimensionality” seriously limits the applicability of DP to problems where we cannot solve (9. let V (t. x) be the maximum value of the objective function over the interval [t. k ≤ ≤ N − 1 .x .124 CHAPTER 9. • Exercise 1: An instructor is preparing to lead his class for a long hike. unless we can ﬁnd a “closed-form” analytic solution to (9. DYNAMIC PROGRAMING x∗ ( + 1) = f ( . x∗ ( )). the DP formulation may necessitate a prohibitive amount of computation since we would have to compute and store the values of V and ψ for all k and x.000. ∆ ≥ 0 . tf ] starting in state x at time t. which is a reasonable amount. suppose n = 10 and the state-space X is a ﬁnite set with 20 elements. How many units of each item should be placed in each knapsack so as to maximize total utility? Formulate this problem by DP. which is quite impractical for existing computers.4) allows us to compute the value function.4) we also obtain the optimum feedback control. He assumes that each person can take up to W pounds in his knapsack.2 Continuous-time DP We consider a continuous-time version of (9.6) In (9.1. f are assumed to satisfy the conditions stated in VIII. Then it is easy to see that V must satisfy V (t. Remark: Theorem 1 and Corollary 2 are the main results of DP. x∗ ( ). t + ∆] → Ω}. u(t)) . t (9. for t0 ≤ t ≤ tf and x ∈ Rn . Φ : Rn → R is assumed differentiable and f0 . Then for N = 10.

ˆ x (9. Suppose there exists a function ψ : [t0 .6). u) + ∂V f (t. ˙ x(t) = x . tf ] and x ∈ Rn . tf ] × Rn → Ω with ψ piecewise continuous in t and Lipschitz in x. ∂t 125 Dividing by ∆ > 0 and letting ∆ approach zero we get the Hamilton-Jacobi. tf ] → Ω be any piecewise continuous control and ˆ let x(τ ) be the solution of ˆ x (τ ) = f (τ. x∗ (τ ))dτ dτ ∂V ∂V ∗ { (τ. x∗ (τ ).9) Theorem 1: Suppose there exists a differentiable function V : [t0 . CONTINUOUS-TIME DP In (9. x(τ ). x) + ∂V f (t. ψ(τ. ˙ x∗ (τ ) = x . Proof: Let t ∈ [t0 . u(τ )) . x∗ (τ ))) . satisfying f0 (t. x)) ∂x = Max{f0 (t. x(τ ) is the solution of x(τ ) = f (τ. x. ψ(t. x.9. tf ] × Rn → R which satisﬁes (9. t ≤ τ ≤ t + ∆ . x∗ (τ ). ∆ > 0 . x. x. x∗ (tf )) − V (t. x∗ (τ ). u(τ ))dτ + Φ(ˆ(tf )) . ∂x (9. x∗ . u)|u ∈ Ω} . Let us suppose that V is differentiable in t and x. x. t ≤ τ ≤ tf . u(τ )) . x. Let u : [t. x.2. (9.12).12) · (9. x∗ (t)) = f tf =∈t tf =− dV (τ.8). u)|u ∈ Ω} = 0. t ≤ τ ≤ tf . ˆ Let x∗ (τ ) be the solution of x∗ (τ ) = f (τ. Let u∗ (τ ) = ψ(τ. (τ )). x∗ (τ ).14) F − 0(τ.Bellman partial differentiable equation for the value function: ∂V ∂t (t. t . x) = Max{f0 (t. x)f (t.7) we get V (t. (9. x. u∗ (τ ))dτ + Φ(x∗ (τ )) f0 (τ. x)∆ + o(∆)|u ∈ Ω}.7). u∗ (τ ))dτ . x)) + ∂V f (t. ˆ ˆ ˆ x(t) = x . x) + Max{f0 (t.9) and the boundary condition (9. and V is the value function. x∗ (τ ) + x (τ )}dτ ˙ ∂τ ∂x t f (9.11) Note that the hypothesis concerning ψ guarantees a solution of (9. u)∆ + V (t.10) Then ψ is an optimal feedback control for the problem (9. u)∆ ∂x + ∂V (t. u) + ∂V ∂x (t. x(τ ). ψ(t.13) ≤ To this end we note that V (tf . t ≤ τ ≤ tf . To show that ψ is an optimal feedback control we must show that tf t tf t f0 (tτ. Then from (9.

The most elegant applications of DP are to various problems in operations research where one can obtain “closed-form” analytic solutions to be recursion equation for the value function. u∗ (τ )) f0 (τ. See (Bellman and Dreyfus [1952]) and (Wagner [1969]). x) = Φ(x∗ (tf )) + ≥ Φ(ˆ(tf )) + x t tf t tf f0 (τ. (9.13) is proved. On the other hand. where P (t) = P (t) is positive semi-deﬁnite. It also follows that V is the maximum value function. ˙ initial condition: x(0) = x0 .14). x(τ ). [Hint: Obtain the partial differential equation satisﬁed by V (t. For an important application of DP to computational considerations for optimal control problems see (Jacobson and Mayne [1970]). [] .15) using (9. Finally.15). x(τ ).8) and the fact that x∗ (t) = x(t) = x we conclude that ˆ V (t.3 Miscellaneous Remarks There is vast literature dealing with the theory and applications of DP.10). For an excellent introduction to this area of application see (Howard [1960]). the book of Bellman [1957] is still excellent reading. 0 ≤ t ≤ T .9). u(τ ))dτ ˆ ˆ ♦ so that (9. DYNAMIC PROGRAMING { ≤− t tf ∂V ∂V · x (τ )}dτ ˜ (τ. • Exercise 1: Obtain the value function and the optimal feedback control for the linear regulatory problem: Minimize 1 x (T )P (T )x(t) + 2 1 2 T0 {x (t)P (t)x(t) +u (t)Q(t)u(t)}dt subject to dynamics: x(t) = A(t)x(t) + B(t)u(t) . u∗ (τ ))dτ . control constraint: u(t) ∈ Rp . (t)) = ˆ ˆ t tf CHAPTER 9. x. V (tf . and Q(t) = Q (t) is positive deﬁnite. ˆ ˆ (9. x) = x R(t)x where R is unknown.] 9. x∗ (τ ). From (9.9). x) and try a solution of the form V (t. (9. Larson [1968] has developed computational techniques which greatly increase the range of applicability of DP where closed-form solutions are not available.126 using (9. x(tf )) − V (t. (9. In the case of sequential decision-making under uncertainties DP is about the only available general method. x(τ )) + ˆ ∂τ ∂x f0 (τ.

IEEE Trans. Essays in Economics and Econometrics.W. [2] M. on Circuit Theory.A. IEEE Trans. CT-16(3). Athans and P.E. Polak. Director and R. Notes for a Second Course on Linear Systems. PhD thesis. Applied Optimal Control. Rohrer.D. Director and R. Banerjee. IEEE Trans. [15] S. 1960. Dynamic Programming. 1963.V. Theory of Optimal Control and Mathematical Programming. 1969b. [7] D. CT-16(3). [3] A. Bellman. Director and R. College of Engineering. IEEE Trans.A. Berkeley. Desoer. On the design of resistance n-port networks by digital computer. Optimal Control. Applied Dynamic Programming. Neustadt. 1970. Princeton University Press. 1962. Decentralization and Computation in Resource in resource Allocation. Theory of Games and Statistical Decisions. 1971. on Circuit Theory.E. (ed. 1969. McGraw-Hill. Computing Methods in Optimization Problems. Cannon. and E. The function of operations research specilists in large urban schools. [13] S. 1969a. Automated network design–the frequency-domain case. 1970. 1966. [8] J.W. SSC-6(4). Cullum. Blaisdell.W. 1957. 1954. C.Bibliography [1] J.W. 1970. Dantzig. University of North Carolina Press.J. McGraw-Hill.A. on Circuit Theory.L. Princeton University Press. Falb. Bryson and Y.A. Bellman and S. [9] A. Rohrer. in Pfouts R. Balakrishnan and L.W. [12] C. 1969c. 127 . on Systems Science and Cybernetics. Blackwell and M. Academic Press. [4] K. Ho.D.E.E. Princeton University Press. Linear Programming and Extensions. 1964. University of California. [10] J. Generalized Lagrange Multipliers in Dynamic Programming. Girshick. Rohrer. CT-16(3). The generalized adjoint network and network sensitivities. [5] R.C. Arrow and L. Van Nostrand Reinhold. Hurwicz.A. John Wiley. [11] G. [14] S. [6] R. Dreyfus. Bruns.).E.

P.M. John Wiley.). and R. [24] A. [27] H. 1958. 1970. on Circuit Theory.A. J. Surrogate mathematical programming. 17. volume 1. New necessary condtions of optimality for problems with state-variable inequality constraints. [18] R. Optimization Techniques. Frank. Dynamic Programming and Markov Processes.Q. 1960. Karlin. 1962. 1970. Jacobson. [28] R. Operations Research.P. Primal resource directive approaches for optimizing nonlinear decomposable programs. Rinehart. Geoffrion. Samuelson.Solow. in Haveman. Functions of Several Variables. [29] R. R.M. Dorfman. Jacobson and D.H.Kushner. to appear.J. Princeton University Press.M.M. [30] D. Public Expenditures and Policy Analysis. and J. 1965. Apl. Fleming. [26] J. 1971. Differential Dynamic Programming. Mathematical Methods and Theory in Games. Greenberg and W.J..(ed. [17] R. 1970. [32] S. [25] F. 1970a. G. American Elsevier Publishing Co. Differential Games. The Rand Corporation.L. Rohrer. Programming. Automated design of biasing circuits. J. Dorfman and H. . Linear Programming and Economic Analysis. Howard. Analysis and Applications. A geometric duality theorem with economic applications. Geoffrion. IEEE Trans. 18.). and Economics.128 BIBLIOGRAPHY [16] R. 1971. Method of Gradients. 1970b. and Margolis. Dowell and R. XXXIV(1). The Economic Theory of Teams. Speyes. CT-18(1). 1969. Pierskalla. SIAM J. [20] W. Operations Research. Markham Publishing Co. 1971. 18. Math. Extensions of lagrange multipliers in nonlinear programming.A. Mayne. Gale. [19] Dowles Foundation Monograph. Kelley. M. 1959. 1967. 1965.H. Addison-Wesley. Holt. Introduction to Stochastic Control.H. Duality in Nonlinear Programming: a Simpliﬁed Application-Oriented Treatment. in Leitmann.D. McGraw-Hill. Production Theory and Indivisible Commodities.J.J. Academic Press. Lele. A Model of Public Decisions Illustrated by a Water Pollution Policy Problem.A.R. 1969. 1971. [21] C.H. [31] D. Review of Economic Studies. Gould. (eds. Addison-Wesley. Math. John Wiley. [33] H. Isaacs.I. [22] D. [23] A. to appear. Jacoby. Memo RM-6134-PR. MIT Press. and Winston.

Games and Decisions. [40] E. August 11-13 1971.G. W. R. 19. 7.W. Analysis and Applications. Autom.). [49] L. P.BIBLIOGRAPHY 129 [34] J. The conjugate gradient method for optimal control problems.B. Siam J.H.S. Nonlinear Programming.S. 1967. and Y. Recent Advance Programming. In Proc. R. S.L. G. Neustadt. D. Kuhn and A. Rao. Mayer.W.. Takahara. Academic Press. [44] S. Lee and L. Academic Press.K.Mischenko. and Wolfe.F. and H. Econometrica. Markus. on Math. in Leitman. Pontryagin. The Simplex Method for Local Separable Programming. Tucker. [47] C. Boltyanski.E. [35] D.S. Academic Press. Theory of Hierarchical.C. The existence of optimal controls in the absence of convexity conditions. Koopmans.D.V. [50] H. [38] R. Larson. Mitter. Washington U. Miller. 1971. on Automatic Control. 1969. and E. Applied Math. [48] L. McGraw-Hill. Berkeley. Raiffa. Math. [52] E. 1963. 1968.. Math. Foundation of Optimal Control Theory. R. Luenberger. and A. Owen.E.W. McReynolds.R. Meditch. Academic Press. Mesarovic. (eds. 1951. AC-12(1). Mangasarian. Saunders & Co. Game Theory. Kopp. [36] T. [37] H. 1970. 1967. Water quality regulation with multiple polluters. Kelley. (ed. Nonlinear programming. J.S. Waren. 1967. 1957. The successive sweep method and dynamic programming. Kendrick. Statistics and Probability. Polak. Macho. . Nikaido. Louis. H. 1969.D. Stochastic Optimal Linear Estimation and Control. R. 1967.E. [45] J. 1969. 1970.V. Computational Methods in Optimization: A Uniﬁed Approach. and C. and outcomes in optimal growth models.. [41] R. Neustadt.W.G. Lasdon. Interscience. Wells. 1968. 1968. 1963. constraints. McGraw-Hill. 3(1). [53] L. John Wiley.H. Topics in Optimization. J. John Wiley. Quasi-convex programming. Singular Extremals. A general theory of extremals.L. 16. J. University of California Press.). Second Berkeley Symp. In Proc. in Graves. [39] L. St. Luce and H. Convex Structures and Economic Theory. Analysis and Applications. 1968. Gamkrelidze. 35(1). Computer and System Sciences.A. Multi-level Systems. Control Conf. [46] M. Objectives. 1971 Jt.. The Mathematical Theory of Optimal Processes. [51] G. IEEE Trans. American Elsevier Publishing Co. 1962.B. McGraw-Hill. [43] O. State Increment Dynamic Programming. [42] D.

1967. [56] L. 1969. Topkis and A. Veinnott Jr. Rockafeller. [59] D. . BIBLIOGRAPHY [55] M.M. 1954. Zangwill. [60] H. Solow. SIAM J. 6(2). Wonham. [62] W. John Wiley. Van Nostrand Reinhold. Science. [57] K.M. 1971. Wolfe. 1970. On the seperation theorem of optimal control. [58] R. [61] P.T. Convex Analysis.M. Savage. 1971. 173(3996).M. Essays in the Theory of Optimal Economic Growth. Principles of Operations Research. SIAM J. Wagner. Prentice-Hall. Nonlinear Programming: A Uniﬁed Approach. 1967. On the convergence of some feasible directions algorithms for nonlinear programming.J. The Foundation of Statistics. 1959. Econometrica. 1969. 27. Notes on Linear Programming. MIT Press. [63] E. on Control. Sakarovitch. Prentice-Hall. 1968. Princeton University Press. The economist’s approach to pollution and its control. on Control. 5(2).130 [54] R. The simplex method for quadratic programming. Shell.

123.H. LP duality theorem. 23 Afﬁne function. 54 Langrangian multipliers.LP optimality condition. 124 Epigraph. DP optimality conditions. 80 Dual problem. 72 algorithm. 8 Hamilton-Jacobi-Bellman equation. 45. 49 Game theory . 101. 125 ˜ Hamiltonian H. 5 Gradient. 35 Langrangian function. 21 Linear programming. 78. 39 basic variable. 37 properties. 105 continuous-time. 91. 78 problem formulation. 33. 80 Minimum fuel problem. 35 problem formulation. 37 Derivative. 39 Certainty-equivalence principle. 91. 103 problem formulation. 55 Convex set. 33. 101.Index Active constraint. 55 Continuous-time optimal control necessary condition. 85 Adjoint equation augmented. 34 Constraint qualiﬁcation deﬁnition. 37. 67 Convex function deﬁnition. 91 discrete-time. 42 Linear programming. 5 Complementary slackness. 71 Feasible solution. 21. 124 Lagrange multipliers. 64 Farkas’ Lemma. 32 Feasible direction. 80 Adjoint network. 58 131 Duality theorem. 101 Hypograph. 121. 34 Maximum principle continuous-time. 77 sufﬁcient condition. 123 Discrete-time optimality control sufﬁcient condition. 125 problem formulation. 103 discrete-time. 107 . 8 Design of resistive network. 37 Lagrangian function. 33. 98 continuous-time. 103 sufﬁcient condition. 54. 101. 53 sufﬁcent conditions. 86. 15 Discrete-time optimal control necessary condition. 61 Equilibrium of an economy. 61 Knapsack problem. 125 Control of water quality. 31 theory of the ﬁrm. 99 ˜ Hamiltonian HH. 81 Minimum-time problem. 33. 63 Dynamic programming. 54 Basic feasible solution. 50 Adjoint Equation augmented.

QP optimality condition. 84 Tangent. 70 Shadow-prices. 113. 39 Nonlinear programming. 60 Supporting hyperplane. 117 discrete-time problem. 117 Optimal feedback control. 1 Optimal economic growth. 63 necessary condition. 81. 33 Quadratic cost. 71 Recursion equation for dynamic programming. 38 Weak duality theorem.132 example. 113 Slack variable. 41 Phase II. 77 Subgradient. 125 Optimization over open set necessary condition. 81. 123 Variable ﬁnal time. 70 problem formulation. 2. 37. 73 Separation theorem for stochastic control. 65 Separation theorem for convex sets. 17 sufﬁcient condition. 112 Resource allocation problem. 123. 54 Optimal decision. 39 Simplex algorithm. 108 Non-degeneracy condition. 45. 49 suﬁcient condition. 11 sufﬁcient condition. 4 Optimization with equality constraints necessary condition. 5 Shadow prices. 124 Regulator problem. 33. 70 Primal problem. 58 Wolfe algorithm. 60 Supergradient. 39 Singular case for control. 71 INDEX . 32 State-space constraint continuous-time problem. 70 Wolfe algorithm. 50 Transversality condition continuous-time problem. 37 Phase I. 80 Value function. 112 Quadratic programming. 21 Optimum tax. 50. 61. NP duality theorem. 91 discrete-time problem. 13 Optimization under uncertainty. 103 Vertex. 53 problem formulation.