State of Palestine
Ministry of Transport
DEVELOPMENT OF MODE CHOICE
MODEL FOR WORK TRIPS IN GAZA CITY
Sadi I. S. AL-Raee
EuroMed Regional Transport Project: Road, Rail and Urban Transport (RRU)
Symposium on Transport
Ramallah, 13-14 February 2013
Live video conference with Gaza
Study Objectives
To develop a
mode choice
model for work
trips in Gaza
city
• To provide a quantitative explanation of the
choices of travel modes for work trips in
Gaza city.
• To study the factors affecting the mode
choice
• To specify the most significant factors which
affect mode choice.
• To study the various types of mode choice
models.
• To choose the most suitable model.
• To calibrate and estimate the chosen mode
choice model.
• To validate the developed models.
Problem Statement




Gaza city is currently facing urbanization and economic growth, with
this, demand for private and public transport have been increasing.
To meet the increasing of travel demand without increasing the
congestion problem there is a need to adopt suitable transport
policies. And this is couldn't be achieved without understanding the
travelers’ needs and preference of using the modes
Developing countries including Gaza Strip often use the mode choice
models that are developed by the developed countries. These
models are not suitable to be used as the original form because of
the different conditions and circumstances in developing countries.
Therefore, there is a need to develop mode choice model for Gaza
in order to help in predicting the future demand for each mode of
transport and adopting the suitable transport policies to solve the
congestion problem.
Urban Transportation planning
PRE-ANALYSIS
PHASE
POST-ANALYSIS
PHASE
•Problem/Issue Identification
TECHNICAL –
ANALYSIS
PHASE
•Formulation of Goals and
•Urban Transportation
•Decision Making.
Objectives
• Model System.
•Implementation.
•Data Collection
•Discrete choice Modeling
•Monitoring
•Generation of Alternatives
•Evaluation of Alternatives.
Research Methodology
Review the
Literature
Conclusions and
Recommendations
Initial Survey
Research
Methodology
Validation of
Model
Final Survey
Calibration of
Model
Research Methodology
Review the literature
Initial survey
Final survey
Transportation planning process.
Design of initial survey
questionnaire.
Design of final survey
questionnaire.
Types of mode choice models.
Pilot study
Determination sample size
Model estimation techniques.
Analysis of pilot study
Distribution and collection
Sampling and data collection
General analysis
Research Methodology
Calibration
Calibration of N number of models
Validation
a) Likelihood ratio test LRTs
b) Estimation of prediction ratio
Conclusions
&recommendations
Conclude the main findings
Recommendations
Comparison
between models in
terms of
a) coeff-Estimators
b) t-Statistics
C) Stnd error
d)Overall fit
Comparison LRTS
with critical chi
square value at
95% confidence
level
Analysis _ General information
General information
Frequency
Percent
Male
378
68.5%
Female
174
31.5%
Married
437
79.2%
Single
115
20.8%
Governmental employee
222
40.2%
Private sector employee
138
25%
UN employee
75
13.6%
Business man or special works
21
3.8%
Waged worker
87
15.8%
others
9
1.6%
18-25 years
38
6.9%
26-30 years
107
19.4%
31-35 years
114
20.7%
36-40 years
130
23.6%
41-45 years
85
15.3%
46-50 years
55
9.9%
51-55 years
17
3%
>55 years
6
1.1%
Gender of respondents
Marital Status
Jobs of respondents
Age of respondents
Analysis _ General information
General information
Frequency
Percent
Less than 1000 ILS
16
2.9%
1001-1500 ILS
56
10.1%
1501-2000 ILS
111
20.1%
2001-3000 ILS
181
32.8%
3001-4000 ILS
115
20.8%
4001-5000 ILS
46
8.3%
More than 5000 ILS
27
4.9%
1-6 persons
386
69.9%
7-10 persons
158
28.6%
8
1.5%
Private car
91
16.5%
Motorcycle
112
20.3%
Bicycle
22
4.0%
No means
327
59.2%
Average monthly income
Family size
More than 10 persons
Ownership of transportation means
Trip length
0.3-1.0 km
37
6.7%
1.1-2.0 km
121
21.9%
2.1-3.0km
164
29.7%
3.1-4.0 km
93
16.9%
4.10-5.0 km
77
13.9%
5.1-6.0 km
49
8.9%
More than 6.0 km
11
2.0%
Analysis _ General information
General information
Frequency
Percent
Private car
76
13.8%
Shared taxi
245
44.4%
Taxi
39
7.1%
Motorcycle
90
16.3%
Bicycle
13
2.4%
Walking
89
16.1%
Choice riders
396
71.7%
Captive riders
156
28.3%
The means of transport usually used by respondents
Captivity
Calibration of Model



Multinomial Logit model was used
Maximum likelihood function was used for determining the
estimators.
The Easy Logit Model (ELM) software was used for estimation
of the models
Calibration Criteria






Wrong sign coefficient variables were dropped from the
model.
Variables with insignificant coefficients were dropped from the
model except the level of service variables (travel time and
travel cost).
Some variables with insignificant coefficient were considered
based on its improving the statistics of the model.
The level of service variables were considered in different
forms (strait forward as cost and travel time) or in ration form
such as cost over income.
Some of intuitively important variables which have been
dropped from the model were reconsidered.
The mode specific constants were considered in spite of the
significance of coefficients of the variables.
The Selected Revealed Model
Parameters
Estimated value
Generic Parameters
TT
TC/PINC
Alternative Specific Parameters
CONSTANT
S_Taxi
CONSTANT
Taxi
CONSTANT
Motorcycle
CONSTANT
Bicycle
CONSTANT
Walking
AGE
Bicycle
OWTM
S_Taxi
DIST
Walking
FINC
Motorcycle
FINC
Walking
Model Statistics
Log Likelihood at Zero
Log Likelihood at Constants
Log Likelihood at Convergence
Rho Squared w.r.t. Zero
Rho Squared w.r.t Constants
Adjusted Rho Squared w.r.t. Zero
Adjusted Rho Squared w.r.t Constants
Number of Cases
Number of iterations
Estimation status
Model_8 (MNL)
t-statistics
-0.1299
-227.5075
-3.264
-2.7647
1.0773
-0.5826
4.9794
-15.3013
7.366
0.3979
-1.0626
-2.16
-0.0021
-0.001
1.4379
-0.987
2.7175
-2.2932
4.2017
2.4455
-1.9061
-4.3961
-2.6153
-2.7697
-190.9438
-144.6343
-105.8891
0.4454
0.2679
0.3826
0.2122
368
15
converged, with contants, with zeros, valid lic.
The Utility Functions of Revealed Model
The Selected Stated Preference Model
Parameters
model_S5 (MNL)
Estimated value
Generic Parameters
TT
FREQ
FARE/PINC
Alternative Specific Parameters
CONSTANT
Minibus
CONSTANT
Bus
AGE
Minibus
AGE
Bus
DIST
Bus
FINC
Minibus
FINC
Bus
Model Statistics
Log Likelihood at Zero
Log Likelihood at Constants
Log Likelihood at Convergence
Rho Squared w.r.t. Zero
Rho Squared w.r.t Constants
Adjusted Rho Squared w.r.t. Zero
Adjusted Rho Squared w.r.t Constants
Number of Cases
Number of iterations
Estimation status
t-statistics
-0.3213
-0.0518
-372.5231
-2.5138
-2.127
-2.003
1.8179
1.1165
0.0493
0.069
0.6234
-0.0018
-0.0039
2.7047
0.7961
2.8485
2.2217
3.6936
-7.6099
-6.6471
-542.7145
-380.0125
-250.54
0.5384
0.3407
0.5199
0.318
494
18
converged, with contants, with zeros, valid lic.
The Utility Functions of Stated
Preference Model
Model Validation
Validation test
Test of
Reasonableness
Likelihood ratio
test (LRTS)
Prediction Ratio
Test of reasonableness



This test is performed during the calibration
process depending on the expected sign of
estimators.
All the models with wrong signs of estimators
would not considered as a valid models.
Based on this criterion, The selected revealed and
stated preference models are considered as a
valid models because all the variables for these
models have correct signs of estimators.
Likelihood Ratio Test (LRTS)

This test is conducted using 1/3rd of data sets.
represents the likelihood ratio test statistics which
restricts the parameters estimated from data j to
be used to predict mode share in data i for same
specifications
is log likelihood ratio value when the parameters
are restricting in data j
is log likelihood ratio value when the parameters
are unrestricted in data j
LRTS for Revealed Model


The calculated chi square value for the Selected revealed
model is
the calculated chi square value can’t lead to reject the null
hypothesis stated that there is no difference between the
predicted and observed behavior because the calculated
chi square value is less than critical chi square value at
95% confidence level and twelve degrees of freedom
(21.026).
LRTS for Stated Preference Model


The calculated chi square value for the chosen stated
preference model is
the calculated chi square value can’t lead to reject the null
hypothesis stated that there is no difference between the
predicted and observed behavior because the calculated
chi square value is less than critical chi square value at
95% confidence level and ten degrees of freedom
(18.31).
Prediction Ratio



The last phase for validation process is calculated the
prediction capability of the calibrated model.
The calculated prediction value for revealed model is
0.69 which means that the model is capable to predict
about 69% of the choices of the trip makers’ correctly.
The calculated prediction value for the stated preference
model is 0.80 which means that the model is capable to
predict about 80% of the choices of the trip makers’
correctly.
Conclusions


For revealed model, the total travel time, total travel cost divided
by personal income, ownership of transport means, age, distance
and average family monthly income are the factors that affect
the mode choice for workers in Gaza city. While the gender and
out of vehicle time are statistically insignificant at 90%
confidence level so they are excluded from the model
For stated preference model, the travel time, fare over personal
income, frequency of service, age, average monthly family
income, and distance have an effect on mode choice decision of
workers as they are statistically significant at 95% confidence
level while the gender variable has no effect on mode choice
decision as it is statistically insignificant even at 90% confidence
level.
Conclusions contd.,


The developed revealed at stated preference models are able
to predict the choice behavior of the workers in Gaza city as the
two models are valid at 95% confidence level.
There are six factors affect the captive riderships which are
gender, job, private car ownership, motorcycle ownership, bicycle
ownership, and distance.
Recommendations




Using the developed revealed model in travel demand analysis
and in developing transport policies for Gaza city.
Using the developed stated preference model in studying the
possibility and the feasibility of introducing the bus services for
transport system in Gaza city.
Using the developed stated preference model in establishing
the time table and in determining the appropriate fare for bus
services in Gaza city.
Awareness campaigns should be implemented to encourage
young people for using a bicycle mode.
Recommendations contd.,




In case on introducing a bus service to transport system in Gaza
city awareness campaigns may be needed to encourage the
young people’s for using bus modes.
Further study for developing mode choice models for trips other
than work trips such as social, recreational and study trips.
Studying the effect of captive travelers on mode choice models.
Calibrating the mode choice using probit and generalized
extreme model and comparing them with logit model