Urban Transportation for the 21st

Century

Lecture in the Later Life Learning Series,

Urban Planning
University of Toronto
March 4, 2016
Eric J. Miller, Ph.D.
Professor, Dept. of Civil Engineering
Director, UTTRI
University of Toronto

Presentation Outline
Transportation & cities: a very brief histroical
overview.
The physics of transportation systems.
Urban form and travel behaviour.
Accessibility.
Some implications.

Throughout history the location, size, shape and economic & social functioning
of cities has been fundamentally influenced by transportation technology,
infrastructure and services.
Transportation networks literally give spatial definition to land and,
by making the land accessible, they make it useful.
To date we have experienced 3 eras of
transportation technology and, hence,
Central Vancouver
urban form.
Montreal

1. The Pre-Industrial
City
Pre-industrial cities were small, dense
and extremely compact.
Walking and animal-powered
vehicles were the only options for
intra-city travel:
To be accessible, everything
had to be close at hand.
Wind and animal-powered
transport were the only inter-city
options:
Cities were largely tied to
rivers, lakes and ocean ports:
water-based travel was so
much faster and cost-effective
than land travel.
Largely a person-based (individual)
supply of transportation services.

2. The Industrial City

With the Industrial Revolution steam engines
and, later, electrically-powered vehicles made
mechanized transport possible for the first time,
providing much higher speed and higher
capacity transport services. This permitted the
size of cities to expand dramatically, leading to:
Lower (but by modern standards still high)
densities.
Vastly larger cities.
The ability to accommodate much larger
populations (& economic activity).

3. The Automobile City

In the 20th Century, with its much higher speeds, flexibility of
use and its provision of an affordable personal means of
transportation, the auto again revolutionized
transportation by freeing people from fixed-service public
transit and freeing urban development to further spread out
(sprawl) in a dramatically increased, indeed explosive, way.
Suburbs first started to develop with the introduction of
commuter railways and (even more so) electric urban street
railways. But the private automobile made possible the
development of the post-WWII low density mass suburbs
that characterize much of current North American cities.

Toronto

Delhi

Bristol, UK

Beijing

Dubai

As cities grow, they simply cannot

build enough roadways to carry all the
traffic.

Dallas

Fundamental Flow Equation

v = average speed (m/min)
k = average density (veh/m)
q = average flow (veh/min)
A = XT (m-min)
q = (Si xi)/A
k = (Si ti)/A

or, alternatively: q = (Ndx)/(Tdx) = N/T

or, alternatively: k = (Mdt)/(Xdt) = M/X

v = (Si xi)/(Si ti) = [(Si xi)/A][(Si ti)/A] = q/k or, q = vk

Average Flow
= Average Speed * Average Density
This relationship governs the movement of all
vehicles, including bicycles (pedestrians too!).

It fundamentally constrains roadway capacity

and achievable speeds.

Roadway Capacity & Level of Service

A roadways capacity is an
emergent outcome of vehicle
interactions, given vehicle
technology and roadway
geometric design.

Changes in technology &

roadway design can change
the parameters of this
relationship but cannot change
its fundamental nature.
One lane of highway can
carry at most 2400
cars/hour under ideal
conditions

The Traffic Control Problem

The challenge of traffic
control is to keep flow levels
in the undersaturated
regime, which maximizes both
speeds and throughput.

As flow approaches
capacity it becomes
turbulent and breaks
down into an
oversaturated regime.
Not only do speeds
dramatically decline, but
the capacity of the
roadway actually
decreases significantly
as well.

Travel time (minutes)

Because of these fundamental limits of roadway

capacity and performance, over & above
environmental and land use impacts, roadways
become very inefficient means of moving people
as they become increasingly congested.
Increasing delay

Increasing congestion
Free flow travel time

Capacity

Roadway traffic volume (cars/hour)

For short trips, walking & biking are

by far the best way to move people.
But in a large urban area, many trips
are too long to be viable by walk or
even bike.

Public transit can be

much more efficient in
carrying large volumes
of people.

Torontos Yonge St.

subway has a capacity
of up to 35,000
persons/hour. This is
the equivalent of up to
14 lanes of highway!

Tokyo

London

New York

An excellent transit
system is an
essential
component of
every great city:
roads along cannot
possibly carry the
trip volumes.
Paris

Transit System Performance

Walk times to/from transit depend
on:
Stop spacing
Network density
Wait & transfer times depend of
service frequencies. Maximum
achievable frequencies depend on:
Station dwell times
Control system
Vehicle technology

In-vehicle travel times depend on:

Stop-to-stop speeds (vehicle
technology)
Stop dwell times
Stop spacing (short spacing
reduces average speeds)
Passenger boarding/alighting
times
Transit line capacity depends on:
Service frequency
Vehicle capacity

As with roads, there are fundamental

limits to the speeds & capacities
achievable with transit, as defined by
transit vehicle and guideway
technology and the routes control
system.
Capacities, however, can be much
higher than for roads.

15

Transit Networks
In order to provide connectivity, coverage and high quality
service levels, the transit network must be designed in a
hierarchical fashion (high capacity trunk lines, feeder
services; long-distance line-haul, local accessibility).
Travellers need to be able to get to/from high-order
services.
Door-to-door service is required to compete with
the auto.
Transit is built line by line, but it is the network that
makes it work.
Munich Rail System

Local bus
Express Bus, LRT, etc.
Subway, Commuter rail, etc.

Transit Usage
Travel times and costs, among other factors affect trip-makers
choice of travel mode.
For transit, walk, wait and in-vehicle time all affect transit usage.
The utility of each mode can be expressed as:
V(transit) = b1 + b2*(In-vehicle time)
+ b3*(Wait time)
+ b4*(Walk time)
And then the probability that a person takes transit for a given trip
can be expressed as:
P(transit) = eV(transit) / Sm eV(m)

Prof. Daniel McFadden 2000 Economics

Nobel Laureate for development of
discrete choice modelling

Probability of Taking Transit

The challenge is to build a transit

system that is sufficiently competitive
with the automobile to provide an
attractive alternative, in a costeffective manner.

Unfortunately, a considerable majority of trips in typical North American cities

are made under conditions that are not transit-supportive.

19

The viability of transit is

tied directly to urban form:
how we build our urban
region directly determines
how much and what type of
transit will be costeffective.

Urban Activity & Transportation

Key System Elements
System Interactions
T transport system
T

III

II
F

Source: Manheim, M.L. (1978) Fundamentals of Transportation Systems Analysis

Volume 1: Basic Concepts, MIT Press

A activity system
F flows & transport
system performance
System Interactions/Feedbacks
I Market demand-supply
interactions determine
flows & system performance
II System performance
(accessibility) influences
activity system markets
III Govt, public & private
service providers respond to
system demand & performance

Number of Vehicles per Household

of

Ol
d
Ne Tor
w on
To to
ro
Du nto
rh
am
M
Y
iss
or
Re issa k
st ug
of a
Pe
Ha el
Re
lt
st
of Ham on
Ha ilt
m on
-W
Ni ent
a
W gara
at
er
l
Gu oo
W elp
el
h
Or ling
t
an on
ge
vil
l
Ba e
rri
Si e
m
Pe
co
Pe ter
V
te bo ict e
rb
or roug oria
ou
g h h Ci
Co ty
u
Ex nty
tre
na
l
st

Re

Average Vehicles per Household

Household auto ownership

levels vary with urban
form, with fewer autos per
household in denser, older
urban areas.
2.00

1.60

1.20

0.80

0.40

0.00

Region of Residence

1.8

1.6

1.4

1.2

0.8

0.6

0.4

0.2

0
0-1000
1000-2000
2000-3000
3000-4000
4000-5000

Zonal Household Density (households/sq. km.)

5000-6000
6000+

First Trip to Work

Trip Length
18
16
14
12
10
8
6
4
2
0
Toronto

Durham

York

Peel

Region of Household

Halton

Hamilton

Average Daily VKT per Household

Median Trip Length (km)

Trip lengths & total auto usage

also vary with urban form.

1986
1991
1996
2001

20

100
90
80
70
60
50
40
30
20
10
0
0-1000

1000-2000 2000-3000 3000-4000 4000-5000 5000-6000

Zonal Household Density (households/sq. km.)

6000+

So too does environmental impact.

 and average annual

transportation costs per
household

1996 Average Annual Total Travel Costs

0 - 6000
6000 - 10000
10000 - 14000
14000 - 20000
20000 - 45000
No Data

Macro vs. Micro Design

Average Annual GHG Travel Emissions Per Household
12000

Annual GHG Emissions Per Hhld

(kg CO2 equivalent)

11000

10000

9300
8700

8400

Neighbourhood Designs

8000
7000

Nbhd 1

6100
6000

Nbhd 2
5000

Nbhd 3

4500
4000

1. Conventional suburban
2. Medium density
3. Neo-traditional

3500

2000

0
Inner City

Inner Suburb

Outer Suburb

Neighbourhood Location

Source: Greenhouse Gas Emissions from Urban Travel: Tool for Evaluating Neighbourhood Sustainability,
Prepared by IBI Group for CMHC and Natural Resources Canada, Feb. 2000

Both are important in the determination of travel demand and transportation

system sustainability. But, macro location effects tend to dominate micro
neighbourhood design impacts.

Urban form is
defined at both a
macro level
(spatial
distribution of
people, jobs,
activities land
use) and the
micro level of
detailed
neighbourhood
design (street
layouts, density,
fine-grain mix of
uses, etc.).

Accessibility
The role of a transportation system is to provide accessibility
to land, people and activities.
Accessibility embodies the concept of the capability or the
potential to participate in activities that are dispersed in both
space and time. This can be contrasted with that of mobility,
which is concerned with the realization of this potential in
terms of actual travel from place to place.
Several measures are used to quantify the concept of
accessibility. These measures all are:
defined for a specific point in space
a function of the magnitude/attractiveness of alternative locations
a function of the distance/time required to reach these locations (and
hence vary by mode of travel and time of day)

Accessibility
The role of a transportation system is to provide accessibility
to land, people and activities.
Accessibility embodies the concept of the capability or the
potential to participate in activities that are dispersed in both
space and time. This can be contrasted with that of mobility,
which is concerned with the realization of this potential in
terms of actual travel from place to place.
Several measures are used to quantify the concept of
accessibility. These measures all are:
defined for a specific point in space
a function of the magnitude/attractiveness of alternative locations
a function of the distance/time required to reach these locations (and
hence vary by mode of travel and time of day)
Land use & transportation intertwined
within the concept of accessibility

A Simple Accessibility
Measure

1996 - Fraction of total GTA Emp within 30 min drive (am)

0 - 0.1
0.1 - 0.2
0.2 - 0.4
0.4 - 0.6
0.6 - 0.8

Number (or fraction) of jobs

(other activities) within x km
(or min.) of a point:

x
i

Ai =

Ej

Jobs w/in 30 minute auto drive time

jSx|i
1996 - Fraction of total GTA Emp within 30 min transit (am)
0 - 0.01
0.01 - 0.03
0.03 - 0.07
0.07 - 0.14
0.14 - 0.23
No Data

Ai = Accessibility of zone i to employment

Ej = Employment in zone j
Sx|i = Set of employment zones within x km
of zone i

Jobs w/in 30 minute transit travel time time

Accessibility & Consumer Surplus

In Random Utility Theory, a logical definition of accessibility is the expected utility of a
trip for a given purpose. Since trip-makers are assumed to choose their best alternative, this
means that accessibility is defined as the expected maximum utility of the trip.
For the special case of multinomial logit models, this expected maximum utility takes on a
convenient analytical closed form. I.e., if the underlying spatial choice process is given by:
P(j|i) = exp(Vj|i/f) / S exp(Vj|i/f)
jCi,m

then the expected maximum utility associated with this choice (and, hence the accessibility)
is the so-called inclusive value or logsum term:
A(i) = loge [ S exp(Vj|i/f) ]
jCi,m

It can be shown that this term is also the consumer surplus for this choice, thereby
providing a direct link between accessibility and social welfare.

Developing a transportation system that maximizes

accessibility maximizes consumer surplus!

The Auto Accessibility Dilemma

The appeal of the car is that, in the absence of
congestion, it provides high-speed mobility; i.e.,
access to activity opportunities over long distances.
BUT:
The auto-based transportation network encourages
low-density sprawl. Thus, the density of activities
that can be reached decreases. Does accessibility go
up or down?
Congestion reduces the travel times considerably.
Accessibility definitely declines as networks become
increasingly congested without associated changes in
land use as urban areas grow.

31

Policy Implications

Maximizing accessibility requires:

A multi-modal approach in which
technologies and services are balanced
within markets.
Integrating urban design/development
with transportation system design.

Information technology is having an

increasing impact on both system
performance and travel behaviour it
brings a new dimension to
accessibility.
New transportation technology (notably
autonomous, connected vehicles) will
undoubtedly have significant impact,
but will not be a silver bullet.
Transit applications need serious
exploration (especially wrt the first/last
mile problem).

New service concepts (auto/bike-share,

etc.) also can significantly change the
accessibility frontier
Affordability
Ownership
Parking

Big, successful cities will always be

congested. The challenge is to find the
right balance between modes and an
urban form that is sustainable (as
opposed to pathological) from a
transportation perspective while also
meeting a wide variety of social &
economic goals.
32

A post-auto city requires a

fundamental rebalancing of our usage
of private (car) versus public (transit)
and mechanized (car, transit) versus
non-motorized (walk/bike).

Zurich

Copenhagen

Cars will still be a major

mode of transport, but
they need to be electric.
And they will be
increasingly autonomous.

GM Volt

But these technological improvements will not solve the fundamental

challenge of how to maintain and improve quality of life, economic
productivity, and environmental sustainability as our urban areas
continue to grow.
Rather than primarily technological, the 21st-century revolution
requires a cultural and political shift in how we view our cities and a
fundamental re-think of how we design and build them. We need to
design cities for people rather than cars. We need to think holistically
about streets and their myriad of functions they are more than mere
conduits for motorized vehicles.

And, most important, we need to incorporate this thinking into the design of every
new suburb, brownfield redevelopment, and master plan. We need to recognize that
the design principles of the 20th-century auto-first city are not working and have not
been working for some time. The longer we cling to them, the more we will
compound our problems and the less likely we will be able to build the cities that we
and our children need. In many cases, this means
a return to earlier, traditional forms of
neighbourhood design.

 Need a coordinated,
systematic plan for both
the short and long runs.
Design & build
networks, not
individual lines.
Increase frequency.
Improve reliability:
give transit priority
Exclusive right-ofways (ROW).
On-street signal
priority.
Choose technology
appropriate to the task
at hand.
Put all the pieces
together.

Keys to Better Transit

Transportation
Building
Blocks
4. Transportation
Building
Blocks
Road
Improvements

Parking

Standards

On Street
Regulations

Widenings

New
Roads

Pricing

Traffic
Control

Pedestrians
& Cycling

Transit
Priority

Reserved
Lanes

Turn
Restrictions

Reduced
Parking

Signal
Preemption

Transit
Improvements

New Rapid
Transit

Frequency
of Service

New
Routes

Alternative
Services

Fare
Integration

Modest Investment

Policy

Source: City of Toronto Official

Plan

Short- and LongRun Plans

Incentives

Removing
Impediments

Intensification

Mixed
Uses

Local
Urban
Design

Enforcement

Capital Intensive

Land Use
Planning

Access for
Disabled &
Seniors

Urban regions need a

systematic, comprehensive
plan for both the short and the
long run.
In the short run, much lowhanging fruit can be
undertaken at low to modest
cost.
At the same time we must
be working consistently on
implementing longer-term,
bigger-ticket improvements.

If I were Czar
Every new development and every
redevelopment would have to:
Incorporate transit and active transportation
supportive design.
Be designed to facilitate the cost-effective
expansion of transit services and networks.
Show how it will increase transit and active
transportation mode shares.

Congestion is not solely a transportation

problem. It is fundamentally a problem in
urban design.

Example: City of Vancouver

Target: 2/3 of all
trips originating in
CoV in 2040 will be
made by foot, bike
or transit.
For this to happen
the absolute number
of auto trips must
decline and all net
growth in travel
must be by non-auto
modes.

40

A sustainable urban
transportation system
must be built upon four
pillars:
Good governance
ECONOMY
Adequate &
sustainable financing

ENVIRONMENT
The Four Pillars of Sustainable Urban
Transportation, Kennedy, et. al (2005)

SOCIETY
SUSTAINABLE URBAN TRANSPORTATION

which lead to
Sound infrastructure
Good urban form and
neighbourhood design

GOVERNANCE

FINANCING

NEIGHBOURINFRAHOODS
STRUCTURE

A sustainable urban
transportation system
must be built upon four
pillars:
Good governance
ECONOMY
Adequate &
sustainable financing

ENVIRONMENT
The Four Pillars of Sustainable Urban
Transportation, Kennedy, et. al (2005)

SOCIETY
SUSTAINABLE URBAN TRANSPORTATION

which lead to
Sound infrastructure
Good urban form and
neighbourhood design
ACCESSIBILITY!
GOVERNANCE

FINANCING

NEIGHBOURINFRAHOODS
STRUCTURE

The sustainability
and liveability of
our cities depends
on us acting now
to build the better
city of tomorrow.
Failure to do so is
not an option.

Thank you for

your attention.
Questions?