MBA 643 - Analyzing Quantitative Models

I
i f ' ~ AJ.,.;:,ll U...l,>JI

CANADIAN UNIVERSITY OF DUBAI
Quantitative Analysis for Managers (MBA 643)
Section 2 February 2014
Assignment
Name (Print Clearly): ____________ Student 10 Number _____ _
Name (Print Clearly):
____________ Student 10 Number _____ _
Due date: February 25, 2014
The objecti ve of this research assignment is to identify one research paper where each one of the
techniques listed below has been used. You will need to use the database resources available
at CUD librar-y.
1. Multiple Linear Regression Analysis
2. Linear Programming
3. Network Models
For each identified paper, you are required to explain in a short paragraph ( with a maximum of
I 0 li nes) how the technique mentioned in 1, 2 or 3 above has been used.
P.S: Only two students may submit the same assignment.
MBA 643 - Canadian University ofDubai - February 2014
Hussein Akberali S1314000563

Canadian University of Dubai
MBA 643-2 QAM
Assignment 1 Analyzing Forecasting, Linear Programming and Networking Models
Hussein Akberali
2/23/2014

American Journal of Applied Sciences, 10 (4): 322-330, 2013
ISSN: 1546-9239
2013 Science Publication
doi:10.3844/ajassp.2013.322.330 Published Online 10 (4) 2013 (http://www.thescipub.com/ajas.toc)
Corresponding Author: Noraini Ibrahim, Department of Industrial Computing and Modelling Mathematics,
Faculty of Computer Science and Information Systems, 81310, UTM Johor Bahru, Johor, Malaysia

322
Science Publications

AJAS
PARTIAL LEAST SQUARES
REGRESSION BASED VARIABLES
SELECTION FOR WATER LEVEL PREDICTIONS
Noraini Ibrahim and Antoni Wibowo

Department of Industrial Computing and Modelling Mathematics,

Faculty of Computer Science and Information Systems, 81310, UTM Johor Bahru, Johor, Malaysia

Received 2012-05-25, Revised 2012-07-30; Accepted 2013-04-25
ABSTRACT
Floods are common phenomenon in the state of Kuala Krai, specifically in Kelantan-Malaysia. Every year,
floods affecting biodiversity on this region and also causing property loss of this residential area. The
residents in Kelantan always suffered from floods since the water overflows to the areas adjoining to the
rivers, lakes or dams. Months, average monthly rainfall, temperature, relative humidity and surface wind
were used as predictors while the water level of Galas River was used as response. The selection of suitable
predictor variables becomes an important issue for developing prediction model since the analysis data uses
many variables from meteorological and hydrogical departments. In this study, we conduct K-fold Cross-
Validation (CV) to select the important variables for the water level predictions. A suitable prediction
model is needed to forecast the water level in Galas River by adopting the Ordinary Linear Regression
(OLR) and Partial Least Squares Regression (PLSR). However, we need to perform pre-processing data of
the datasets since the original data contain missing data. We perform two types of pre-processing data
which are using mean of the corresponding months (type I pre-processing data) and OLR (type II pre-
processing data) of missing data. Based on the experiment, PLSR is more suitable model rather than OLR
for predicting the water level in Galas River and the use of the type I pre-processing data gives higher
accuracy than the type II pre-processing data.

Keywords: Cross-Validation (CV), Ordinary Linear Regression (OLR), Partial Least Squares Regression
(PLSR), Galas River, Water Level
1. INTRODUCTION
Floods are common phenomenon which can be
defined as the presence of excess of water in the place
that is normally dry. Floods are often cited as being the
most lethal of all natural disasters (Noji, 1997;
Alexander, 1993; Jonkman and Kelman, 2005). The
flooding of Malaysian rivers is mainly due to the high
amount of rainfall in river basins because of the climate
is greatly influenced by the monsoon winds. The worst
flood in Malaysia was recorded in 1926 which has been
described as having caused the most extensive damage
to the natural environment. Subsequent major floods
were recorded in 1931, 1947, 1954, 1957, 1967 and
1971. Most countries in Malaysia suffer from floods
during monsoon season especially in Kedah, Kelantan,
Terengganu, Pahang and Johor. Kelantan is a state in
the east coast of Peninsular Malaysia that has never
missed a flooding event, which occurs every year
during the northeast monsoon period.
Floods affect many of the engineering structures
such as bridges, embankments, tanks, reservoirs and
significantly disrupt or interfere with human and societal
activity. Kuala Krai is one of the districts in Kelantan
that always affected with flood. The factor that cause
flood at Kuala Krai district of Kelantan state was due to
Noraini Ibrahim and Antoni Wibowo / American Journal of Applied Sciences, 10 (4): 322-330, 2013

323 Science Publications

AJAS
a combination of physical factors such as elevation and
its close proximity to the sea apart from heavy rainfall
experienced during monsoon period. The severe floods
all over Kelantan are resulted from heavy rainfall
during the north east monsoon season especially in
November and December. In order to facilitate the
prediction of flooding in the river and the warning
beforehand, this study aims to build a model on the
relation between the selected predictors and the water
level of Galas River by adopting the OLR and PLSR.
Kelantan state consists of more than 25 rivers and
having seven major river basins that are Galas, Kelantan,
Golok, Semerak, Pengkalan Chepa, Pengkalan Datu and
Kemasin river basins. Kelantan River Basin is the
biggest river basin in Kelantan and it drains a catchment
area of about 12,000 km
2
in north-east Malaysia
including part of the National Park and flows northwards
into the South China Sea (Rohasliney, 2010). The
Kelantan River is about 248 km long and occupying
more than 85% of the State of Kelantan. It divides into
Galas and Lebir Rivers near Kuala Krai, about 100 km
from the river mouth which means that Kelantan River is
the main river while Galas and Lebir Rivers are the
tributary rivers. For this study, we focused on one main
tributary of Kelantan River which is Galas River in
Kuala Krai, Kelantan. Figure 1 shows the location of the
study area which is Galas River.
The data for this analysis are collected from Water
Resources Management and Hydrology Division and
Malaysian Meteorological Department. It is noticed that
the original data contain missing data. Missing data are
common issue for data quality and most real datasets
consist of missing data. There are four types of serious
data quality problems in real datasets which are
incomplete, redundant, inconsistent and noisy data.
Based on our observation, the data has incomplete data
which missing values in certain months. Due to the
presence of missing data, the two methods can be
inappropriate to be used directly for water level
prediction, therefore, we need to perform a pre-
processing data of the dataset. There are five factors that
were identified and related to the level of the Galas River
which can lead to the occurrence of flood phenomenon
in Kuala Krai, Kelantan: (1) Months from January until
December for 11 years starting from 2001 until 2011, (2)
Monthly mean of rainfall, (3) Monthly mean of
temperature, (4) Monthly mean of relative humidity and
(5) Monthly mean of surface wind.

Fig. 1. Galas Rivers


AJAS
Different approaches for water level predictions can
be found in hydrology science literature. The most
common approaches for predicting the water level are
stepwise regression (Zou et al., 2010), Artificial Neural
Network (ANN) (Bustami et al., 2007) and ANN
combined with PLSR (Shu et al., 2008). The previuos
research on floods monitoring was conducted which is
the utilizing of the GPS data for monitoring the severe
flood in Kuala Krai Kelantan in order to detect the
influence of heavy rainfall towards severe floods
(Suparta et al., 2012). Another previuos research on the
influence of groundwater flow systems towards climate
change was reviewed to recommend the solutions that
are more economical and enviromentally in managing the
flooding water (Carrillo-Rivera and Cardona, 2012). A
number of papers have previously reviewed on variables
selection such as N-PLSR as empirical downscaling tool
in climate change studies (Bergant and Kajfez-Bogataj,
2005) and application of PLSR as downscaling tool for
Pichola lake in India (Goyal and Ojha, 2010). PLSR is
successful mostly in chemometrics since the origin of
PLSR lies in chemistry. It is useful when the factors are
many highly collinear for constructing predictive
models. In this study, we apply this method for variables
selection to develope water level models. The following
sections present an approach to the development of the
water level models. Materials and methods are discussed
in Section 2 while the results are described in Section 3.
The discussion is reported in Section 4 and finally the
conclusion is given in Section 5.
2. MATERIALS AND METHODS
The linear regression model is given as in Equation
(1) (Mevik and Cederkvist, 2004):

n 0 1
y 1 X = + + (1)

where, y is an n1 vector of observations on the
response variable, X is an np matrix consisting of n
observations and p predictors,
0
is an unknown
constant,
1

is an p1 vector of unknown regression
coefficient,1
n
is an n1 ones vector and is an n1
vector of errors identically and independently distributed
with mean zero and variance
2
>0, respectively.
2.1. Ordinary Linear Regression
OLR often being used in fitting models to make an
observation which is applied by minimizing the sum of
the squared residuals between the predicted and actual
response. When matrix X
1
= [1
n
X] has full rank of p, the
OLR estimator of
1
T T
0
( , ) = , say
T
OLS OLS OLS OLS
p 0 1

[ , , , ] =
, is estimated in Equation (2):

T 1 T
1 1 1 OLS
(X X ) X y
(2)

The prediction of y is given in Equation (3):

1 OLS
y X =

(3)

The model for OLR can be represented by
Equation (4):

1 p OLS0 OLS1 OLSp

f (x) x x = + + + (4)

where, x = [x
1
,x
2
,,x
p
]
T
R
P

2.2. Partial Least Squares Regression
Partial Least Squares (PLS) has been proven to be
an effective approach to solve the problems in
chemometrics such as by predicting the bioactivity of
molecules to facilitate discovery of novel
pharmaceuticals. The PLS approach was originated
around 1975 by Herman Wold for modeling the
complicated datasets in terms of matrices blocks which
called path models (Joreskog, 1982). The PLS method
has been introduced in the chemical literature as an
algorithm and it is only recently that its numerical and
statistical properties have become more apparent (Stone,
1974). PLSR is a technique for modeling a linear
relationship between a set of output variables (response)
n L
i i 1
{ } R y
=
with L-dimensional responses and a set of
input variables (regressors)
n p
i i 1
{x } R
=

with p number of
variables (Rosipal and Trejo, 2002). The data matrices
X and y in this analysis are assumed to be centered as a
first step to perform PLSR.
In this study, we only use one dimensional response
which is L equals to one. PLS is a method for modeling
relations between sets of observed variables by means of
latent variables which are linear combinations of the
original regressors while maintaining most information
in the input variables. PLS is useful when the number
of explanatory variables exceeds the number of
observation and high level of multicollinearity among
those variables is assumed. The weights used to
determine the linear combinations of the original
regressors are proportional to the covariance among
input and output variables (Helland, 1988).


AJAS
2.3. Partial Least Squares Regression Using
SIMPLS Algorithm
SIMPLS algorithm was used to compute the
regression coefficient in order to find the model for
predicting water level in Dungun River of Terengganu.
SIMPLS algorithms work very well, resistant to be more
appropriate, fast, easy to implement and simple to tune
(Bennett and Embrechts, 2003). In PLSR approach, we
need to obtain the PLSR estimator, say
PLSR
B
T
PLSR1 PLSR2 PLSRp

[B , B , , B ] , = and it starts with computing
the cross-product of (Jong, 1993; Ibrahim and Wibowo,
2012) as shown in Equation (5):

T
S X y = (5)

Then, the computing of the iteration is followed
starting from 1 until A latent variables where A is
determined in advanced and 1A p. The algorithm of
SIMPLS is given as follows:
For a =1to A:

If a = 1, then do the singular value decomposition
(svd) of S: [u, u, v] = svd (S)
Otherwise, if a > 1, we compute the svd of:

T 1 T
[u,s, v] svd (S P(P P) P S)
=

Get weights for r which is the first singular vector: r
= u (:, 1)
Compute the scores: t = Xr

Compute the loadings: p = X
T
t/(t
T
t)
The vector r, t and p are stored into R, T and P
respectively

The last step is computing a regression coefficient
can be shown in Equation (6):

T 1 T
PLSR
B R(T T) T y
= (6)

Then, the estimate of PLSR is given in Equation (7):

PLSR
y XB = (7)

The model for PLSR can be represented by
Equation (8):

1 1 p p PLSR1 PLSRp

g(x) y B (x x ) B (x x ) = + + + (8)
where, y is the mean of response y
i
and
p
x is the mean of
observation data of x
p
.
2.4. Evaluating the Quality of the Prediction
The quality of the prediction is evaluated using A
latent variables,
i
y and y
i
(Helland, 1988; Ibrahim and
Wibowo, 2012). CV technique is used to estimate the
prediction capacity and the data are separated between
the training data set to build the model and testing data set
to test the model. The CV is applied in three cases which
are in performance estimation, model selection and tuning
learning model parameters. In this study, CV is used in
predictors selection and model selection for predicting
water level of Dungun River. The CV is a statistical method
to evaluate the algorithms by dividing the data into two
segments which are for training and validation and the
basic form of cross-validation is K-fold CV. The idea
for CV was originated in the 1930s (Larson, 1931;
Refaeilzadeh et al., 2008; Ibrahim and Wibowo,
2012). In 1970s, CV was employed as means for
choosing proper model parameters, as opposed to
using cross-validation purely for estimating model
performance (Geisser, 1975; Sjgstrgm et al., 1983;
Ibrahim and Wibowo, 2012).
Stratified 10-fold CV was recommended as the best
model selection method since it tends to provide less
biased estimation of the accuracy compared to regular
cross-validation, leave-one-out CV and bootstrap methods
(Refaeilzadeh et al., 2008; Ibrahim and Wibowo, 2012).
For this analysis, we used 10-fold CV because it can give
accurate performance estimation and it suitable for small
samples of performance estimation. We were using this
type of CV to choose an appropriate model between
normalized original data and cleansing data by
comparing the value of Mean Squared Error of Cross-
Validation (MSECV) based on OLR and PLSR. The data
are divided into K segments of roughly equal size and the
inner sum of MSECV is taken over the observations in the
kth segment (Davison and Hinkley, 1997; Mevik and
Cederkvist, 2004; Ibrahim and Wibowo, 2012). For each
of K experiments, the K-fold CV uses K-1 folds for
training and the remaining one for testing. There is an
advantage of using K-fold CV which is all the examples
in the dataset are eventually used for both training and
testing. For this type of CV, we used the function in
Matlab software called crossval to obtain the value of
MSECV which is a scalar containing a 10-fold CV
estimate of mean-squared error. We will select a better
model according to lowest value of MSECV and it is a
measure of how well the model fits the data.


AJAS
2.5. Data
As predictors in predicting water level of Galas
River, months (x
1
), average monthly rainfall (x
2
),
temperature (x
3
), relative humidity (x
4
) and surface
wind (x
5
) were identified and related to the occurrence
of flood phenomenon in Kuala Krai, Kelantan.
Observed predictors and response for the period 2001-
2011 were extracted from the Water Resources
Management and Hydrology Division in Kuala
Lumpur and Malaysian Meteorological Department in
Selangor. Variable selection is performed to select the
suitable predictors in predicting the water level based
on the MSECV of OLR and PLSR. It is noted that the
data consist of missing values for rainfall and water
level and we performed cleaning data to replace these
missing values. The data are separated into two sub
data which are 120 data for developing models and
variables selection using 10-fold CV and 12 data for
validating the models. The data that were used in this
analysis are shown in Table 1.
2.6. Original Data
The data set is cover from January until December
for 11 years and yet it has shown a total of 132 data.
Table 1-4 describe the predictors and response used over
training period in predicting water level of Galas River.
The first column, second column, third column, fourth
column, fifth column and sixth column represent
months, rainfall, temperature, relative humidity, surface
wind and water level data, respectively. They show the
raw data and 47th month is in November 2004 and the
NA values means that there are missing values of rainfall
in November and December 2004.
2.7. Pre-Processing Data
Data preprocessing is the process that was
performed to the original data in order to prepare it for
next processing procedure. Thus, it will transform the
data into the format that more effective according to
our purpose of analysis. Data preprocessing is
important since the real world data normally are noisy
which are containing errors and outliers. There are
five tasks in performing data preprocessing which are
data cleaning, data integration, data transformation,
data reduction and data discretization. For this
analysis, we performed two types of data cleaning
which are using mean of the corresponding months
throughout 11 years and OLR to replace the missing
values of rainfall and water level.
Table 1. Details of the data
Station Period Data
Kuala Krai 2001-2011 Monthly 24 h Mean Temperature
Monthly 24 h
Mean relative humidity
Monthly mean surface wind
Dabong 2001-2011 Monthly mean
rainfall and water level

Table 2. The snapshot of original data of Galas River
Relative Surface Water
Month Rainfall Temperature humidity wind level
37 9.840 25.7 86.5 0.5 27.58
38 0.810 26.4 81.5 0.7 27.00
39 6.510 27.4 82.8 0.7 26.84
40 1.930 27.8 82.1 0.5 26.29
41 4.120 28.0 83.4 0.5 26.48
42 12.450 27.6 83.1 0.6 26.17
43 1.875 26.7 83.5 0.7 26.08
44 7.390 27.2 83.6 0.7 26.12
45 13.180 26.5 85.6 0.6 28.26
46 9.600 25.9 88.5 0.3 28.49
47 NA 26.0 88.9 0.2 27.76
48 NA 25.0 89.3 0.2 29.12

Table 3. The snapshot of pre-processing data of galas river
using type I pre-processing data
37 9.840 25.7 86.5 0.5 27.58
38 0.810 26.4 81.5 0.7 27.00
39 6.510 27.4 82.8 0.7 26.84
40 1.930 27.8 82.1 0.5 26.29
41 4.120 28.0 83.4 0.5 26.48
42 12.450 27.6 83.1 0.6 26.17
43 1.875 26.7 83.5 0.7 26.08
44 7.390 27.2 83.6 0.7 26.12
45 13.180 26.5 85.6 0.6 28.26
46 9.600 25.9 88.5 0.3 28.49
47 12.750 26.0 88.9 0.2 27.76
48 163.280 25.0 89.3 0.2 29.12

Table 4. The snapshot of pre-processing data of Galas River
using type ii pre-processing data
37 9.840 25.7 86.5 0.5 27.58
38 0.810 26.4 81.5 0.7 27.00
39 6.510 27.4 82.8 0.7 26.84
40 1.930 27.8 82.1 0.5 26.29
41 4.120 28.0 83.4 0.5 26.48
42 12.450 27.6 83.1 0.6 26.17
43 1.875 26.7 83.5 0.7 26.08
44 7.390 27.2 83.6 0.7 26.12
45 13.180 26.5 85.6 0.6 28.26
46 9.600 25.9 88.5 0.3 28.49
47 6.570 26.0 88.9 0.2 27.76
48 6.570 25.0 89.3 0.2 29.12


AJAS
Table 5. Msecv for variable selection of Galas River
MSECV
----------------------------------------------------------------------------------------------------------
PLSR
Pre-processing Data Predictors OLR ncomp = 1 ncomp = 2 ncomp = 3 ncomp = 4 ncomp = 5
x
1
, x
2
, x
3
, x
4
,x
5
0.7281 1.2239 0.7989 0.6420 0.5688 0.5623
x
1
, x
2
, x
3
, x
4
0.7218 0.6359 0.6556 0.6393 0.5934
Type I x
1
, x
2
, x
4
, x
5
0.7095 0.7038 0.7122 0.7121 0.6777
x
1
, x
2
, x
3
, x
5
2.2966 0.6122 0.5975 0.6315 0.6185
x
1
, x
2
, x
3
2.4732 0.6208 0.6138 0.5828
x
1
, x
2
, x
4
0.7189 0.7261 0.7366 0.6972
x
1
, x
2
, x
5
48.6901 0.8776 0.9144 0.9158
x
1
, x
2
1.2298 0.9870 0.9596
x
1
, x
2
, x
3
, x
4
,x
5
0.7253 0.5769 0.5883 0.5809 0.6031 0.5765
x
1
, x
2
, x
3
, x
4
0.7354 0.5789 0.5804 0.5988 0.5995
Type II x
1
, x
2
, x
4
, x
5
0.7209 0.6805 0.6803 0.6933 0.6916
x
1
, x
2
, x
3
, x
5
2.3629 0.6474 0.6455 0.6496 0.6386
x
1
, x
2
, x
3
2.4326 0.6493 0.6223 0.6308
x
1
, x
2
, x
4
0.7343 0.7107 0.6851 0.726
x
1
, x
2
, x
5
50.0777 0.9258 0.8853 0.8849
x
1
, x
2
123.2602 0.9925 0.9953

2.8. Pre-processing Data Using Mean of the
Corresponding Months
For this subsection, we used mean of the
corresponding months which is represented by type I
pre-processing data to replace these missing values. For
example, NA value of rainfall in November and
December 2004 for Galas River are replaced by the
means of the corresponding months throughout 11
years. Table 4 presents the snapshot of the pre-
processing data using mean of the corresponding
months for Galas River.
2.9. Pre-processing Data Using Ordinary Linear
Regression
The second type of cleaning data that we used is
OLR and we represent it as type II pre-processing data.
We performed OLR to replace the missing values of the
dataset in Galas River. Table 3 shows the snapshot of
the pre-processing data using OLR for Galas River.
The model to replace the missing value of the water
level for Galas River is given in Equation (9):

1 1 OLR
WL1
f (x ) 26.5767 0.0135x = + (9)

The model to replace the missing values of the
rainfall for Galas River is represented by Equation (10):

1 1 OLR
RF1
f (x ) 6.4736 0.0021x = + (10)

2.10. Selection of Predictors
The selection of appropriate predictors is one of the
most important steps in predicting the water level of
Galas River. The predictors are chosen based on the
smallest value of MSECV and the result is compared
between two types ofpre-processing data which are type
I pre-processing data and type II pre-processing data. It
can be seen from Table 5 that five predictor variables
namely months (x
1
), average monthly rainfall (x
2
),
temperature (x
3
), relative humidity (x
4
) and surface wind
(x
5
) with type I pre-processing data have their lowest
value of MSECV when ncomp is equals to five. Hence,
these variables are used in the water level predictions.
3. RESULTS
3.1. Models Development
The models for water level predictions of Galas
River were developed using OLR and PLSR. The results
were compared between these two approaches and
between two types of pre-processing data.
3.2. Ordinary Linear Regression
LR is performed in this experiment to build the model
for water level in Galas River. This subsection presents the
results of the experiment which are the prediction models
for water level over training period based on two types of
pre-processing data. The prediction model for water level


AJAS
using type I pre-processing data of Galas River is given by
Equation (11):

1 2 3 4 5
1 2
3
4 5
OLRGalasI
f (x , x , x , x , x ) 31.6220
0.0038x 0.0323x
0.5443x
0.1165x 0.4555x
=
+ +
+
(11)

The prediction model for water level using type II
pre-processing data of Galas River is given in
Equation (12).

1 2 3 4 5 1
2
3
4 5
OLRGalasII
f (x , x , x , x , x ) 30.0678 0.0036x
0.0309x
0.5085x
0.1233x 0.3873x
= +
+
+
(12)

3.3. Partial Least Squares Regression:
PLSR is another method that we use in this
experiment in order to get the prediction model and the
results based on these two methods are being compared
between original data and cleansing data. Validation
method is used for choosing number of components of
PLS and the model with the lowest MSECV is
considered to be the optimal one.
The prediction model for water level of Galas River
using type I pre-processing data is represented by
Equation (13):

PLSRGalasI 1 2 3 4 5
1 2
3
4 5
g (x , x , x , x , x ) 33.1985
0.0037x 0.0340x
0.5712x
0.1064x 0.4552x
= +
+
+
(13)

The prediction model for water level using type II
pre-processing data of Galas River is given in
Equation (14):
1 2 3 4 5
1 2
3
4 5
PLSRGalasII
g (x , x , x , x , x ) 31.5444
0.0036x 0.0324x
0.5340x
0.1139x 0.3889x
= +
+
+
(14)

3.4. Model Selection
In this study, we will restrict ourselves to the common
variants of CV called K-fold CV, where the calibration
objects are divided in k segments and for this experiment
we use k = 10 (Breiman, 1984; Wiklund et al., 2007;
Ibrahim and Wibowo, 2012). The selected number of
components using k-fold CV correctly find this range,
the actual value of the number of components is
immaterial as long as the prediction error is close to its
minimum (Wiklund et al., 2007; Ibrahim and Wibowo,
2012). We used 10-fold CV to obtain the appropriate
model for predicting water level at Galas River of
Kuala Krai using two types of pre-processing data. The
data were analyzed using OLR and PLSR and the
results are compared between these two types of pre-
processing data to obtain a better model according to
lowest value of MSECV.
4. DISCUSSION
Table 6 illustrates the comparison of MSECV for
Water Level in Galas River using 10-fold CV of OLR
and PLSR. From this result, it shows that PLSR with
type I pre-processing data of ncomp equals to 5 has the
smallest MSECV. Therefore, this PLSR is considered as
the best model. Figure 2 shows the comparison between
actual and prediction monthly water level for Galas
River with test data in 2011 using type I pre-processing
data and Fig. 3 presents the comparison between
predicted and actual water level in Galas River with test
data using type II pre-processing data. From these graph,
it is clear that the use of type I pre-processing data
achieves closer agreement between actual and predicted
water level rather than using type II pre-processing data.

Table 6. Msecv for variables selection of Galas River
MSECV
--------------------------------------------------------------------------------------------------------------------------------
PLSR
------------------------------------------------------------------------------------------------------------
Pre-processing data OLR ncomp = 1 ncomp = 2 ncomp = 3 ncomp = 4 ncomp = 5
Type I 0.7281 1.2239 0.7989 0.642 0.5688 0.5623
Type II 0.7253 0.5769 0.5883 0.5809 0.6031 0.5765


AJAS

Fig. 2. A comparison between actual and prediction monthly water level for Galas River with test data of 2011 using type I pre-
processing data

Fig. 3. A comparison between actual and prediction monthly water level for Galas River with test data of 2011 using type II pre-
processing data

5. CONCLUSION
In Kuala Krai district, rising water levels of the
river become critical issues since it can induce flood and
destroy a lot of things. We had compared between two
types of pre-processing data which are type I and type II
pre-processing data using OLR and PLSR approaches
for variables selection and model selection. The
experiment had shown that PLSR is a suitable method
in variables selection and model development since it
give higher accuracy than using OLR. Our further
research will focus on the use of nonlinear method and
compare them to PLSR model.
6. ACKNOWLEDGEMENT
This project is funded by the Short Term Research
Grant-Foreign Academic Visitor Fund (vot number:
4D051). The authors would like to thank the Research
Management Centre for supporting this research and
Drainage and Irrigation Department of Malaysia for
general assistant. The first author would like to thank for


AJAS
Zamalah scholarship for supporting her master by
research program.
7. REFERENCES
Alexander, D.E., 1993. Natural Disasters. 1st Edn.,
Springer Science and Business, New York, ISBN-
10: 0412047411, pp: 632.
Bennett, K.P. and M.J. Embrechts, 2003. An optimization
Perspective on Kernel Partial Least Squares
Regression. In: Advances in Learning Theory:
Methods, Models and Applications, Suykens, J.A.K.
(Ed.), IOS Press, Amsterdam, ISBN-10:
1586033417, pp: 227-250.
Bergant, K. and L. Kajfez-Bogataj, 2005. N-PLS
regression as empirical downscaling tool in climate
change studies. Theoritical Applied Climatol., 81:
11-23. DOI: 10.1007/s00704-004-0083-2
Breiman, L., 1984. Classification and Regression Trees.
1st Edn., Wadsworth International Group, Belmont,
Calif., ISBN-10: 0534980538, pp: 358.
Bustami, R., N. Bessaih, C. Bong and S. Suhaili, 2007.
Artificial neural network for precipitation and water
level predictions of Bedup River. IAENG Int. J.
Comput. Sci.
Carrillo-Rivera, J.J. and A. Cardona, 2012. Groundwater
flow systems and their response to climate change: A
need for a water-system view approach. Am. J.
Environ. Sci., 8: 220-235. DOI:
10.3844/ajessp.2012.220.235
Davison, A.C. and D.V. Hinkley, 1997. Bootstrap
Methods and their Application. 2nd Edn., Cambridge
University Press, Cambridge, ISBN-10:
0521574714, pp: 582.
Geisser, S., 1975. The predictive sample reuse method
with applications. J. Am. Stat. Assoc., 70: 320-328.
DOI: 10.1080/01621459.1975.10479865
Goyal, M.K. and C.S.P. Ojha, 2010. Application of PLS-
regression as downscaling tool for pichola lake basin
in India. Int. J. Geosci., 1: 51-57. DOI:
10.4236/ijg.2010.12007
Helland, I.S., 1988. On the structure of partial least
squares regression. Commun. Stat. Elem. Simul.
Comput., 17: 581-607. DOI:
10.1080/03610918808812681
Ibrahim, N. and A. Wibowo, 2012. Predictions of water
level in Dungun River Terengganu using partial least
squares regression. Int. J. Basic Applied Sci., 12: 1-7.
Jong, S.D., 1993. SIMPLS: An alternative approach to
partial least squares regression. Chemometr. Intell.
Laboratory Syst., 18: 251-263. DOI: 10.1016/0169-
7439(93)85002-X
Jonkman, S.N. and I. Kelman, 2005. An analysis of the
causes and circumstances of flood disaster deaths.
Disasters, 29: 75-97. DOI: 10.1111/j.0361-
3666.2005.00275.x
Joreskog, K.G., 1982. Systems under Indirect
Observation. 1st Edn., North-Holland, Amsterdam,
ISBN-10: 044486301X, pp: 636.
Larson, S.C., 1931. The shrinkage of the coefficient of
multiple correlation. J. Educ. Phychol., 22: 45-55.
DOI: 10.1037/h0072400
Mevik, B.H. and H.R. Cederkvist, 2004. Mean Squared
Error of Prediction (MSEP) estimates for Principal
Component Regression (PCR) and Partial Least
Squares Regression (PLSR). J. Chemometr., 18:
422-429. DOI: 10.1002/cem.887
Noji, E.K., 1997. The Public Health Consequences of
Disasters. 1st Edn., Oxford University Press, USA.,
ISBN-10: 0195095707, pp: 468.
Refaeilzadeh, P., L. Tang and H. Liu, 2008. Cross-
Validation. Arizona State University.
Rohasliney, H., 2010. Status of river fisheries in
Kelantan, Peninsular Malaysia, Malaysia. World
Acad. Sci., Eng. Technol., 65: 829-834.
Rosipal, R. and L.J. Trejo, 2002. Kernel partial least
squares regression in reproducing kernel hilbert
space. J. Mach. Learn. Res., 2: 97-123.
Shu, L., G. Dong, L. Liu, Y. Tao and M. Wang, 2008.
Water level variation and prediction of the Pingshan
Sinkhole in Guizhou, Southwestern China. Am. Soc.
Civil Eng. DOI: 10.1061/41003(327)40
Sjgstrgm, M., S. Wold, W. Lindberg, J.A. Persson and H.
Martens, 1983. A multivariate calibration problem in
analytical chemistry solved by partial least-squares
models in latent variables. Anal. Chim. Acta, 150:
61-70. DOI: 10.1016/S0003-2670(00)85460-4
Stone, M., 1974. Cross-validatory choice and assessment of
statistical predictions. J. Royal Stat. Soc., 36: 111-147.
Suparta, W., J. Adnan and M.A.M. Ali, 2012. Monitoring
of GPS precipitable water vapor during the severe
flood in Kelantan. Am. J. Applied Sci., 9: 825-831.
DOI: 10. 384/ajassp.2012.825.831
Wiklund, S., D. Nilsson, L. Eriksson, M. Sjostrom and S.
Wold et al., 2007. A randomization test for PLS
component selection. J. Chemom., 21: 427-439.
DOI: 10.1002/cem.1086
Zou, Y., W. Zhou and M. Zhong, 2010. A model on the
relation between the rainfall in Poyang Lake Basin
and its Water Level. Proceedings of the 4th
International Conference on Bioinformatics and
Biomedical Engineering, Jun. 18-20, IEEE Xplore
Press, Chengdu, pp: 1-4. DOI:
10.1109/ICBBE.2010.5516861
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.
Hussein Akberali
S1314000563
Page 1 of 4
1 PARTIAL LEAST SQUARES REGRESSION BASED
VARIABLES, SELECTION FOR WATER LEVEL
PREDICTIONS
AMERICAN JOURNAL OF APPLIED SCIENCES, 10 (4) PG. 322-330, 2013
The paper that I have chosen to analyze the Least Squares Regression Model is an analysis to predict the
flooding due to the Monsoon season in Galas, Malaysia by looking at the selected predictors and the
water level of the Galas River. There are five (5) independent variables that were selected that would
lead to flooding, these were; 1) Months from January to December from 2001 to 2011, 2) Monthly mean
of rainfall, 3) Monthly mean of temperature, 4) Monthly Mean of relative humidity, 5) Monthly mean of
surface wind.
The methods they determined that would work best for this analysis was the linear regression model
using the equation Y = 0 + 1(X) + (error). The methods they attempted are known as Ordinary
Linear Regression and Partial Least of Squares Regression.
It was determined that the model with the lowest Mean Squared Error of Cross Validation (MSECV),
which is where the data is divided into K segments, each segment being roughly same size and the inner
sum of MSECV is taken over the observation of the Kth segment. The lower the value of MSECV, the
better the model, therefore, the objective of this study is to find which model would give the lowest
MSECV and be the best predictor of the flooding in Galas, Malaysia
When using the Type 1 Pre-processing data under the Ordinary Linear Regression, the equation was
calculated as fOLRGalasI f (x1, x2, x3, x4, x5) = 31.6220 +0.0038x1 +0.0323x2 - 0.5443x3 + 0.1165x4 -
0.4555x5.
When using the Type 1I Pre-processing data under the Ordinary Linear Regression, the equation was
calculated as fOLRGalasII f (x1, x2, x3, x4, x5) = 30.0678 +0.0036x1 +0.0309x2 - 0.5085x3 + 0.1233x4 -
0.3873x5.
When using the Type 1 Pre-processing data under the Partial Least Squares Regression, the equation
was calculated as gPLSRGalasI f (x1, x2, x3, x4, x5) = 33.1985 +0.0037x1 +0.0340x2 - 0.5712x3 + 0.1064x4 -
0.4552x5.
When using the Type 1I Pre-processing data under the Ordinary Linear Regression, the equation was
calculated as gPLSRGalasII f (x1, x2, x3, x4, x5) = 31.5444 +0.0036x1 +0.0324x2 - 0.5340x3 + 0.1139x4 -
0.3889x5.
Based on the results obtained versus actuals ( (error)), and the lowest MSECV results, it was
determined that the Partial Least Squares Regression is a suitable method and gives a higher accuracy
than the Ordinary Linear Regression Model. Based on this, the Partial Least Squares Regression can be
used to warn people of any impending floods in the area as the water levels can be confidently
predicted.
ijcrb.webs.com
INTERDISCIPLINARY JOURNAL OF CONTEMPORARY RESEARCH IN BUSINESS
COPY RIGHT 2013 Institute of Interdisciplinary Business Research

955

MAY 2013
VOL 5, NO 1

Application of Linear Programming to Improve the Transport of Wheat

Majid Yahiavi
1
, Soheil Talebian
2
, Mohamad Poorkhaefdoost
3
Abas Abasi
4
Samira Dejkam
Foomani
5

1,2,3,4,5-M A Student of Business Management, Science and Research Branch, Islamic Azad University,
Rasht, Iran

Abstract
Solve the transportation problems is one of the first and most important applications of linear programming methods. The first
application of this approach is seen during World War II. This paper using the methods of linear programming, Provide a
distribution method that benefits both are supplier side, Transportation Services (goods transport companies and drivers) and the
applicant Transportation Services (The Wheat State) will be considered. In continue, we compared this method with the distribution
method which used in 1379 (2000) for imports of wheat and calculate the savings that could be income in this way. And secondl y,
in provincial capitals (Which are considered as transportation destinations) the required grain storage facilities in each province are
there in a sufficiently way.

Keywords: supply, demand, linear programming

Introduction
In terms of the amount of goods moved in country wheat is in the second place. In other words, wheat allocated between 6 to 7
percent of the total transferred products to itself. On the other hand, wheat is one of the essential commodities that the supply and
distribution of it is in government monopoly. According to studies, the transportation of wheat in the country is faced with
problems, and this is the cause of reducing in productivity of wheat discussion. For example, imported wheat in country is
evacuated from mega ports, and some considerations in this subject arent seen, such as the distance between the consumption
centers and economic issues of transportation (2). Since wheat is one of goods that the government pays subsidies for it, it is the
cause of national capital loss. And in a year, because of the necessity of supplying wheat to all areas of the country and
coordinating the quality of wheat, they proceed to fill and empty the silo or silos is a province for several times, which led to the
creation of additional transportation. But being economical for transportation organizations is an issue that should not be hidden of
sight of policy makers in transportation sector; this means that planning should be done so that the equilibrium between the two
sides, the applicant of transportation services and transport services supplier is established. So the problem can be stated as follows:
What is the grain of wheat from each distribution center (mega-port state and internal centers) to each of the consumption centers
(centers for domestic consumption) must be carried to on behalf of the applicant of transportation services meet all demands and
minimize the total cost of transportation, and also for the transportation supplier, In consideration of the issue costs, best and most
profitable way to obtain for transportation suppliers(1).

Problem-solving approach
To evaluate and compare the distribution wheat could be studied in three cases; so the researchers compared them with what
happened, to reach a satisfactory conclusion.
First case: the case where the objective is to minimize the cost of transportation service for the applicant of transportation services.
Second case: the case where the objective is to maximize profit for the transportation services supplier (5).

Third case: is a equilibrium, so in this case the simultaneously target is minimizing costs for the applicant of transportation
services and maximizing profits for the transportation services supplier.
But at first the distribution of imported wheat in 1379 (2000) is reviewed and analyzed according to the statistics of countrys
transportation and terminals organization, it should be noted that these statistics obtained from the bill of lading of countrys
transportation and terminals organization. On this basis in 2000, according to the statistics of wheat purchased from foreign
countries is loading and unloading in Bandar Imam, Bandar Abbas, Chabahar port, Bandar Anzali, Noushahr port and Sarakhs. The
distribution of wheat is about 5/5 million tons of wheat between the 28 provinces of the country. In the first case, you should see
what is the distribution of wheat? to have a low cost of transportation for the applicant's services, For this purpose, a
ijcrb.webs.com

956

MAY 2013
VOL 5, NO 1

mathematical expression is composed that the variables indicative of wheat shipped from the port to the centers of the provinces
and the coefficients of the variables is representing each ton of wheats transport rent from the source to the destination which this
expression should be minimized by taking the following constraints. These limitations include the loading and unloading capacity
of each entry, These limitations include the loading and unloading capacity of each mega-port, which is here it considered as equal
to the sum of wheat shipped from each mega-port to each provincial capitals in 1379 (2000). In other words, the first constraint or
limitation is that the total distributed wheat from each mega-port must not be greater than the loading and unloading capacity of
each input source. The second limitation relates to the needs of each province's to imported wheat. This quantity is considered as
equivalent to the amount of imported wheat into the each province in 1379 (2000). Thus, with the main terms and the limitations
which is mentioned above and using the LINDO software and after 40 times repetition, the result is a distribution method (6). If the
imported wheat in 1379 (2000) was distributed in accordance with this approach, then the minimum cost will paid by applicant of
transportation services. In the second case, the objective is to maximize profit of the transportation supplier. Thus, the distribution
of wheat from mentioned mega-port to the provinces, how should be to make the maximum profits for suppliers of transport
services. In this situation, variable coefficients is defined as the subtracting of the rent of one tone of wheat transportation cost from
a specific source from the cost of carrying one ton of wheat from the same source to the same destination. The limits in this case
and the first case are exactly the same. With completing the problem and by using of LINDO software and after 44 times repetition,
the result is another distributed approach. In other words, if the imported wheat in 1379 (2000), was distributed according to this
approach, transportation suppliers would be reach to maximize profits (4).
In the third case, the final purpose is creating balance between demands of sides, the applicant of transportation services and the
transportation services supplier; this means that distribution of imported wheat must be done in a way that simultaneously and
relatively obtain minimal cost for applicant of transportation services and maximum benefit to the transportation suppliers. In this
part the dual (DUALL) is used. In the dual form the maximum of each function is equal to its negative minimum and the minimum
of each function is equal to its negative maximum. So the expression is set as follows:

(The supply function demand function) MIN

We solved the above expression with the same constraints which mentioned previously by using LINDO software and after 37
times of repetition and some values were obtained. In other words, if the distribution of imported wheat in 1379 (2000) was made
according to this approach, the supplier's profit reached to maximize, and the costs were minimized for the applicant of
transportation services (8).

History of research
Mohammad Bagher Fjrzad - Yazd University - In a study entitled Analysis and optimization of transportation systems by using
linear programming, network models has concluded that systematic linear programming method is an Organized and logical
solution to analyze transportation projects (3). Thus, a good control acquired for the analyst, and the important personnel and those
who have somehow involved in project. Since the time, cost, satisfaction and human resource is important to transport. Therefore,
with the transport model, we have seen reduction of costs, satisfaction of vehicle drivers and passengers. Following this plan results
were as follows:
1 - Reduce Costs 64% of the week (using the appropriate vehicle, instead of prioritizing the use of available rental vehicles, using
hourly rental vehicles instead of daily rental vehicles).
2 - Increased passenger satisfaction (regular and timely movement of vehicles, passengers get to their destination on time, enough
chairs to sit on).
3 - Increase driver satisfaction and unit (there is a regular schedule for all vehicles, there is no overcrowding of passengers in
vehicles, reducing time and coordination meetings between officials and drivers, reducing down time of vehicle, personal time off
program at time of low demand and overtime in high demand).

ijcrb.webs.com

957

MAY 2013
VOL 5, NO 1

Concepts and Terminology
Linear programming is a method which by using it a desirable planning will determined for economic activity related to the limited
resources in a given period of time. In general, determining the desirable level is maximizing and minimizing of a function
(function of generate, profits, cost) due to the limitations of resources (Work force, capital). Linear programming is widely used in
military, industry, government, urban planning, management, economics, agriculture and other areas. It has many advantages,
including:
1 - Modeling the most complex and important issues in the management of linear programming.
2 - Ability to solve problems in a reasonable time. Resource constraints, is the main motivation to optimize the production process
and consumption. For this reason, various optimization methods are used in the economy. Linear programming as a static
optimization technique is relatively widely used in economies (9).

Applications:
Linear programming has numerous applications in military, government, industry and Urban Planning and also often used as part
of the computational planes, nonlinear programming problem solving, discrete programming, combinatorial problems, optimal
control problems and contingency planning. According to several reasons, nonlinear programming is an important area in
optimization, such as: Many scientific problems in operations research can be expressed as a linear programming problem and also
a number of other algorithms of optimization problems solve by linear programming, and works as a sub-problem. Historically, the
ideas of linear programming are inspiration to many of the basic concepts of optimization theory, such as: duality, decomposition,
the importance of convexity (11).

Analysis
First case: minimizing the cost of transport for applicant: Variable coefficients in main expression for this case are multiplied by
the rental ton - kilometers carrying wheat from the source to the destination in the distance between the source and destination.
Because in this case the problem is minimizing the cost of shipped imported wheat to consumption centers, the model results
should also help minimize the total cost of transportation. Based on the responses obtained from solving the problem, the sum of
minimum cost is 8/575 billions of Rails that caused from the solving the problem with 40 times of repetition. These minimal costs
are established, based on the optimal values of the wheats transportation which have been obtained in model results.
Second case: maximizing the profit of transportation supplier: In this case, profit maximization for transportation supplier (Carrier)
is our goal. So it have to maximize the total benefits from cargo trips, according to obtained responses, the total profits of wheat
carrying based on the received maximum benefits can be about 9/15 billions of Rails, which this result is obtained with 44 ti mes of
repetition. It should be mentioned that in objective function the supply side also, is variable of the same amount transportation of
goods, but the coefficient of this variable to maximize profits for the benefit of any origin to any destination. Thus, the benefits of
carrying one ton of wheat from any entry origin of wheat to any province in order to calculate this amount we required to know
income and expenses for each trip, which with subtracting expenses from income profits are achieved. Accordingly, the variabl e
coefficient in this function (supply side) is a factor of two partial, like (Cij-Pij). Pij is the rent of each pair of source - destination
wheat transport and Cij is costs of each per of a ton of wheat transportation from the same source to the same destination.
Conclusion it can be stated as follows: Distribution of wheat which was in 1379(2000) for the owner of the goods (the government)
had 581 331 951 500 rails of costs, but if this distribution had to be done based on the optimal pattern of demand, this cost would
reduce to 575818207100 RLS, as noted above, the transportation system is currently being detrimental. Thus, for comparison, the
loss rate in the transport service provider is both realistic and optimal contrast. According to the distribution of wheat in 1379
(2000), this distribution for the transportation supplier (vehicles, Carrier) has been 278776887700 RLS of the loss, However, if the
distribution had to be done according to provide optimal model, the value of these losses decreased to 243472250000 RLS.
Third case: equilibrium: As you can see the benefits of applicant of the transportation (owners of goods - here is the government)
in order to reduce the rent off transformation, if the supplier of transport services (carriers of wheat) to get out of the profit losses
case and earn profits tend to increase rent of transportation. Strategy should be considered here, to create balance between the two
sides of this interact. In other words, a model should be presented for the distribution which in the benefits of this model is split
relatively evenly between the two sides. This model is a balance between sides of supply and demand of transportation which in a
dual form (DUALL) function is used.

Conclusions
The result can be expressed in another way. Thus, if the distribution of imported wheat from ports in to the provinces, the total cost
which community to pay as a unit would be 848830240000 RLS. While currently value of the cost based on actions in 1379 (2000)
is 895 814 363 100 RLS. And that means 10984123100 RLS surcharge, in other words one billion and one hundred million
Tomans to the national capital loss which can be prevented by an optimal planning. Finally it should be added that the answers to
questions such as: Current capacity of the Mega-port whit the needs of Society based on efficient use of resources is consistent or
ijcrb.webs.com

958

MAY 2013
VOL 5, NO 1

not? In case of Increase capacity, how much of the needs of each province should be provided from each mega-port? And excess
capacity and a lack of capacity is in what port and how much? Increase the number of mega-ports of wheat (due to the limitations
of any new source) has what effect in the answer of problem? All require a new solution to the problem, which can continue to pay
for them (10).

ijcrb.webs.com

959

MAY 2013
VOL 5, NO 1

References

1.Vineston,V,(2001), Linear programming, Publication termeh .Tehran.
2.Azar,F &Adel,B.(2004), Operations research concepts and applications of linear programming, Publishing of modern science,
third edition.
3.Akbari,K &Sharif,N.(1999). Agricultural Economics, Allameh Tabatabai University Press, September.
4.Soltani.GH&Kahkha,A.(2008).Application of mathematical programming in agricultural education and agricultural extension
publications, research organizations.
5.Abdshahi, A &Torkamani,J.(1998). Estimated demand function units Fattening calf ingredients in Fars Province, Journal of
Agricultural Economics and Development, 9-27.
6.Jafari Samimi, A &Tehranchian,A.(2005). Mathematical economics, BABOLSAR, Mazandaran University Press
7- Alphonce, B.A.(1997).Application of the analytic hierarchy process in agriculture in developing countries, Agricultural Systems,
53:97-112.
8- Berkeley,H.(2006). An introduction to economics, concepts for students of agriculture and the Rural sector, 238-267.
9- Burton, R.O. Gidly, J. S. Reda- Wilson.(1987). Nearly optimal linear Programming solution. Some application, American
Journal of Agricultural Economics, 69(4): 813-818.
10- Havlicek, J&. Kucera,P.(1997). Decision making in risk environment based on fuzzy linear programming, ScientialAgriculture
Bohemica,28(3): 227-244.
11- Chung-HsingYeh.(2000). Fuzzy multi criteria analysis for performance evaluation of bus companies: European journal of
Operational Research 126

Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.
Hussein Akberali
S1314000563
Page 2 of 4
2 LINEAR PROGRAMMING TO IMPROVE THE TRANSPORT
OF WHEAT
INTERDISCIPLINARY JOURNAL OF CONTEMPORARY RESEARCH IN BUSINESS MAY 2013, VOL. 5, NO. 1
This paper analyzes the transportation problem of wheat in Iran using the Linear Programming Methods.
I have selected this paper as it gives a good background as to the challenges of transporting an integral
and vital product, which constitutes of roughly between 6-7 % of the total goods transported in Iran.
As the paper explains, since Wheat is subsidized, it equates to a national capital loss, which is why the
Government has a responsibility to ensure that the transportation cost is minimized and as well
maximize the profit for shipping companies (capital gain).
The variables for this challenge are defined as the Ports where the Wheat is being shipped from
(Supply), and the Center of the Provinces where the items are being delivered to (Demand). The
variables were subject to the following constraints; loading and unloading capacity of each mega-port,
the second constraint being the demand should be fulfilled in each Provincial Center. This study was
conducted in the year 2000, and with the resources available at the time, they decided to use the
LINDO software for calculating based on the above variables and constraints. After several trial and
error scenarios, they were able to determine the ideal distribution of Wheat to be 1379 Tons to satisfy
the demand.
Based on the calculations and the determined amount of Tons of Wheat, the minimum cost of
transportation was realized, this was 8/575 Billion Iranian Rials. Using the second scenario, the profit
was maximized for the shipping companies at 9/15 Billions Iranian Rials. Based on the Equilibrium, the
dual function was used to gather where the Capital Loss was the least where the Capital Profit was the
highest. This is the balanced approach.
In conclusion, if the balanced approach is utilized, the result would lead to a lower National Capital Loss
by at least 1 Billion, 100 Million Iranian Tomans and therefore, tells us that in order to minimize National
Cost, but at the same time maximizing Profit for the Transport Company would lead to a lower overall
National deficit.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LOCATING RURAL SOCIAL SERVICE CENTERS IN INDIA
Patel, Nitin R
Management Science (pre-1986); Jan 1979; 25, 1; ProQuest Central
pg. 22
Hussein Akberali
S1314000563
Page 3 of 4
3 LOCATING RURAL SOCIAL SERVICE CENTERS IN INDIA
MINIMUM SPANNING TREE METHOD
MANAGEMENT SCIENCE VOL. 25, NO. 1 JANUARY 1979 (PG. 22)
The paper analyzes the research techniques in planning new roads and service centers in Dharampur,
India. This was crucial in determining the number and location of the service centers over a five year
period.
The objective of the project was to produce and integrated plan of Rural Development aimed at the
rural poor in the region. The service Center was to involve Village Level Workers, Talatis, School
Teachers, Health Center Workers, Cooperative Secretaries, Post Office functionaries. Based on the
analysis of the Poverty, having these functions would help fight two of the four major reasons for
poverty which were inadequate and to an extent inappropriate institutional efforts for development and
Physical Isolation of the community, particularly in hilly areas.
A location and distance analysis was done and it was determined that there would be 44 service centers
in the 237 villages in Taluka, each no further than the hat, eight to ten miles apart. Costs were taken
into consideration and an estimate of Rs. 5.5 Million for the service centers and Rs. 90 Million for the
roads was budgeted for this activity. The Government of Gujarat kept a cap of Rs. 130 Miliion. The
proposed map (Minimal Spanning Tree) with the service centers look like the below Figure:

The possible locations were then researched using a mix between a Linear (minimizing costs) and Non-
Linear (maximizing people and minimizing cost). A very large analysis was done on an IBM 360/44
machine. A construction schedule was built using Dynamic Programming, where an 11 dimensional
Boolean Vector was used to solve this conundrum to maintain and stay within the budget constraints.
Hussein Akberali
S1314000563
Page 4 of 4
These models explained above have proved to be a very valuable tool in Infrastructure Planning for
Regional Development. The savings realized from using these models were substantial, however, is
highly dependable on the overall objective set out at the beginning of the project. At the end of it all, the
statistics derived and realized in such an analysis, gives basis to proceed with such an action and reduces
the overall effectiveness of favoring certain locations based on power, but rather increases the overall
effectiveness in effective planning and budget control.

MBA 643 - Analyzing Quantitative Models

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

MBA 643 - Analyzing Quantitative Models

Enviado por

Direitos autorais:

Formatos disponíveis

I

i f ' ~ AJ.,.;:,ll U...l,>JI

, is estimated in Equation (2):

Você também pode gostar