THE DISSERTATION ENTITLED

DEVELOPING TRIP GENERATION AND DISTRIBUTION MODELS FOR

THE CITY OF SURAT USING TRANSCAD

SUBMITTED TO
SARDAR VALLABHBHAI NATIONAL INSTITUTE OF TECHNOLOGY
IN PARTIAL FULFILLMENT OF THE REQUIREMENT
FOR THE AWARD OF THE DEGREE OF

MASTER OF TECHNOLOGY
IN

TRANSPORTATION ENGINEERING AND PLANNING

SUBMITTED BY
PARTH DOSHI
SVNIT

RESEARCH SUPERVISOR
DR.G.J.JOSHI
ASSOCIATE PROFESSOR
SVNIT

2011-12
P.G. CENTRE IN TRANSPORTATION ENGINEERING AND PLANNING,
DEPARTMENT OF CIVIL ENGINEERING,
SARDAR VALLABHBHAI NATIONAL INSTITUTE OF TECHNOLOGY,
SURAT-395007

SARDAR VALLABHBHAI NATIONAL INSTITUTE OF TECHNOLOGY,

SURAT

CERTIFICATE
This is to certify that the dissertation entitled Developing trip generation and
distribution models for the city of Surat using TransCAD submitted by Mr.
Parth P. Doshi (P10TP495) in partial fulfillment of the requirement for the award
of the degree of Master of Technology in Transportation Engineering and
Planning of the Sardar Vallabhbhai National Institute of Technology, Surat, is
a record of his own work carried out under my supervision and guidance. The report
has been approved as it satisfies the academic requirements in respect of the
dissertation work prescribed for the M.Tech degree.

(Dr. G. J. Joshi)
Associate professor
& P.G. In-charge (TE&P)
Department of Civil Engineering,
SVNIT

(Dr. N. C. Shah)
Professor
& Section Head (TE&P)
Department of Civil Engineering,
SVNIT

(Dr. J. N. Patel)
Professor
& Head of the Department
Department of Civil Engineering,
SVNI

SARDAR VALLABHBHAI NATIONAL INSTITUTE OF TECHNOLOGY,

SURAT

EXAMINERS CERTIFICATE
This is to certify that the dissertation entitled Developing trip generation and
distribution models for the city of Surat using TransCAD submitted by Mr.
Parth P. Doshi (P10TP495) in partial fulfillment of the requirement for the award
of the degree of Master of Technology in Transportation Engineering and
Planning of the Sardar Vallabhbhai National Institute of Technology, Surat, is
here by approved for the award of the degree.

Date:

Examiners:
1.
2.

ACKNOWLEDGEMENTS
Though only my name appears on the cover of this dissertation, a great many people have
contributed to its production. I owe my gratitude to all those people who have made this
dissertation possible and because of whom my post graduation experience has been one
that I will cherish forever.
First and foremost I would like to acknowledge my Research Guide Dr. G.J. Joshi, Associate
Professor & P.G. Incharge of Transportation Engineering and Planning division, Civil
Engineering Department, S.V.N.I.T., who has given me a new vision to think in the direction
of research. It has been an honor to be his student.
I am very much grateful to Dr. N.C. Shah, Professor & Section Head, and Dr. J.N. Patel,
Head of the Department, and al other faculty members of civil engineering department for
their encouragement.
Many friends have helped me stay sane through these difficult years. Their support and
care helped me overcome setbacks and stay focused on my study. I greatly value their
friendship and I deeply appreciate their belief in me.
Special thanks to my hostel mates for helping me in carrying out surveys.

Parth Doshi

ABBREVIATIONS
CBD

Central business district

FSM

Four stage model

HBO

Home based other

HBW

Home based work

HH

Household

HTR

Household trip rate

MLR

Multiple linear regression

MRT

Mass rapid transit

NH

National Highway

NHB

Non home based

OD

Origin-Destination

PA

Production-Attraction

PCTR

Per capita trip rate

ROW

Right of way

RTO

Regional transport office

SH

State highway

SMC

Surat municipal corporation

SUDA

Surat urban development authority

TAZ

Traffic analysis zone

TDM

Travel demand modeling

TSM

Transportation system management

VOC

Vehicle operating cost

TABLE OF CONTENTS
ABBREVIATIONS
I
ABSTRACT
VII
1.

2.

INTRODUCTION ...................................................................................................... 1
1.1

GENERAL .................................................................................................................. 1

1.2

NEED FOR THE STUDY .............................................................................................. 1

1.3

OBJECTIVES OF THE STUDY ...................................................................................... 2

1.4

SCOPE OF THE WORK ............................................................................................... 2

1.5

RESEARCH METHODOLOGY ..................................................................................... 2

1.6

REPORT ORGANIZATION .......................................................................................... 4

LITERATURE REVIEW ............................................................................................... 5

2.1
2.1.1

URBANIZATION IN INDIA ....................................................................................................... 5

2.1.2

URBANIZATION IMPACT ON TRAVEL ........................................................................................ 6

2.2

URBAN TRANSPORTATION PLANNING .................................................................... 6

2.3

TRAVEL DEMAND FORECASTING ............................................................................. 7

2.3.1

2.4

HISTORY OF TRAVEL DEMAND MODELING ................................................................................ 7

A FOUR STAGE TRAVEL DEMAND MODEL ................................................................ 7

2.4.1

TRIP GENERATION ............................................................................................................... 9

2.4.2

TRIP DISTRIBUTION ............................................................................................................ 10

2.4.3

MODAL SPLIT.................................................................................................................... 12

2.4.4

TRAFFIC ASSIGNMENT ........................................................................................................ 12

2.4.5

MODEL VALIDATION .......................................................................................................... 13

2.4.6

MODEL CALIBRATION ......................................................................................................... 13

2.4.7

TRIP GENERATION CALIBRATION/VALIDATION........................................................................ 14

2.5
3.

URBANIZATION ........................................................................................................ 5

RECENT STUDIES .................................................................................................... 14

STUDY AREA PROFILE ............................................................................................ 19

3.1

LOCATION .............................................................................................................. 19

3.2

DEMOGRAPHICS AND GROWTH ............................................................................ 21

3.3

TRANSPORTATION SYSTEM ................................................................................... 22

3.3.1

ROAD NETWORK ............................................................................................................... 22

II

3.4
4.

DATA COLLECTION ................................................................................................ 29

4.1

5.

6.

STUDY AREA DELINEATION .................................................................................... 23

SURVEYS CARRIED OUT .......................................................................................... 29

4.1.1

INVENTORY DATA .............................................................................................................. 29

4.1.2

FIELD SURVEYS.................................................................................................................. 29

4.1.3

HOME INTERVIEW SURVEYS................................................................................................. 30

4.1.4

DESIGN OF QUESTIONNAIRE ................................................................................................ 30

4.2

TRAFFIC ANALYSIS ZONES ...................................................................................... 30

4.3

SAMPLE DISTRIBUTION .......................................................................................... 35

DATA ANALYSIS .................................................................................................... 36

5.1

HOUSEHOLD STRUCTURE ...................................................................................... 36

5.2

INCOME ................................................................................................................. 39

5.3

WORKING MEMBERS ............................................................................................. 41

5.4

SCHOOL/ COLLEGE GOING MEMBERS ................................................................... 43

5.5

VEHICULAR OWNERSHIP ....................................................................................... 45

5.6

OBSERVED TRIPS .................................................................................................... 47

5.7

DEVELOPMENT OF TRIP PRODUCTION CURVES .................................................... 60

DEVELOPMENT OF MODELS .................................................................................. 62

6.1

TRIP GENERATION MODEL ..................................................................................... 62

6.1.1

REGRESSION APPROACH ..................................................................................................... 62

6.1.2

MODEL STRUCTURE ........................................................................................................... 62

6.1.3

INPUTS ............................................................................................................................ 63

6.1.4

CORRELATION ANALYSIS ..................................................................................................... 63

6.1.5

BUILDING A REGRESSION MODEL IN TRANSCAD ..................................................................... 64

6.1.6

OUTPUT .......................................................................................................................... 65

6.1.7

MODEL EVALUATION ......................................................................................................... 67

6.2

TRIP PRODUCTIONS ............................................................................................... 69

6.2.1

WARD WISE TRIP PRODUCTIONS .......................................................................................... 69

6.2.2

ZONAL TRIP PRODUCTIONS .................................................................................................. 71

6.2.3

CITY LEVEL TRIP PRODUCTIONS ............................................................................................ 71

6.3

TRIP DISTRIBUTION MODEL ................................................................................... 71

III

6.3.1

GRAVITY APPROACH .......................................................................................................... 71

6.3.2

MODEL STRUCTURE ........................................................................................................... 72

6.3.3

INPUTS ............................................................................................................................ 72

6.3.4

DEVELOPING GRAVITY MODEL IN TRANSCAD ......................................................................... 73

6.3.5

OUTPUT .......................................................................................................................... 74

7.

CONCLUSIONS ...................................................................................................... 78

8.

FUTURE SCOPE AND RECOMMENDATIONS ............................................................ 80

9.

REFERENCES ......................................................................................................... 81

ANNEXURES
ANNEXURE A: BREAK UP OF STUDY AREA
ANNEXURE B: HIS SURVEY FORMAT
ANNEXURE C: DATA COLLECTED FROM HIS
ANNEXURE D: WORKING WITH TRANSCAD

IV

TABLE OF FIGURES
FIGURE 2.1 POPULATION GROWTH TRENDS IN INDIA ........................................................................... 5
FIGURE 2.2 FOUR STAGE TRAVEL DEMAND MODEL .............................................................................. 8
FIGURE 3.1 LOCATION OF THE SURAT CITY IN GUJARAT REGION ........................................................... 19
FIGURE 3.2 CITY MAP INDICATING OLD LIMITS, NEW LIMITS AND SUDA LIMITS ....................................... 20
FIGURE 3.3 POPULATION PROJECTION FOR THE CITY .......................................................................... 21
FIGURE 3.4 ZONAL MAP OF THE CITY AS PER NEW LIMITS .................................................................... 24
FIGURE 3.5 ELECTION WARD MAP OF THE CITY AS PER NEW BOUNDARIES ............................................... 25
FIGURE 3.6 MAP SHOWING CENSUS WARDS OF THE CITY .................................................................... 26
FIGURE 4.1 MAP SHOWING AREA OF EACH ZONE .............................................................................. 32
FIGURE 4.2 MAP SHOWING ZONE WISE POPULATION ......................................................................... 33
FIGURE 4.3 MAP SHOWING POPULATION DENSITY OVER THE ZONES...................................................... 34
FIGURE 5.1 MAP REPRESENTING ZONAL POPULATION AND NO. OF HHS ................................................ 38
FIGURE 5.2 MAP REPRESENTING AVG. NO. OF WORKING MEMBERS PER HH ........................................... 42
FIGURE 5.3 MAP REPRESENTING ZONEWISE AVG. NO. OF LEARNERS PER HH .......................................... 44
FIGURE 5.4 MAP REPRESENTING ZONE WISE AVG. VEHICULAR OWNERSHIP ............................................. 46
FIGURE 5.5 GRAPHICAL REPRESENTATION OF ZONAL AVERAGE TRIP RATES.............................................. 54
FIGURE 5.6 ZONAL AVERAGE VALUES OF SURVEY FINDINGS ................................................................. 56
FIGURE 6.1 DEVELOPING A REGRESSION MODEL IN TRANSCAD ........................................................... 64
FIGURE 6.2 CHOOSING DEPENDENT AND INDEPENDENT VARIABLES ....................................................... 65
FIGURE 6.3 PLOT OF OBSERVED VS. EXPECTED TRIP RATES .................................................................. 68
FIGURE 6.4 PLOT OF WARD LEVEL AVG. OBSERVED VS. COMPUTED TRIP RATES........................................ 69
FIGURE 6.5 PROCEDURE TO BUILD GRAVITY MODEL IN TRANSCAD ....................................................... 73
FIGURE 6.6 DESIRE LINE DIAGRAM SHOWING ZONE TO ZONE TRAFFIC FLOWS .......................................... 77

LIST OF TABLES
TABLE 3.1 ZONE AREA AND POPULATION AS PER 2012 ...................................................................... 23
TABLE 4.1 STUDY AREA DETAILS..................................................................................................... 31
TABLE 4.2 NO. OF SAMPLES COLLECTED FROM EACH ZONE .................................................................. 35
TABLE 5.1 ZONE WISE HOUSEHOLD STRUCTURE ................................................................................ 36
TABLE 5.2 ZONE WISE INCOME CHARACTERISTICS .............................................................................. 40
TABLE 5.3 PURPOSE WISE TRIP COMPOSITION AT ZONAL LEVEL ............................................................ 53
TABLE 5.4 CITY LEVEL AVG. VALUES OF PARAMETERS ......................................................................... 55
TABLE 5.5 COMPARISON OF PARAMETERS ....................................................................................... 55
TABLE 5.6 PER CAPITA DAILY TRIP RATE ........................................................................................... 57
TABLE 5.7 VEHICLE OWNERSHIP MODEL .......................................................................................... 60
TABLE 5.8 TRIP PRODUCTION MODEL ............................................................................................. 61
TABLE 6.1 CORRELATION COEFFICIENTS BETWEEN PARAMETERS ........................................................... 64
TABLE 6.2 WARD WISE DAILY TOTAL TRIP PRODUCTION ...................................................................... 70
TABLE 6.3 ZONE WISE TOTAL DAILY TRIP PRODUCTION ....................................................................... 71
TABLE 6.4 ZONE WISE TRIP PRODUCTIONS AND ATTRACTIONS .............................................................. 72
TABLE 6.5 FRICTION FACTOR MATRIX .............................................................................................. 72
TABLE 6.6 ZONAL TRAFFIC FLOWS .................................................................................................. 76

VI

LIST OF CHARTS
CHART 3.1 NO. OF REGISTERED VEHICLES OVER THE PERIOD OF LAST 5 YEARS ......................................... 22
CHART 3.2 ELECTION WARD WISE POPULATION ................................................................................ 27
CHART 3.3 CENSUS WARD WISE POPULATION................................................................................... 28
CHART 5.1 WARD WISE AVG. HH SIZE ............................................................................................ 36
CHART 5.2 WARD WISE TOTAL NO. OF HH ...................................................................................... 37
CHART 5.3 ZONE WISE AVG. HH SIZE ............................................................................................. 38
CHART 5.4 WARD WISE AVG. HH INCOME/ MONTH .......................................................................... 39
CHART 5.5 ZONE WISE AVG. HH MONTHLY INCOME .......................................................................... 40
CHART 5.6 WARD WISE AVG. NO. OF WORKING MEMBERS / HH.......................................................... 41
CHART 5.7 ZONE WISE AVG. NO. OF WORKING MEMBERS PER HH ........................................................ 42
CHART 5.8 WARD WISE AVG. NO. OF LEARNERS / HH ........................................................................ 43
CHART 5.9 ZONE WISE AVG. NO. OF LEARNERS PER HH ...................................................................... 44
CHART 5.10 WARD WISE AVG. VEHICULAR OWNERSHIP / HH ............................................................. 45
CHART 5.11 ZONE WISE AVG. VEHICULAR OWNERSHIP ....................................................................... 46
CHART 5.12 WARD WISE AVG. OBSERVED WORK TRIPS / HH / DAY ...................................................... 47
CHART 5.13 WARD WISE AVG. NO. OF EDUCATIONAL TRIPS / HH / DAY ................................................ 48
CHART 5.14 WARD WISE AVG. NO. OF OTHER TRIPS / HH / DAY .......................................................... 49
CHART 5.15 WARD WISE AVG. NO. OF TOTAL TRIPS / HH / DAY .......................................................... 50
CHART 5.16 ZONE WISE AVG. NO. OF WORK, EDUCATION AND OTHER TRIPS / HH / DAY........................... 51
CHART 5.17 ZONAL TRIP RATE / HH / DAY ...................................................................................... 52
CHART 5.18 ZONE WISE TRIP COMPOSITION .................................................................................... 53
CHART 5.19 AUTO OWNERSHIP SUB MODEL .................................................................................... 60
CHART 5.20 TRIP PRODUCTION MODEL........................................................................................... 61

VII

ABSTRACT
Urbanization is an index of transformation from traditional rural economies to modern
industrial one. Urbanization is a natural expansion of an existing population, namely the
proportion of total population or area in urban localities or areas or the increase of this
proportion over time. The demand for travel in urban areas is derived from the size of the
settlement, distribution and characteristics of population and activities over space. India
shares most characteristic features of urbanization in the developing countries. Surat is the
fastest growing Indian city. The city economy is characterized by large number of small and
medium size unorganized industries. It is experiencing a phenomenal traffic growth due to
the addition of private vehicles and at the same time the city is experiencing a number of
issues related to traffic and transportation. Hence, for sustainable transport and to overcome
current issues and tackle future issues, it is necessary to plan and implement various
planning measures as soon as possible. For this, it is essential to estimate current and future
travel demand. The history of travel demand forecasting has been dominated by modeling
approach and therefore this dissertation is an endeavor to develop 1st and 2nd stage of
aggregate travel demand model to predict travel demand of people. The conventional four
stage method has been widely adopted by planners all over the world. This method
encompasses four phases. These phases are: 1. Trip generation 2. Trip distribution 3. Mode
split and 4. Traffic assignment. For this dissertation, Generation and Distribution models are
developed through transportation planning software package TransCAD for the year 2012.
Population fir the current year has been projected from census 2011 data. Delineation was
done as per zones and election ward boundaries defined by SMC. For primary data home
interview surveys were carried out. The sample size was limited to 705 no. 15 samples were
taken from each 47 election wards of the city. The trip generation model is developed based
on regression technique. Trip production curves are developed based on FHWA simplified
curve technique. The R2 value for the model (aggregate) is 0.67, which is low but
acceptable. With the help of model, ward wise, zone wise and at city level total no. of daily
trip production has been estimated. Per capita per day trip rates are compared for the year
2012, 2011 and 2001. Models for work trips, education trips and other trips are also
developed. Trip distribution model is developed through gravity approach at zonal level.
Finally, traffic flows are represented by desire line diagram.

VIII

1. INTRODUCTION
1.1

GENERAL

Transportation plays an enormous role in our everyday lives. Each of us travels

somewhere almost every day, whether it is to get to work or school, to go shopping, or for
entertainment purposes. In addition, almost everything we consume or use has been
transported at some point. The need for the travel arises due to the separation of activities in
multiple locations. The level and characteristics of travel demand are influenced by the
activity system and the transportation opportunities in the area. In order to analyze, design
and manage existing and future transportation systems, it is necessary to estimate the
existing travel demand and to predict the changes in it that will result from the projects
being studied and/or from changes in external factors.
Urban Travel behavior means movement of peoples form one place to other place
for them purpose by choosing any travel mode. Travel behavior in developing countries has
been varying substantially in recent years with growing vehicle ownership and vehicle use.
Among those behavioral changes, complicated mode-choice and trip chaining patterns are
very common in urban transportation. Because there is a strong relationship between
household members in developing countries regarding their daily travel, the investigation of
household travel behavior is found to be important. Urban Travel behavior is main factor in
transportation planning. Urban travel demand is growing at an exponential rate as a result of
multitude of activities in urban centers coupled with population growth. The travel demand
of an urban area is influence by the various attributes of the area grouped under main two
categories: urban area characteristics & urban population characteristics. Urban area
characteristics include land area urban form, floor area ratio, edge of area, transportation
network, land use pattern, employment density and economic based of the city. Urban
population characteristics includes socio-economic attributes like total population, family
size, Age structure of family, family income, vehicular ownership, and population density.
1.2

NEED FOR THE STUDY

Like many others the city of Surat is attempting to develop an integrated solution to
meet its mobility and access requirements. The city has been experiencing rapid growth in
population during past 20 years. The contours of the city landscape underwent basic and
cosmetic changes in the mid-1990s along with growth of infrastructural facilities, but lack
of a viable transport system, has left the city with a crumbling face. In the absence of any
traffic planning, over 7.5 lakh vehicles leave the city roads gasping for space. The public
transport services have remained as they were years back and need for a proper system to
cater to the growing population is felt overwhelming. Along with this the needs for
infrastructure the needs of augmentation have also arise rapidly. Surat city has good road
network. Though roads are generally wide, network design and development is not
complete. However, the road network in the walled city is congested due to insufficient road
widths. The road based public transport system is bus operated by GSRTC.
1

At present there are only about 100 city buses in Surat. The share of mass
transportation in Surat is very poor. This in turn prompted the people to go in for private
transport. In general, an efficient, affordable and environmental friendly public transport
system reduces the increasing usage of private vehicles and results in saving in travel time
and VOC as well as improvement in environment. To provide an efficient and environment
friendly mass transportation system for Surat, Government of Gujarat initiated a study on
Integrated Public Transit System for the city of Surat & for any urban area to study current
transportation system and to manage it for the future, there is a high need to forecast future
travel demand which will be used in 4 elements of planning: TSM, Long range planning,
Plan refinement, & Updating.
1.3

OBJECTIVES OF THE STUDY

1.4

To understand travel behavior of the people of the city and development of trip
production curves,
Development of trip generation model for the city through transportation planning
software package TransCAD,
Prediction of daily trip productions for the city, and
Development of trip distribution model for the city through transportation planning
software package TransCAD.
SCOPE OF THE WORK

1.5

The models will be developed for the entire geographical area within the jurisdiction
of SMCs latest boundaries (326.51 sq. km.).
Home interview surveys will be carried out for weekdays with a strong statistical
approach.
Generation model will be developed through multiple linear regression technique.
Distribution model will be developed through singly constrained gravity model.
Through trips, internal to external and external to internal trips are excluded here.
Base year is 2012 for the present study. Population for the present year is projected
from census 2011 data.
RESEARCH METHODOLOGY

The research methodology adopted for the study is described below. Methodology
flowchart is presented in fig. 1.1.

The first step is the identification of issue which forms the basis for problem
statement.
Then, Problem statement is defined by objectives, need for the study and study
scope.
After defining problem, literature review is carried out. Books, journal papers,
dissertation reports and various web resources are referred.
Surat city is selected as study area for the present study.
2

Study area is delineated into no. of zones for the ease of data collection and analysis.
Data collection consists of inventory database collection and home interview
surveys.
Data analysis identifies the household and personal characteristics of the people and
their travel demand characteristics.
After data analysis, model is developed through software package TransCAD.
Dissertation concludes with summary of findings.

Figure 1.1 Research methodology

3

1.6

REPORT ORGANIZATION
This report contains seven chapters:

Chapter one gives overview of the dissertation topic, background of the city and
study carried out. It outlines the needs, objectives and scope of the work. It also covers
study methodology.
Chapter two presents the literature review of previous studies related to this search.
It also covers fundamentals of travel demand forecasting, travel demand modeling and its
history and 4 stage travel demand model. It extensively covers trip generation, multiple
linear regressions, trip distribution and gravity modeling approach.
Chapter three outlines the study area profile and its delineation (i.e. making TAZs).
It also covers area demographics, present land use pattern and existing transportation
system.
Chapter four presents the efforts conducted to collect primary and secondary data to
be used in for the analysis in this research.
Chapter five summarizes data analysis of home interview survey findings. It shows
ward and zone wise HH size, income, working members, learners and vehicular ownership
distribution. It also contains purpose wise trip behavior. Further, it explains travel behavior
of the people of the city and represents them through trip production curves. Here various
socio economic and household parameters are compared for the current year with 2011 and
2001.
Chapter six describes development of trip generation and distribution models
through the software and their evaluation. It tells avg. daily trip rate per household per ward,
zone and at city level. Finally total no. of daily trips at city level is forecasted. Gravity
model is represented through desire line diagram. It also discusses some of the basic
features of the software.
Thesis concludes with Chapter seven, which contains conclusions extracted from
data analysis and models and recommendations for future works.

2. LITERATURE REVIEW
2.1

URBANIZATION

Urbanization is an index of transformation from traditional rural economies to

modern industrial one. It is a long term process. Urbanization is a natural expansion of an
existing population, namely the proportion of total population or area in urban localities or
areas, or the increase of this proportion over time. It can thus represent a level of urban
population relative to total population of the area, or the rate at which the urban proportion
is increasing. Both can be expressed in percentage terms, the rate of change expressed as a
percentage per year, period between censuses. Mobility of urban population expressed in
person-km or person-trips per capita per year, has been generally increasing in recent
decades due to rapid urbanization particularly in developing countries like India. Urban
Travel behavior is main factor in transportation planning. Urban travel demand is growing
at an exponential rate as a result of multitude of activities in urban centers coupled with
population growth. The travel demand of an urban area is influence by the various attributes
of the area grouped under main two heads (1) Urban area characteristics (2) Urban
population characteristics. Urban area characteristics include land area urban form, floor
area ratio, edge of area, transportation network, land use pattern, employment density and
economic based of the city. Urban population characteristics includes socio-economic
attributes like total population, family size, Age structure of family, family income,
vehicular ownership, and population density.
2.1.1 Urbanization in India
India shares most characteristic features of urbanization in the developing countries.
Number of urban agglomeration has grown from 5161 in 2001 to 6243 in 2011. Number of
total population has increased from 102.7 crores in 2001 to 120 crores in 2011, whereas
number of population residing in urban areas has increased from 28.53 crores in 2001 to
37.5 crores in 2011. It reflects a gradual Increasing trend of urbanization. India is at
acceleration stage of the process of urbanization. According to 2011 census, in India out of
total population of 120 million about 37.5 million live in urban areas and 82.5 million live
in rural areas. Fig. 2.1 shows population growth in India.
Fig 2.1 Process of Urbanisation in India
1200

population

1000
Total population (in
millions)

800

Urban population (in

millions)

600
400

Rural population (in

millions)

200
0
1901

1921

1941

1961

1981

2001

Census Year

Figure 2.1 Population growth trends in India

5

2.1.2 Urbanization impact on travel

Increased urbanization and growth in the number of motor vehicles and their usage,
longer trip lengths due to expanding city creates a chaotic and agonizing situation in
absence of actuate and appropriate transportation system. Loose land use patrol, inadequate
public transport, incompatible modal mix and efficient transport management and control
add to the malaise. All these factors combined together cause a host of traffic problems.
These problems manifest themselves in the form of deteriorating road safety, traffic
congestion, defilement of environment, noise pollution and marginalization of the less
privileged and wasteful consumption of energy.
2.2

URBAN TRANSPORTATION PLANNING

Planning is defined as the activity or process that examines the potential of future
actions to guide a situation or a by stem towards a desired direction. Planning is oriented
towards the future. As the fundamental goal of transportation is to provide efficient access
to various activities to satisfy human needs, the general aim of transportation planning is
accommodate this need for mobility. In developing countries like ours, the emphasis is more
on urban transportation planning. However, the principles and methodology can well be
applied in regional or national context with due modifications. It was realized quite early
that traffic is a function of land use and various activities on land, generates different
amounts of trips. The number of trips generated forms the basis of the entire transportation
planning process. The important stages in transportation planning process are as follows:

Survey and analysis of existing conditions,

Forecast analysis of future conditions,
Evaluation,
Programmed adoption and implementation, and Continuing study.

Surveys are prerequisite to the planning process. They are carried out to collect data
on various factors that influence the travel behavior. These include inventory of existing
travel pattern, that If existing transportation facilities and land use-economic activities.
Study area for the purpose must be well defined at national, regional or urban level. The
urban area should encompass the existing and potential expanse. An external cordon is laid
to limit the survey area and the area-within is studied in great detail. To facilitate the survey
work, study area is divided into smaller units called zones. Land use, economic factors,
regular shape etc. are the criteria adopted to formulate the zones. Surveys are conducted to
collect data on the following basic movements in between study zones. 1) Intra zonal. 2)
Through movements. 3) Internal to External and viz. versa. This inventory also helps in
identifying the deficiency in requirement. The inventory on land use includes the following:

Type of land use and its intensity,

Population estimation, and
Employment patterns

2.3

TRAVEL DEMAND FORECASTING

Travel demand forecasting refers to the process of estimating future travel demands
and assigning these to an existing or future transportation network using a transportation
demand model.
2.3.1 History of travel demand modeling
A model is an information construction used to represent and process relationships
between a set of concepts, ideas, and beliefs. A travel-demand model can be defined as a
mathematical relationship between travel- demand flows and their characteristics on the one
hand, and given activity and transportation supply systems and their characteristics. The
history of demand modeling for person travel has been dominated by the modeling approach
that has come to be referred to as the FSM. The literature on transportation demand models
is broad and covers a period of more than 40 years. The first partial share demand model
systems were formulated in the 1950s and 1960s. With time they have undergone a number
of developments, both formal and interpretive. Travel, always viewed in theory as derived
from the demand for activity participation, and in practice has been modeled with trip-based
rather than activity-based methods. Trip origin-destination rather than activity surveys form
the principle database. The influence of activity characteristics decreases and that of trip
characteristics increases, as the conventional forecasting sequence proceeds.
Transportation modelers and planners need knowledge of travel behavior, including
route choice, mode choice, destination choice, travel frequency, activity scheduling,
commuting behavior, and pre-travel and en route travel decision making. Since the 1970s,
most modeling emphasis has been based on random utility theory. Different travel options
are assumed to have an associated utility, which is defined as a function of the attributes of
the alternative and the decision makers characteristics. Ben-Akiva et al. (1997) provide a
summary of the state of the art in modeling individual travel choices. They claimed that
there are few satisfactory existing structural models and that there is a need for behavioral
realism, which involves considering heterogeneity of travel preferences, a variety of
decision strategies, differentiation between individual and joint decision making for travel,
improved consideration of information, and travelers states of knowledge.
2.4

A FOUR STAGE TRAVEL DEMAND MODEL

Fig. 2.2 shows a typical four stage travel demand model. It consists of 4 sub models.
These sub models are discussed below. The model starts with defining the study area and
dividing them into a number of zones to achieve homogeneity and considering the entire
transport network in the system. The database also includes the base year population,
economic activities like employment, shopping space, educational, public spaces and leisure
facilities of each zone etc. The current FSM might best be viewed in two stages. In the first
stage, various characteristics of the traveler and the land use - activity system (and to a
varying degree the transportation system) are evaluated, calibrated, and validated to produce
a non-equilibrated measure of travel demand.

In the second stage, this demand is loaded onto the transportation network in a
process that amounts to formal equilibration of route choice only, not of other choice
dimensions such as destination, mode, time-of-day, or whether to travel at all (feedback to
prior stages has often been introduced, but not in a consistent and convergent manner). This
approach has been moderately successful in the aggregate. All four stages of this model are
described briefly after fig. 2.2.

Figure 2.2 Four stage travel demand model

2.4.1 Trip generation

The first step in travel forecasting is trip generation. In this step, information from
land use, population and economic forecasts are used to estimate how many person trips
will be made to and from each zone. This is done separately by trip purpose. Trip purposes
that can be used include: home based work trips (work trips that begin or end at home),
home based shopping trips, home based other trips, school trips, non-home based trips (trips
that neither begin nor end at home) and transit trips. Trip generation uses trip rates that are
averages for large section of the study area.
Trip productions are based on household characteristics such as the number of
people in the household, their income and the number of vehicles available. For example, a
household with four people and two vehicles may be assumed to produce 4.00 work trips
per day. Trips per household are then expanded to trips per zone. Trip attractions are
typically based on the level of employment in a zone. For example a zone could be assumed
to attract 1.50 home based work trips for every person employed in that zone. Trip
generation is used to calculate person trips. These are later adjusted in the mode split/auto
occupancy step to determine vehicle trips.
2.4.1.1 Factors influencing trip generation
The trip generation models are based upon the principle that Land Use Generates
Traffic. The factors influencing trip generation are:
(a) Socio-economic variables: These are variables largely quantify the potential of
households for making trips. Family size, income, vehicle ownership, number of working
and school going member are major socio-economic variables.
(b) Accessibility variables: These are variables usually account for the influence public
transport on trip generation.
Four approaches are available to develop trip generation models.
Cross-Classification: Cross-classification methods separate the population in an
urban area into relatively homogenous groups based on certain socioeconomic
characteristics. Then, average trip production rates per household or individual are
empirically estimated for each classification. This creates a lookup table that may be used to
forecast trip productions.
Regression Models: The first method uses data aggregated at the zonal level, with
the average number of trips per household in the zone as the dependent variable and average
zonal characteristics as the explanatory variables. The 2nd method uses disaggregate data at
the household or individual level, with the number of trips made by an hh or individual as
dependent variable and the hh and personal characteristics as the explanatory variables.
Population Synthesis: It is a micro-simulation technique.
9

Discrete Choice Models: Discrete choice models use disaggregates household or

individual level data to estimate the probability with which any household or individual will
make trips. The outcome can then be aggregated to predict the number of trips produced.
2.4.1.2 Models for Trip Attractions
It is a common practice to use aggregate models in the form of regression equations
for trip attractions. The dependent variable for these aggregate models is the total number of
trip attractions for a specific trip purpose in a traffic zone. The independent variables
usually are employment related and they represent zonal total values. The development of
aggregate models usually requires a full-scale O-D survey. The coefficients of the
regression equations would vary from area to area. The choice of independent variables and
trip purpose categories also may vary from one study to another.
2.4.1.3 Balancing of Productions and Attractions
Due to the definition of productions and attractions, home-based productions in a
zone may not be equal to the corresponding attractions in the same zone. Non-home-based
productions in a zone should equal to corresponding attractions in the same zone. However,
area wide productions of any trip purpose -- home-based or non-home-based should be
equal to the corresponding area wide attractions. Thus,
(HBW P)i = (HBW A)j
(HBNW P)i = (HBNW A)j
(NHB P)i = (NHB P)j
2.4.2 Trip distribution
Trip generation finds the number of trips that begin or end at a particular zone.
These trip ends are linked together to form an origin-destination pattern of trips through the
process of trip distribution. Trip distribution is used to represent the process of destination
choice, i.e. "I need to go shopping but where should I go to meet my shopping needs?. Trip
distribution leads to a large increase in the amount of data which needs to be dealt with.
Origin-destination tables are very large. For example a 120 zone study area would have
14,400 possible trip combinations in its O-D table. Separate tables are also done for each
trip purpose. Here, available approaches are:

Growth factor methods:

Future Travel Patterns are found by applying a growth factor (Expansion factor) to
the present travel pattern. Various forms of growth method are:

Uniform factor

Average factor

Fratar method

Furness method

10

Synthetic methods:

Synthetic methods utilize the existing data to discern a relationship between trip
making, the resistance to travel between the zones and the relative attractiveness of the
zones for travel.
Various forms of synthetic methods are:

Gravity model
Opportunity model

The most commonly used procedure for trip distribution is the 'gravity model'. The
gravity model takes the trips produced at one zone and distributes to other zones based on
the size of the other zones (as measured by their attractiveness) and on the basis of the
distance to other zones. A zone with higher number of trip attractions will receive a greater
number of distributed trips than one with a small number of trip attractions. Distance to
possible destinations is the other factor used in the gravity model. The number of trips to a
given destination decreases with the distance to that destination (i.e. distance is inversely
proportional to the attractiveness).
The distance effect is found through a calibration process which tries to lead to a
distribution of trips from the model similar to that found from field data. Distance can be
measured several ways. The simplest way this is done is to use auto travel times between
zones as the measurement of distance. Other ways might be to use a combination of auto
travel time and cost as the measurement of distance. Another way is to use a combination of
transit and auto times and costs. This method involves using multiplying auto travel times
and costs by a percentage and transit time/cost another percentage to get a composite time
and cost of both modes. Because of calculation procedures, the model must be iterated a
number of times in order to balance the trip numbers to match the trip productions and
attractions found in trip generation.
2.4.2.1 Formulation of a Gravity Model
The basic hypothesis underlying a gravity model is that the probability that a trip of
a particular purpose k produced at zone i will be attracted to zone j, is proportional to the
attractiveness or pull of zone j, which depends on two factors. One factor is the magnitude
of activities related to the trip purpose k in zone j, and the other is the spatial separation of
the zones i and j. The magnitude of activities related to trip purpose k in a zone j can he
expressed by the number of zonal trip attractions of the same purpose, and the effect of
spatial separation between zones i and j can be expressed by a friction factor, Fij k, which is
inversely proportional to an appropriate measure of impedance, usually travel time. The
attractiveness or pull of zone j with respect to zone i is proportional to Aj k Fij k. The
magnitude of trips of purpose k produced in zone i and attracted to zone j, Tij k, of course,
also depends on the number of trips being produced at zone i, Pi k. This can be expressed
mathematically as follows:
11

Tijk = f(Pik, Ajk, Fijk)

The above formulation is not sufficient for estimating the Tij k values because it yet
does not reflect any considerations for other zones that are competing as alternative
destinations for the trips Pi k. Actually, the effective attractiveness of a zone is relative to
others and it can be expressed as the ratio of its own attractiveness with respect to the total.
Thus the relative attractiveness of a zone for trips of purpose k being produced in zone i is
expressed by the ratio:

Ajk Fijk / Ajk Fijk

Dropping the subscript k, the trip distribution model can be written as follows:

Tij =
2.4.3 Modal split
Mode choice is one of the most critical parts of the travel demand modeling process.
It is the step where trips between a given origin and destination are split into trips using
transit, trips by para-transit or as automobile passengers and trips by automobile drivers.
Calculations are conducted that compare the attractiveness of travel by different modes to
determine their relative usage. All proposals to improve public transit or to change the ease
of using the automobile are passed through the mode split/auto occupancy process as part of
their evaluation. It is important to understand what factors are used and how the process is
conducted in order to plan, design and implement new systems of transportation. The most
commonly used process for mode split is to use the logit model. This involves a
comparison of the "disutility" of travel between two points for the different modes that are
available. Disutility is a term used to represent a combination of the travel time, cost and
convenience of a mode between OD. It is found by placing multipliers (weights) on these
factors and adding them together.
Travel time is divided into two components: in-vehicle time to represent the time
when a traveler is actually in a vehicle and out-of-vehicle time which includes time spent
traveling which occurs outside of the vehicle (time to walk to and from transit stops or
parking places, waiting time, transfer time). Out-of-vehicle time is used to represent
"convenience" and is typically multiplied by a factor of 2.0 to 7.0 to give it greater
importance in the calculations. This is because travelers do not like to wait or walk long
distances to their destinations or accessibility to a transit.
2.4.4 Traffic assignment
While the shortest distance between any two points on a plane is described by a
straight line, it is often not possible to actually travel that way. When using an automobile
or a bicycle you must, for the most part, use a path that travels along existing roads; when
using a bus or train you must use a path that consists of different predefined route segments;
even when flying you often must use a path that consists of different flight legs.
12

In some respects, it is pretty remarkable that people are able to make path choices at
all, given the enormous umber of possible paths that can be used to travel from origin to
destination. Luckily, people are able to make these choices and it is possible to model them.
The basic idea which underlies almost all path choice models is that people choose the
best path available to them (where the best may be measured in terms of travel time,
travel cost, convenience, safety, comfort, etc.). In general, this assumption may fail to hold.
For example, infrequent travelers may not have enough information to choose the best path
and may, instead, choose the most obvious path. As another example, in some instances it
may be too difficult to even calculate what the actual best path is, as is sometimes the case
with complicated transit paths that involve numerous transfers or when a shopper needs to
choose the best way to get from one point to several destinations. However, this relatively
simplistic approach does seem to work fairly well in practice.
Traffic assignment is highly time consuming and data intensive phase in the overall
process and is done differently for highway trips and transit trips. The process first involves
the calculation of the shortest path from each origin to all destinations (usually the
minimum time path is used). Trips for each O-D pair are then assigned to the links in the
minimum path and the trips are added up for each link. The assigned trip volume is then
compared to the capacity of the link to see if it is congested. If a link is congested the speed
on the link needs to be reduced to result in a longer travel time on that link. Changes in
travel times mean that the shortest path may change. Hence the whole process is repeated
several times until equilibrium between travel demand and travel supply is attained. So it is
an iterative process. Trips on congested links will be shifted to uncongested links until
equilibrium conditions occur.
2.4.5 Model validation
Validation of travel demand models typically involves the refinement and
adjustment of model components and parameters to ensure that the forecasts replicate baseyear travel conditions and statistics within an acceptable margin of error. Examples of
measures frequently used in the validation process include aggregate measures of travel
such as vehicle miles traveled, vehicle hours traveled, mode split, trip length distribution,
and total trips and trip rates. The traditional approach has focused on replicating observed
base- year conditions within a certain margin of acceptable error.
2.4.6 Model calibration
Calibration is the process of adjusting the appropriate parameters used in the
validation year travel model to achieve an adequate level of accuracy in replicating traffic
counts, and other relevant measures of travel. If the differences between the base year
model results and the count data are within the acceptable limits, it is assumed that the
horizon year model will then reasonably predict future travel demand. The calibration
process begins with trip generation. The relationships are expressed in the form of
mathematical equations that calculate productions and attractions for basic trip purposes.
Based on the zonal socio-economic data, a series of regression functions are used to
calculate the local trip production and attraction total by purpose for each TAZ.
13

The calibration process is driven by the underlying trip length frequency

distribution. In the basic process, either this distribution or its mean is used to judge
convergence. The relative distribution of trip interchanges (matrix cells) is not directly
considered. On one hand, it is difficult to relate any policy variables to these factors, thus, it
is difficult to assess their validity in the future. On the other hand, the resultant base trip
matrix will more closely reflect observed behavior.
2.4.7 Trip Generation Calibration/Validation
Trip generation calibration/validation involves three components: evaluation of the
base year zonal socioeconomic estimates, development and application of a trip production
model, and the development and application of trip attraction model. In a trip generation
step, the trip production rate or model is applied using zonal socioeconomic estimates to
produce trip productions. The analyst is concerned with the reasonableness of the base year
zonal socioeconomic estimates, the appropriate functions or mathematical expressions used
for the trip production model and trip attraction model, and the parameters associated with
these models. The analyst is also concerned with the trip P & A balance. Household travel
surveys are used to estimate trip rates. Surveys do not provide direct estimates of zonal trip
ends. Therefore, there is not an observed number of trip ends to compare with modelestimated trip ends. The analyst can compare summary statistics such as vehicle trips per
household, person trips per household by trip purpose, and the percentage of person trips by
trip purpose for the study area with values for similar study areas.
2.5

RECENT STUDIES

Ashish Vaghani (2011) in his M.Tech. thesis developed trip generation and
distribution models for the city of Surat based on census 2011 data and compared travel
demand behavior with respect to change in land use for the city for year 2011 with that of
2001. He used multiple linear regression technique to develop trip rate equation. Trip
distribution model was developed using gravity model approach. The city was delineated as
per municipal zones and further by election wards. Various household characteristics were
taken as independent variables and trip rate was taken as dependent variables. 600 random
HH samples were taken as a primary data source. From the survey data and secondary data
citys travel demand behavior was analyzed. In his MLR equation, he got coefficient of
correlation 0.56, which is law and unacceptable. It might have been due to small sample
size.
Vaibhav Aggarwal & Ashish Verma (2010) developed Trip-end models for the city
of Thane in Maharashtra based on data from Thane MRT study carried out in 2000. Each
zone in the study area was either classified as a Residential or a CBD zone, and separate
models were developed for trip productions and attractions tend to depend on the total
number of residential workers and the number of school going children in the age group of
6 to 18 years while the attractions trips depend on the zonal employment characteristics. 22
zonal variables used in Model formulation. Zonal trip rate analysis was also carried out on
the existing data. The results obtained after development of the models, trip forecasts have
14

been made for each zone for the target year 2031. Trip estimates for the target year have
also been obtained through trip rate analysis approach. The results from the two methods
have been compared graphically. It was found that if the most significant variables are
considered, the trip rate approach yielded higher values for the estimated trips.
Louis de Grange, Rodrigo Troncoso, and et. al (2009) had presented an alternative
method for estimating gravity models by multiple linear regressions that is based on proxy
variables, thus circumventing the endogeneity problems arising when least-squares
estimators are used. The proxy variable approach generates consistent estimators for a
gravity model without endogeneity bias. The presence of endogeneity is tested for using
statistical tests developed specifically for our application. We conclude that proxy variables
eliminate the endogeneity and produce consistent estimators in gravity models estimated
using least squares. We also find, however, that endogeneity bias has no significant impact
either on gravity model prediction or on urban transportation system planning processes
based on such models.
Dr. Kayitha Ravinder, Dr. S. Velmurugan, J. Nataraju & Dr. S. Gangopadhyay
(2008) attempted to model the travel prediction using the readily available parameters.
About six models were developed utilizing various parameters such as population, average
household income, road network length and area of the city. Among the six models
developed, the predicted values from model 1 and model 6 were very close to the observed
values of Durgapur and Gwalior cities. Other models were predicting more trips than the
observed trips and this might be attributed to the average household income used in the
models for predicting PCTR. The methodology adopted was: (1) Collection of data relating
to traffic flows on strategically selected links of the urban road networks in India. (2)
Collection of details regarding PCTR, HTR, purpose of the trips, mode split, route choices
etc. (3) Collection of data on influencing parameters includes population, population
density, vehicle occupancy, socio-economic characteristics (Income, sex, age, educational
levels, vehicle ownership, occupation etc.) & network characteristics (Type of road and
length etc.) (4) Estimation of passenger flows on the links of road network as per the modes
of travel. (5) Employing different techniques available such as Artificial Neural Networks
etc. to develop models, which relate passenger / vehicular flows, PCTR and HTR with one
or more of the above said influencing parameters. (6) Validation of the models for their
suitability and practicality by selecting one or two typical cities in India for which data is
available from secondary sources / primary sources.
As can be noted from the PCTR models, the average household income was taken
from field observations. However, the reported average income may not be true
representative of the prevailing income levels of the households in the candidate cities
surveyed and this might be due to incorrect information provided by households.
Furthermore after observing the developed models it would be more appropriate to classify
the cities which are having similar demographic and socio economic characteristics to
predict the travel demand estimation reasonably well.

15

Yao liya, Guan Hongzhi & Yan Hai (2008) discussed trip generation model based
on destination attractiveness. In this paper, attempts were made to reduce the forecasting
error produced by uniform trip generation rates for different traffic zones. The behavior of
each traveler was studied instead of the characteristics of the traffic zone. It also discussed
method for calculating the trip efficiency and the effect of traffic zones combined with a
destination selection model based on disaggregate theory for trip generation. Beijing data
was used with the trip generation method to predict trip volumes. The results showed that
the disaggregate model in that paper was more accurate than the traditional method. An
analysis of the factors influencing traveler behavior and destination selection showed that
the attractiveness of the traffic zone strongly affects the trip generation volume. According
to the authors, the traffic zone attractiveness can be calculated for each zone with the final
forecasting result showing that the accuracy of the disaggregate trip generation model is
greatly improved considering the destination attractiveness. There concluding remarks
were: 1) The traffic zone attractiveness evidently influences the trip generation volume.
Factors that affect the traffic zone attractiveness are age, income, objective, having a car or
not, payment mode, time and cost. 2) The traffic zone attractiveness can be calculated for
each zone with the final forecasting result showing that the accuracy of the disaggregate trip
generation model is greatly improved considering the destination attractiveness. 3) The
factors that influence the traffic zone attractiveness should be studied further to improve the
model precision.
Ravin Tailor and Dr.G.J.Joshi (2005) developed trip rate models for the analysis of
the passenger trip generation and trip intensity profile at the zonal and sector levels. Multi
linear regression were used in the study and the explanatory variables for the prediction of
trip rates was household size, income, number of working members per household, number
of school going member per household, number of vehicles per household with R2 value of
0.54. The zonal trip generation pattern has been mapped by the model for forecasting trip
rates for future growth.
Abdel Aal (2004) applied multiple classification analysis in estimating a trip
generation model for the city of Alexandria, Egypt. In this study, household size and
number of vehicles owned by the household were used for defining homogeneous
household groups. Household size had six categories and vehicle ownership three
categories, thus the resultant cross-classification matrix had 18 cells in total. The travel
survey used in the study had 170 households with complete records. With such a small
overall sample size, none of the cells had 30 observations, a statistical requirement for
computing a reliable trip rate. Further, two cells had no observations and eight cells had
observations of 10 or fewer. However, the application of multiple classification analysis
resulted in the cross-classification trip rate matrix that had trip rates in all the cells and the
expected progression in column and row trip-rates, respectively. As an example, for the
same household size, households with a greater number of vehicles had a higher trip rate.
Though the MCA model is data-efficient because it uses information from the entire sample
of households in estimating trip rates, the prediction error in this study was more than 42%
which is not acceptable.
16

Masaya Kawano, Tetsunobu Yoshitake & Takeshi Chishaki (2003) introduced a

concept of zonal trip production to trip generation analysis, which is trip production in the
zonal context, and then defines the zonal-trip production with a mathematical model, after
giving the sequence for analysis of trip generation. In this paper, they focused on how to
develop a functional relationship between regional-trip production and zonal-trip
production. A production generation table similar to an OD table was introduced, which can
specify distribution of trips between a producing zone and a generating zone. Finally, they
discussed the estimation scheme of trip generation analysis and investigated the
transferability of the proposed model in time and space as well as the reproducibility of the
model. The proposed approach could provide the estimates of the trip generation with high
accuracy. The high transferability enables this approach to have the robustness in
forecasting the trip generation in different times and in different spaces. The conclusions of
the paper were: (1) Introducing the zonal-trip production succeeded in representing
explicitly the precise relation between trip production and trip generation. The proposed
approach to trip generation analysis consists of three sequential phases; regional-trip
production, zonal-trip production and trip generation. The figure of distribution of trip-rate
suggests that the regional-trip production should be considered as the regional trip-rate
distribution. The personal attributes, which are sex, age, occupation and the travel purposes,
were chosen as the explanatory variables for the variation of the regional trip-rate
distribution. The travel purpose was endogenously considered and the regional trip-rate
distribution was finally defined as the joint probability of the trips per person in a day and
the travel purpose. (2) The regional trip-rate distribution model was obtained by the
multiplication of the trip-rate distributions considering individually the categorization of
sex, age and occupation. The model parameters were exogenously given by use of Fstatistics of three-way layout ANOVA. The zonal trip-rate distribution for the zonal-trip
production phase was modeled in the same way as the regional trip-rate distribution model
after the identification of statistical distribution. The stochastic method was successfully
developed for balancing of the zonal trip-rate distribution with the regional trip-rate
distribution. (3) The PG table was introduced with a view to representing the distribution of
trips between the producing zone and the generating zone. The PG model was composed of
one part for home-based generating rate and another part for non-home-based generating
rate. The latter was further divided into the discrimination model and the NHB generation
model. The HB generation model and the NHB generation model were formulated using
logit type model, while the discrimination model was developed with linear discriminant
function. Three models were all certified to have the high reproducibility through the model
calibration.
Thomas F. Golob (2000) proposed a simultaneous model of household activity
participation and trip chain generation in which trips are generated in conjunction with outof-home activities, and time spent traveling is another component of overall time use. The
model jointly forecasts three sets of endogenous variables, namely: 1. Activity
participation, 2. Travel time (together making up total out-of-home time use), and 3. Trip
generation as a function of household characteristics and accessibility indices.

17

It is estimated with data from the Portland, Oregon 1994 Activity and Travel Survey.
Results showed that the basic model, which has 10 endogenous time use and trip generation
variables and 13 exogenous variables, fits well, and all postulated relationships were upheld.
Test showed that the basic model, which divides activities into work and non-work, can be
extended to a three-way breakdown of subsistence, discretionary and obligatory activities.
The model can also capture the effects of in-home work on trip chaining and activity
participation. He concluded: Based on these initial tests results, it is not difficult to specify
and estimate families of trip generation models that provide insights about travel behavior
that are not obtainable using conventional approaches that treat trip generation as an isolated
set of events. The approach was to model trip generation jointly with time use. The models
can reveal how the generation of simple and complex trip chains is interrelated with demand
for out-of-home and in-home activities, and how travel time budge' effects can affect
activity demand and trip generation. The models reveal how the time-use and trip
generation interrelationships are affected by household characteristics and by residential
accessibility. Joint models of time use and trip generation can also be used to forecast the
effects of exogenous shocks to the endogenous variables because they capture the effects of
activity participation, trip chaining, and travel times on all the endogenous variables. For
example, we can trace the effects of travel time changes due to congestion on activity
participation and trip generation. Sensitivity to network variables was also demonstrated,
and it will not be difficult to apply additional exogenous accessibility variables. Models can
be expanded to break down travel by mode, adding network-based levels-of-service
exogenous variables. With such a model, forecasts of trip generation would be sensitive to
network link volumes and other location-dependent measures of transportation system
attributes.
Ajay Kumar & David Levinson (1992) discussed the development of an afternoon
peak period trip generation model for both work and non-work trips. Three data sources
were used in model development, a HH travel survey, a census-update Survey, and a trip
generation study. Seven one-direction trip purposes were defined, specifically accounting
for stops made on the return trip from work to home. Trips were classified by OD activities
rather than by production and attraction, so reframing the conventional scheme of homebased and non-home-based trips. Prior to estimating the model, the HH travel surveys were
demographically calibrated against the Census- update to minimize demographic bias. A
model of home-end trip generation was estimated using the HH travel surveys as a crossclassification of the demographic factors of age and household size in addition to dwelling
type. The model was validated by comparison with a site based trip generation study, which
revealed an under-reporting of the relatively short and less regular shopping trips.

18

3. STUDY AREA PROFILE

Surat city is the administrative capital of Surat district and one of the fastest growing
cities in India with a population in 2011 nearly the double of that in 2001. The city has been
ranked 9th in the list of Indias largest cities. The city is well connected by road, rail and air
transport. Several National and State Highways passes through the city. It is also connected
through sea port. As a result traffic coming to as well as passing through the city is very
high. The city transportation system is predominantly road based. With the absence of a
public transport system, dependence on individual modes and intermediate public transport
is high.
3.1

LOCATION

Surat is located in well-developed South Gujarat region. It is 13 m above the mean sea
level. The city occupies a pivotal position on the Ahmedabad Mumbai regional corridor
centrally located at a distance of 260 km North of Bombay and 224 km South of
Ahmedabad, as well as on the 225 km long industrial belt, having direct linkages with the
industrial urban centers of Vadodara, Ankleshwar and Vapi. The National Highway No. 8
passes within 16 km of the SMC boundary and is one of the busiest inter-state trunk routes
in the country. The city is well connected by road and rail with the major cities and towns of
the state as well as states of Maharashtra and Madhya Pradesh. Fig. below presents the
location of city in the Gujarat State.

Figure 3.1 Location of the Surat city in Gujarat region

19

Figure 3.2 City map indicating old limits, new limits and SUDA limits
20

3.2

DEMOGRAPHICS AND GROWTH

On the whole, rapid pace of economic growth continuing, the rate of growth of
population in the city region is likely to remain high. However, the diamond and textile
sector are becoming semi-capital intensive. This, along with the growth of observed in the
large and medium scale industries in the region, especially in the petrochemical sector
would have impact on the nature and size of employment generation. With industries
becoming capital intensive, nature of labor force requirements would also change. The wage
levels would also improve. The growth of large and medium scale industries in the region,
especially in the down stream petrochemical activities would further add to the change.
Hence some moderation in the population growth rate is expected. Further, as per the
forecast made by Registrar General, population growth is likely to be moderate. Keeping
these in context three scenarios have been developed. While this would bring dampening
effect on the population growth rate, the quality of jobs would be better both in terms of
education and salaries. Better quality workforce, but in reduced numbers would be in
demand. Consequently, demand for transport both in quality and quantity would alter. In the
absence of quality public transport, high growth in personalized vehicles may be expected.
Fig. 3.3 shows projection of city population based on past growth trends.

Figure 3.3 Population projection for the city

21

3.3

TRANSPORTATION SYSTEM

The city is well connected by road and rail transport. Several NHs and SHs pass
through the city. It is also connected through sea port. As a result traffic coming to as well
as passing through the city is very high. The city transportation system is predominantly
road based. With the absence of a public transport system, dependence on individual modes
and intermediate public transport is high. The vehicles registered in Surat RTO area has
raised from four lakh in 1994 to fourteen lakh in 2008. Two wheelers comprise nearly 80%
of the total number of vehicles while cars constitute about 10.3%. In terms of composition,
there is a significant shift towards space consuming modes of transport. This shift is from
bicycles to scooters and three wheelers; and from scooters to four wheelers. The rate of
growth has remained high at 10 to 11%. In the absence of public transport system in the
city, the rate of increase in auto rickshaws has been rapid and increase in motor cars is at
12% per year. Chart below shows the total no. of registered vehicles over the last five years.

Total no. of vehicles

1800000

1650000

1500000

1350000

1200000
2007

2008

2009

2010

2011

Year

Chart 3.1 No. of registered vehicles over the period of last 5 years
3.3.1 Road network
The roadway system of Surat is around 1150 Km. The two urban local bodies; SMC
and SUDA, are responsible for developing, operating and maintaining road infrastructure. A
total of 9.2 % of SUDA area is proposed under Transport and Communication, which is
relatively less. The street network in Surat may be classified as ring radial form and grid
iron pattern is observed on the smaller roads of the local network. The rings and the radials
are incomplete. As a result a small portion of the network acts as arterial roads. The paths of
movement in Surat are very concentric, and roads leading to these places are less. Majority
of the vehicular load is taken by the inner ring road which connects to all important roads.
This results in congestion on the ring road. As far as ROW is concerned, the city is well
placed with about 35% of its network having width exceeding 30 meters. It is to be noted
that many of these roads form part of NH/ SH network. Another 25% of the network is
wider than 18 meters. Within the developed area, road density is higher but roads are
narrower.
22

3.4

STUDY AREA DELINEATION

For the present study, area of Surat city is divided in 7 zones same as Surat
Municipal Corporation. Fig. 3.4 shows zones of the city. These 7 zones are further sub
divided in 47 election wards. Fig. 3.5 shows election wards of the city. These election wards
are further sub divided into 101 census wards. These wards are shown in fig. 3.6. Details of
wards population are given in Annexure A. Zone area and population as projected from
census 2011 data is given in table 3.1.
Table 3.1 Zone area and population as per 2012
Zone.
no.

Zone

Popular
name

Area
(Sq.Km.)

Total
Population

Central Zone

CBD

8.89

448760

North Zone

AmroliKatargam

43.90

305799

East Zone

Varachha

27.14

1325286

South Zone

Udhna

55.95

799807

South West
Zone

Athwa

116.05

642159

West Zone

Adajan

51.27

220955

South- East
Zone

Limbayat

19.88

1057233

326.515

4800000

City

23

Figure 3.4 Zonal map of the city as per new limits

24

Figure 3.5 Election ward map of the city as per new boundaries
25

Figure 3.6 Map showing census wards of the city

26

Thousands

Fig. 3.6 and 3.7 show election ward population and census ward population respectively.
400
360
320

280

Population

240
200
160
120
80
40
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Election ward no.

Chart 3.2 Election ward wise population

27

Hundreds

4000

3500

Population

3000

2500

2000

1500

1000

500

0
1

10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100

Census ward no.

Chart 3.3 Census ward wise population

28

4. DATA COLLECTION
For the planning process preparation of database is pre requisite step which is
carried through inventory studies, field studies and personal interviews. Field studies should
emphasize the observation of real user behavior. Without basic data it is difficult to judge
the present condition that prevails and it is difficult to have planning. In transport planning,
after study area delineation, collection of data for studying the past and existing socioeconomic, traffic and travel characteristics is a major activity. Also accurate and large
database is required to clearly assess the problem and formulate policy and plans for future,
therefore proper data collection becomes very important, especially when the same kind of
data is collected from the various sources. The collected, processed, arranged and
interpreted data should be free from any bias. This section of study deals with methodology
of the data collection and types of data sources for the study purpose.
4.1

SURVEYS CARRIED OUT

Various types of surveys are carried out to assess two main components of transport
planning. The first is known through the inventory search and the second is through the
field surveys.
4.1.1 Inventory data
Inventory data helps to formulate the strategy of the field surveys. It also helps in
designing the questionnaire and sample size estimation. The inventory data helps to decide
the type and extent of field surveys as well. The inventory data collected are as follow.

Maps of the city,

Data related to demographics,
Census wards wise population,
Land use pattern and its maps,

4.1.2 Field Surveys

Basically there are three way for field survey.
(1) By self,
(2) By direct inspection, and
(3) By interview.
In self survey method, reply with return post-card is quite economical but it has very
less response rate. In the method of direct inspection field observation is required for
specific objective; because of this reason response of individuals is not required. Third
method either conducted on telephone or personally is found to be most effective method to
gather the travel behavior of the people in a study area with reference to their socioeconomic background. The field interviews can be carried out by any of the following
methods:
29

The data collection at the origin or home,

The data collection during the trip, and
The data collection at the destination or office/workplace.

Each of the above methods has various merits and demerit. Detailed information
regarding household and travel characteristics may be collected by the data collection at the
origin or destination, which is not possible in the case of data collection during the trip. The
limited time and scope does not permit to acquire detailed data about the person being
interviewed.
4.1.3 Home interview surveys
Home Interview Survey method is one of the most reliable types of surveys for
collection of socio-economic and travel data. The information on a travel pattern like
number of trips made, purpose of trip, travel mode, and information on house hold
characteristics like type of dwelling units, number of residents, age, sex, vehicle ownership,
number of drivers, family income is collected in the HIS. These data are useful to relate the
amount of travel to various socio-economic attributes of household and zonal characteristics
and develop trip rate equations and classified matrices for the trip generation rates. For the
present work, home interviews were carried out for 705 households of the city.
4.1.4 Design of Questionnaire
The layout of questionnaires was done in such a way so as to collect all the relevant
information for the purpose to analyze present travel pattern and to forecast for the horizon
year. The questionnaire designed to collect the information was divided into two parts as
follow.
PART A: Socio economic Characteristics: In this part, location of households,
number of residents, working members, numbers of school and college going members,
number of vehicles owned by the family and activities of family members were collected.
PART B: Trip characteristics: In this part travel related data like purpose wise and
mode wise trips data, locations for different activities, travel times and probable daily trips
were collected. The survey format is shown in Annexure B: HIS format. Through trips,
external to internal and internal to external trips have not taken into account.
4.2

TRAFFIC ANALYSIS ZONES

For Home Interview Survey, Surat is considered to be bounded by an external

cordon as per the boundary of SMC. The city spreads over 326.51 Sq. km. with population
of 48 lacs for the year 2012. The census ward is taken as a micro level unit for the study.
Then the study wards are built from 2 to 3 census wards boundaries as the meso level unit. 5
to 9 study wards form zones, matching with the administrative zones of the SMC at macro
level. The number of census wards, study wards and zones in the study area respectively are
101, 47 and 7. Thus 47 study wards are made to cover all the 101-census wards of Surat
city. Samples are taken randomly from all the election wards.
30

The zoning features of study area are described below:

The area of Surat city was divided into 7 zones: Central, North, East, South, SouthEast, South-West and West.
The Seven Zones were further divided in 47 election wards comprising of 2 to 3
census wards each.
The existing census ward boundaries were considered, while delineating the study
wards, as data was available on census ward basis.
The election and census ward names, numbers and population details are shown in
Appendix A: Break up of study area.
Zone wise details of study area study ward numbers, area, population, and density
are summarized in table 4.1.
Table 4.1 Study area details

Zone
No.
1

Zone
Central
Zone
(CBD)
North
Zone
(Katargam)
East
Zone
(Varachha)
South
Zone
(Udhna)
South West
Zone
(Athwa)
West
Zone
(Adajan)
South- East
Zone
(Limbayat)
City level

Election wards

Census wards

Population
(2012)

Density
(ppha)

1,2,3,4,5,6

1,30,2,3,4,7,8,9,
10,5,11,12

448760

505

7,8,9,33,34

29,31,38,39,40,
41,42,70,71,73,
72

305799

70

10,11,12,13,14, 32,43,28,44,45,3
6,46,47,48,74,75
35,36,37
,76,77

1325286

488

15,19,20,21,23, 34,56,57,53,55,6
6,54,87,88,85,86
39,40,41
,82,83,84

799807

143

24,25,26,27,28,
42,43,44,45

37,58,59,60,61,
13,33,89,90,94,
91,92,93,95,96,
97,99,100,101

642159

55

29,30,31,
32,46,47

27,1426,63,64,65,67,
68,69

220955

43

16,17,18,22,38

35,50,49,51,52,7
8,79,80,81

1057233

532

47

101

4800000

147

31

Fig. 4.1 represents map, generated in TransCAD, showing area of each zone.
Similarly fig. 4.2 and 4.3 represent zonal population and density respectively. Central zone
is the densest zone, while South West Zone is having the lowest density.

Figure 4.1 Map showing area of each zone

32

Figure 4.2 Map showing zone wise population

33

Figure 4.3 Map showing population density over the zones

34

4.3

SAMPLE DISTRIBUTION

Generally sample distribution is based on population distribution. But, from

statistical point of view, 30 samples are required from each ward. Initial plan was to collect
30 samples from each ward. Due to time and budget constraints, 15 samples from each ward
have been collected. Table below shows no. of samples taken from each zone.
Table 4.2 No. of samples collected from each zone
ZONE NO.

ZONE NAME

1
2
3
4
5
6
7

Central Zone
North Zone
East Zone
South East Zone
South Zone
South West Zone
West Zone
TOTAL

NO. OF
WARDS/ZONE
6
7
8
8
9
6
7
47

SAMPLES COLLECTED
90
75
120
120
135
90
75
705

35

5. DATA ANALYSIS
For the city planning process the present characteristics of city households are
highly important. To have the effective plan, policy and decisions, present scenario of the
citizen is a governing factor. The city of Surat had undergone expansion in year 2006 from
112.28 sq.km to 326.52 sq.km. This lead to change in land use pattern and because of that
city travel behavior was also changed. The home interview survey was carried out of 705
house hold for the year 2012. Findings of HIS are described below.
5.1

HOUSEHOLD STRUCTURE

Due to joint family concept in India, the family size remains higher. In the study
area the family pattern in most cases was observed between 3 to 6. Table 5.1 shows the
range of household size for various zones of the city and city level average. The household
size varies from minimum 1 to maximum 9 with an average of 4.49. Data collected through
home interview survey is given in Appendix C: Data collected from HIS.
Table 5.1 Zone wise Household structure
MIN. MAX. AVG.
AVG.
AVG.
AVG.
HH
HH
HH
WORKING
OWNED
LEARNERS/HH
SIZE SIZE SIZE MEMBERS/HH
VEHICLES/HH

ZONE
NO.

ZONE

CENTRAL

4.51

1.41

1.20

1.48

NORTH

4.68

1.36

1.36

1.81

EAST

4.70

1.33

1.49

1.97

SOUTH

4.27

1.47

1.22

2.07

SOUTH
WEST

4.39

1.48

1.18

2.16

WEST

4.61

1.42

1.29

2.20

SOUTH
EAST

4.37

1.19

1.43

1.61

36

Chart 5.1 shows ward wise avg. HH size. Ward 37 is having the highest avg. HH size of 5.25, while ward 45 is having the lowest avg.
HH size of 3.51.

Chart 5.1 Ward wise avg. HH size

36

Chart 5.2 shows no. of households in each ward (estimated from surveys). Ward 9 is having the highest no. of HHs, while ward 12 is
having the lowest no. of HHs.

Chart 5.2 Ward wise total no. of HH

37

Chart 5.3 shows zone wise average HH size. South zone is having the lowest avg. HH size of 4.27. East zone is having the highest avg.
HH size of 4.70. Fig. 5.1 represents population of each zone and no. of HHs in each zone on a map of Surat.

Chart 5.3 Zone wise avg. HH size

Figure 5.1 Map representing zonal population and no. of HHs

38

5.2

INCOME

Monthly HH income of each sample HH is shown in annexure B. It is observed that the minimum monthly income per household varies
from Rs. 5000/- to Rs. 10000/- where as the maximum monthly income per household varies from Rs. 300000/- to Rs. 55000/- in various wards.
The zonal average family income is showing a variation from Rs. 22000/- per month to Rs. 22000/- per month. On the whole, a household in the
city is found to have an average income of Rs. 26284 /-. The ward wise average income is presented through chart 5.4. Ward 11 is having the
highest avg. monthly income per HH, while ward 38 is having the lowest avg. monthly income per HH.

Chart 5.4 Ward wise avg. HH income/ month

39

Table 5.2 shows zone wise min., max. and average HH income. The avg. income values are represented graphically in chart 5.5. From
the table it can be seen that HHs of East zone are having the highest avg. monthly income of 28067 Rs., whereas for South East zone the value is
least with 23253 Rs./ month.
Table 5.2 Zone wise income characteristics
Zone no.
Zone name
1
CENTRAL
2
NORTH
3
EAST
4
SOUTH
5
SOUTH WEST
6
WEST
7
SOUTH EAST
CITY LEVEL

Min. Income (k)*

10
10
11
8
5
10
8
5

Max. Income (k)*

60
75
70
55
55
55
50
75

Avg. Income (k)*

25.100
25.693
28.067
26.258
27.126
26.878
23.253
26.284
* Income in Rs / month

Chart 5.5 Zone wise avg. HH monthly income

40

5.3

WORKING MEMBERS

Chart 5.6 shows ward wise avg. no. of working members per HH. Ward 43 is having the highest avg. no. of working members with the
value of 1.90, while ward 22 is having the lowest avg. no. of working members per HH with the value of 0.88.

Chart 5.6 Ward wise avg. no. of working members / HH

41

Chart 5.7 shows zone wise avg. no. of working members per HH. South zone is having the highest no. of work participation with avg.
value of 1.48 working members per HH, while South East zone possesses the least work participation with avg. working members per HH is
1.19.

Chart 5.7 Zone wise avg. no. of working members per HH

Figure 5.2 Map representing avg. no. of working members per HH

42

5.4

SCHOOL/ COLLEGE GOING MEMBERS

Chart 5.8 shows ward wise avg. no. of learners (i.e. school/college going persons) per HH. Ward 37 is having the highest no. of learners,
while ward 45 is having the least no. of learners per HH.

Chart 5.8 Ward wise avg. no. of learners / HH

43

Chart 5.9 shows zone wise avg. no. of learners per HH. East zone is having the highest ratio of school/ college participation with 1.49
learners per HH, while for South West zone, this ratio was observed to be lowest with the value of 1.18. Fig. 5.3 represents the same thing on a
map.

Chart 5.9 Zone wise avg. no. of learners per HH

Figure 5.3 Map representing zonewise avg. no. of learners per HH

44

5.5

VEHICULAR OWNERSHIP

Chart 5.10 shows ward wise avg. no. of owned vehicles per HH. Ward 42 is having the highest vehicular ownership with an avg. value of
2.81 owned vehicles per HH while ward 23 is having the least vehicular ownership with an avg. value of 1.18 vehicles per HH.

Chart 5.10 Ward wise avg. vehicular ownership / HH

45

Chart 5.11 shows zone wise avg. vehicular ownership. Central zone possesses the least vehicular ownership with avg. no. of owned
vehicles are 1.45. Fig. 5.4 represents these average values on a map.

Chart 5.11 Zone wise avg. vehicular ownership

Figure 5.4 Map representing zone wise avg. vehicular ownership

46

5.6

OBSERVED TRIPS

Chart 5.12 shows ward wise avg. no. of observed trips for work purpose per HH per day. Ward 15 is having the highest avg. no. of work
trips with a value of 2.10 work trips per HH per day, while ward 22 did possess the least avg. no. of observed trips for work purpose with a value
of 1.09 daily work trips per HH. Avg. value of work trip rate per day per HH is found to be 1.64 for 705 samples.

Chart 5.12 Ward wise avg. observed work trips / HH / day

47

Chart 5.13 shows ward wise avg. no. of trips for education purposes per HH per day. Ward 25 is having the highest avg. value of 2.20,
while ward 43 is having the lowest avg. value of 0.92. Avg. educational trip rate per HH per day is 1.62 for 705 samples. Ward 40 to 47 was
showing less educational trip rates compared to other wards.

Chart 5.13 Ward wise avg. no. of educational trips / HH / day

48

Chart 5.14 shows ward wise avg. no. of trips for purposes other than work and education per HH per day. Ward 42 and 44 were having
the highest avg. value of 2.77, while ward 30 was having the lowest avg. value of 1.30 other trips per HH per day. Trip rate for avg. other trips
per HH per day is found out to be 1.97 for 705 samples.

Chart 5.14 Ward wise avg. no. of other trips / HH / day

49

Chart 5.15 shows ward wise avg. no. of total trips per HH per day. Ward 15 is having the highest avg. value of 6.10, while ward 34 is
having the lowest avg. value of 4.20. Avg. trip rate per HH per day is found out to be 5.24 for 705 samples.

Chart 5.15 Ward wise avg. no. of total trips / HH / day

50

/ DAY

Fig. below shows zone wise avg. daily work, educational and other trips. South East
zone is having the lowest work trip rate, while South zone is having the highest work trip
rate. South zone is having the least education trip rate, while East zone is having the highest
trip rate for the same. For all the zones the trip rate for other trips was observed in between
1.70 to 2.10.

Fig. 5.16 Zone wise avg. no. of work, education and other trips / HH / day
51

Avg. total trips / day / hh

Fig. below shows zone wise trip rate. The South East zone is having the least trip
rate with 4.91 avg. total daily trips per HH. East zone is having the highest trip rate with
5.54 avg. total daily trips per HH.

Fig. 5.17 Zonal trip rate / HH / day

Table 5.3 shows break up of daily trips by purpose. It indicates that out of all trips
produced in city, 32% are of work trips, 30% are of education trips and 38% are of other
purpose trips. Here other purpose trips include shopping, recreational, social and other
purpose trips. These proportions are against general case. In general, work trips should
contribute about 50%, while education purpose trips should account to 30-40%. Other trips
generally remain lower than work and education trips. South east zone shows lowest
amount of education purpose trips accounting to 20% and highest amount of other trips
accounting to whopping 44%. Chart 5.18 compares purpose wise trips at zonal level.

52

Table 5.3 Purpose wise trip composition at zonal level

Central

% DAILY WORK
TRIPS
33

% DAILY EDUCATION
TRIPS
30

% DAILY OTHER
TRIPS
37

North

28

32

40

3
4

East
South

31
29

33
35

36
36

South west

32

33

35

West

32

26

42

South east

44

20

36

CITY

33

30

37

ZONE NO.

ZONE NAME

45
40
35

% trips/ day

30
25
Work trips

20

Education
trips

15

Other trips

10
5
0

Central

North

East

South

South
west

West

South
east

Zone

Chart 5.18 Zone wise trip composition

53

Fig. 5.5 represents trip rate at zonal level on a city map. The map shows proportion
of trips for various purposes and total trip rate for a particular zone.

Figure 5.5 Graphical representation of zonal average trip rates

54

Table 5.3 shows average values of survey findings of 705 data samples (i.e. city
level avg. values). The avg. HH size for the city is observed to be 4.49, workers per HH are
1.39, learners per HH are 1.30 and avg. vehicular ownership 1.93 per HH. Avg. HH income
at city level is observed to be 26284 Rs./ month.
Table 5.4 City level avg. values of parameters
HH SIZE

HH MONTHLY
INCOME (Rs.)

WORKING
MEMBERS /HH

LEARNERS
/HH

NO. OF OWNED
VEHICLES /HH

4.49

26284

1.39

1.30

1.93

Table 5.5 Comparison of parameters

CITY LEVEL COMPARISON

2001*

2011*

2012

5.07

4.22

4.49

11950

26310

26284

No. of working members

1.30

1.50

1.39

No. of learners

1.03

1.26

1.30

No. of owned vehicles

1.95

2.09

1.93

Family size
Income /month (Rs.)

* Source: Vaghani A. M.Tech. dissertation, 2011

25.00

No.

20.00
15.00
10.00
5.00
0.00
HH size

Income/mont
h (*1000 Rs.)

Working
members

Learners

Vehivular
ownership

2001

5.07

11.95

1.30

1.03

1.95

2011

4.22

26.31

1.50

1.26

2.09

2012

4.49

26.28

1.39

1.30

1.93

Chart 5.19 Comparisons of parameters

55

Fig. 5.6 represents zonal level average values of HH size, monthly income, no. of
working members, no. of learners and vehicular ownership on an area of respective zone
with bars.

Figure 5.6 Zonal average values of survey findings

56

Table 5.6 shows per capita per day trip rate for 2001, 2011 and current year. In 2001
the city was having 32 wards. Results show that it has been increased compared to 2001, but
decreased compared to 2011. For present year ward 39 is showing highest per capita daily
trip rate with 2.01 daily trips, while ward 38 is showing lowest value with 1.31 trips. Chart
5.20 shows per capita per day trip rate comparisons for the year 2001, 2011 and 2012.
Table 5.6 Per capita daily trip rate
Ward
No.

Trips generated /person /day

2001*

2011*

2012

1.78

1.88

1.51

2.05

1.89

1.80

0.93

1.73

1.56

1.31

2.05

1.54

2.27

1.83

1.56

1.61

1.78

1.49

1.06

1.62

1.69

1.54

2.21

1.89

1.64

1.70

1.64

10

1.61

2.44

1.68

11

1.53

1.98

1.93

12

1.55

1.96

1.78

13

1.70

1.94

1.67

14

1.75

1.93

1.77

15

0.80

1.75

1.89

16

1.33

1.50

1.71

17

1.82

1.78

1.61

18

1.56

1.58

1.74

19

1.68

1.49

1.88

20

1.45

1.48

1.56

21

1.34

1.35

1.54

57

22

1.62

1.28

1.75

23

1.81

1.84

1.67

24

2.09

1.49

1.63

25

1.73

1.75

1.88

26

1.70

1.76

1.56

27

1.72

2.90

1.67

28

2.10

1.58

1.53

29

1.80

1.75

1.41

30

2.20

1.79

1.52

31

1.80

2.12

1.78

32

1.90

2.86

1.45

33

1.22

1.57

34

1.96

1.34

35

1.26

1.63

36

1.33

1.45

37

1.86

1.50

38

1.56

1.31

39

1.55

2.01

40

1.21

1.83

41

1.40

1.89

42

1.67

1.87

43

2.50

1.71

44

2.21

1.86

45

1.46

1.80

46

1.77

1.62

47

1.79

1.83

City level

1.65

1.78

1.67

*Source: Vaghani A., M.Tech. dissertation, 2011

58

Ward no.

47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1

2012
2011
2001

0.5

1.5

2.5

Daily trip rate

Chart 5.20 Per capita per day trip rate comparisons for the year 2001, 2011 and 2012
59

5.7

DEVELOPMENT OF TRIP PRODUCTION CURVES

This model relates HH income and auto ownership. From HIS information regarding
no. of trips, vehicle ownership and income for each HH is available. From these data, a
matrix is established based upon income range and vehicle ownership. Table 5.7
summarizes the data from stratification of samples. The number in each cell shows % of
HHs in each category of vehicular ownership and income group. Now smooth curves are
plotted for % of HHs against income groups for various vehicular ownerships. These curves
may be extended depending upon the shape of the curves. Chart 5.21 illustrates the curves
developed from the above data.
Table 5.7 Vehicle ownership model
Income group*
5 15 k
16 25 k
26 35 k
36 45 k
>45 k

0
1.99
0.99
0.14
0.00
0.00

Owned vehicles
1
2
3
7.38
1.28
0.57
18.87 17.73 3.40
5.39
19.72 8.37
0.14
2.70
3.83
0.00
0.28
0.71

>3
0.14
0.99
1.84
1.99
1.56

% HH

*Income in Rs./ month

24
22
20

% of households

18
Vehicular
ownership

16
14

12
1

10

>3

4
2
0
10000

20000

30000

40000

>45000

Income range (per month)

Chart 5.19 Auto ownership sub model

60

This model relates trips made by each HH and independent variables. From HIS
information regarding no. of trips, vehicle ownership and income for each HH is available.
From these data, a matrix is established based upon income range and vehicle ownership.
Table 5.8 summarizes the data from stratification of samples. The number in each cell
shows average trip rate of HHs in each category of vehicular ownership and income group.
Now smooth curves are plotted for average trip rate against income groups for various
vehicular ownerships. These curves may be extended depending upon the shape of the
curves. Chart 5.22 illustrates the trip rate curves developed from the above data. To find trip
rate per day for any household, first select income range of that hh. Now select vehicular
ownership. From a chart given below, draw a horizontal line on Y axis from the intersection
of line of income and particular vehicle ownership curves. That will give you trip rate for a
hh with given monthly income and no. of owned vehicles. Sample trip rate for hhs having
monthly income 20000 and varying vehicular ownership, is shown in chart below.
Table 5.8 Trip production model
Income group*
5 15 k
16 25 k
26 35 k
36 45 k
>45 k

0
2.72
3.86
5
-

Owned vehicles
1
2
3.24 4.33
4.38 4.72
5.07 5.43
6.67 5.48
6.25

>2
4
5.62
6.08
6.73
7.34

Avg.
daily
trip rate
*Income in Rs./ month

Average trip rate per day

7
Vehicular
ownership

0
5

1
2

>2
3
2
10000

20000

30000

40000

>45000

Income range (Rs./month)

Chart 5.20 Trip production model

61

6. DEVELOPMENT OF MODELS
For the present study, models are developed based on the data collected, through
HIS, using comprehensive transportation planning package TransCAD. TransCAD is a fullfeatured geographic information system designed specifically for planning, managing and
analyzing the characteristics of transportation systems and facilities. It provides all the tools
one needs to create and edit maps and geographic data sets, produce thematic maps and
other graphic output, and perform spatial and geographical analyses. It also includes a core
set of transportation network analysis, travel demand modeling, and operations research
models, a set of advanced analytical models for specific applications, and a set of
supporting tools for statistical and econometric analysis. These procedures can be used
individually or in combination to solve problems.
6.1

TRIP GENERATION MODEL

6.1.1 Regression approach

In a regression analysis we study the relationship, called the regression function,
between one variable y, called the dependent variable, and several others xi, called the
independent variables. Regression function also involves a set of unknown parameters bi. If
a regression function is linear in the parameters we term it a linear regression model.
Otherwise, the model is called non-linear. Linear regression models with more than one
independent variable are referred to as multiple linear models. General formula for MLR is:

Y* = C + aixi, Where, i = 1, 2, , n
Where,
Y* = Dependent variable (experimental value),
C = constant (intercept),
a = coefficient of independent variable, &
x = Independent variable.
6.1.2 Model structure
Structure selected for the model is:

Y = C + a1x1 + a2x2 + a3x3 + a4x4 + a5x5

Where,
Y = Total No. of trips generated / HH / Day (Observed value),
C = constant, a1, a2, a3, a4, a5 = model parameters,
x1 = HH Size, x2 = HH monthly income,
x3 = No. of working members/ HH,
x4 = No. of School/college going members/ HH, and
x5 = No. of Vehicles/ HH.
62

x1, x2, x3, x4, x5 are inputs here. Regression method will estimate values of Y, C and
model parameters.
6.1.3 Inputs
The household size, family income, vehicle ownership, working member and
school/college going member have very strong relation with peoples trip making behavior
besides number of other factors, but for the present study these five attributes of a dwelling
unit have considered to predict peoples trip making characteristics. Here total data points
are 705 and underlying assumptions are strong correlation and linear relationships exist
between dependent and independent variables.
Family Size (X1): Generally with higher family size the trip generation rate is observed to
be increasing due to more number of potential travelers. However sometimes trip rate may
decrease with higher family size depending on the age structure of the family members.
Senior citizens and kids generally do not contribute much to the trip generation.
Family income (X2): It is obvious when income increases the affordability of person to
make trips for varied purposes also rises. Particularly the recreation, shopping, social and
recreational purpose trips increases with the rise in income.
No. of working members (X3): The number of working members in a family makes direct,
additive & major influence on the trip generation rate of a family. The data for current study
very well reflects this phenomenon.
No. of school/ college going members (learners) (X4): Very obviously, number of
school/college going members directly affects the total trips generated by a particular
dwelling unit.
Vehicular ownership (X5): As the convenience of travel is better provided by the personal
vehicle, it is commonly observed that higher the number of vehicles owned by a family,
more will be the trips generated for purposes other than compulsive purposes of work and
education and vice versa.
6.1.4 Correlation analysis
The correlation coefficient measures the degree of association between two or more
variables. In the two-variable case, the simple linear correlation coefficient, R, for a set of
sample observations is given by -1 <= R <= 1. R < 0 refers to a perfect negative correlation;
however, R > 1 refers to perfect positive correlation. A zero correlation coefficient means
that there exists no linear relationship whatsoever between X and Y. Correlation analysis
refers simply to the type and degree of association between two variables. Table below
shows correlation coefficients between Y and input parameters. From the table values we
can conclude that positive correlation (although not strong) exists between dependent
variable and independent variables. Except x2 and x5, no strong correlation exists between
input parameters which is a prime requirement for a good MLR equation.
63

Table 6.1 Correlation coefficients between parameters

Parameters
Y
X1
X2
X3
X4
X5

Y
0.48
0.55
0.34
0.56
0.49

X1
0.48
0.30
0.34
0.50
0.27

X2
0.55
0.30
0.19
0.28
0.70

X3
0.34
0.34
0.19
-0.12
0.22

X4
0.56
0.50
0.28
-0.12
0.19

X5
0.49
0.27
0.70
0.22
0.19
-

6.1.5 Building a regression model in TransCAD

Flow chart below shows step by step how to develop a linear regression model in
TransCAD. For more information regarding regression in this software see Annexure D:
Working with TransCAD. Given a set of values for the dependent and independent
variables, the software uses a mathematical method (the estimation method) to determine
the parameters that yield a model that best matches the values found in sample. The least
squares estimation procedure also produces measures that tell you how well the model fits
the sample data and how important each variable is in the model. Together, the regression
parameters and goodness-of-fit measures help making statements about the relationships
present in the data and make predictions of the dependent variable based on other sets of
values for the independent variables. To estimate a trip production regression model, you
provide the dataset to use to estimate the model, specify the dependent variable, and specify
all of the independent variables. Fig. 6.2 shows how to select dependent and independent
variables. TransCAD then estimates the coefficients of the model and creates two output
files: A report file that lists the estimated parameters and goodness-of-fit and importance
measures and a file containing the values of the estimated model parameters.

Prepare
database in MS
Excel/ Access

Import into
TransCAD

Save dataview
file

Statistics -->
Model
estimation

Select
dependent
variables

Select
independent
variables

Select
regression

Save model file

View results in
Browser

Figure 6.1 Developing a regression model in TransCAD

64

Figure 6.2 Choosing dependent and independent variables

6.1.6 Output
Output for 705 samples is shown below.
Procedure Linear Model Estimation with Regression Model on May 15, 2012 (08:11 PM)
Model of '[Observed Trips/HH (Y)]' for final regression data (A Regression estimation of
survey data
Inputs
Field Name

Mean

Std Dev

[HH size (X1)]

4.49362

1.27933

[HH income/month (k) (X2)]

26.28369

9.08164

[Working members (X3)]

1.39291

0.72708

[School/college going members (X4)]

1.30071

0.75777

[No# of owned vehicles (x5)]

1.93333

0.96961

[Observed Trips/HH (Y)]

5.23546

1.75037

65

Dependent field

[Observed Trips/HH (Y)]

Outputs

Standard error

1.0132

Source

df

SS

Mean SS

F Ratio

Model

1439.29842

287.860

280.393

Error

699

717.61506

1.02663

Total

704

2156.91348

R Squared

0.6673

Adjusted R Squared

0.6649

Estimation Results
Field Name

Estimate

Std. Error

T Test

CONSTANT

0.282883

[HH size (X1)]

0.133425

0.0394668

3.38069

[HH income/month (k) (X2)]

0.078979

0.00565074

13.9767

[Working members (X3)]

0.612475

0.0609774

10.0443

[School/college going members (X4)]

0.866420

0.0652720

13.2740

[No# of owned vehicles (x5)]

0.153660

0.0506411

3.03430

Execution Time was 00:00:00.030

Inserted
from
<file://C:\Users\Admin\AppData\Roaming\Caliper\TransCAD
r4\Report.xml#IDAFIZZB>

5.0

----------------------------66

6.1.7 Model evaluation

Trip generation model equation is as follows. This equation predicts trip rate per HH
per day.
Y = 0.282883 + 0.133425 x1 + 0.078979 x2 + 0.612475 x3 +
0.866420 x4 + 0.153660 x5 (R2 = 0.67)
Here, R2 (coefficient of determination) measures the percentage of variation in the
dependent variable that can be explained by the linear relationship between x and y. That is,
how accurate the linear regression model is at predicting the trip rate of a dwelling unit
based on its characteristics. R2 value is 0.67 which is comparatively low but acceptable for a
travel forecasting model for comprehensive transportation planning. The value of multiple
R is 0.82, which is a correlation coefficient. It measures how well the data clusters around
regression line. The closer the value to 1, the more linear the data is. Here, the value of R is
near to 1, which indicates that relationship between input variables and output is very much
linear. Here, df stands for degree of freedom (df = n-1 = 704) and SS stands for sum of
square. Standard error value is 1.0132. It shows the amount of error in the prediction of Y
(calculated) for an individual sample. Lower the value of standard error stronger the
equation is. F ratio is used to compare two population variances. The T- test is a test for
equality of the population means that underlie each sample. It is obtained by dividing
estimated value of parameter by its standard error. High T-statistics (over 2) means the
variables are having significant effects on a dependent variable. All the factors have
addictive effect on trip rate. From the equation, it can be concluded that school/college
going members have highest effects on trip rate. As described above, model for work trips,
education trips and for other purpose trips can be developed. These models are as below.
Work purpose trip equation:
Y = 0.221266 + 0.015999 x1 + 0.000747 x2 + 0.863750 x3 0.008038 x4 + 0.065601 x5 (R2 = 0.53)
Educational purpose trip equation:
Y = 0.099116 - 0.000736 x1 + 0.004114 x2 - 0.069892 x3 +
1.064492 x4 + 0.061400 x5 (R2 = 0.82)
Other purpose trip equation:
Y = 0.642661 - 0.001846 x1 + 0.031906 x2 - 0.347368 x3 +
0.098838 x4 + 0.174325 x5 (R2 = 0.52)
Fig. 6.3 shows plot of observed trip rate versus estimated trip rate for 705 samples.
From the plot we get equation Y = 0.8626 x + 0.5379 (R2 = 0.56). Fig. 6.4 represents same
thing at a ward level (i.e. average values at ward level). Where Y = 0.6856 x + 1.6452 (R2 =
0.6). In both the cases we get poor coefficient of determination.
67

y = 0.8626x + 0.5379
R = 0.5649

12

10

Observed trips / HH / day

0
0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

Computed trips / HH / day

Figure 6.3 Plot of observed Vs. expected trip rates

68

6.50
y = 0.6856x + 1.6452
R = 0.597

Average Observed trips / HH / day

6.00

5.50

5.00

4.50

4.00

3.50
3.50

3.80

4.10

4.40

4.70

5.00

5.30

5.60

5.90

6.20

6.50

6.80

Average Computed trips / HH / day

Figure 6.4 Plot of ward level avg. observed Vs. computed trip rates
6.2

TRIP PRODUCTIONS

6.2.1 Ward wise trip productions

Trip rate for each ward is estimated by taking average values of input parameters at
ward level. Total trips produced by each ward can be estimated by multiplying trip rate to
the no. of HHs in each ward (obtained by avg. HH size and ward population). These data
are summarized in table 6.1.
69

Table 6.2 Ward wise daily total trip production

Ward
no.

Calculated
trip rate
/day

Trips produced
/day

Ward
no.

Calculated
trip rate
/day

Trips produced
/day

4.92

69527

25

5.54

81439

4.80

11432

26

5.56

68533

5.27

110494

27

5.41

75058

5.21

74874

28

4.71

45174

4.82

60070

29

5.44

232205

4.98

60889

30

4.69

142457

5.62

162826

31

5.91

39752

5.29

83858

32

5.85

7992

5.18

389127

33

5.53

182808

10

5.42

82204

34

4.51

31990

11

6.19

293568

35

6.01

43112

12

5.28

451

36

5.08

59676

13

5.66

79610

37

5.67

410087

14

4.99

349149

38

5.01

291151

15

5.87

81603

39

5.13

160612

16

5.14

148303

40

4.74

11640

17

4.83

76718

41

5.13

117440

18

4.85

152294

42

5.79

48620

19

5.44

357746

43

5.38

8926

20

5.48

132682

44

4.90

16938

21

4.93

110083

45

4.39

25422

22

4.73

195125

46

5.10

75736

23

4.86

67052

47

5.09

26598

24

5.73

30129

Total

5303024

70

6.2.2 Zonal trip productions

Taking values at zonal level and proceeding as described above zonal trip rate and
total daily trips produced from each zone can be estimated. These data are shown in table
6.2.
Table 6.3 Zone wise total daily trip production
Zone no.

Calculated avg. trip

rate /day /hh

Trips produced
/day

5.00

732192

5.23

980389

5.54

551113

5.20

249174

5.27

344350

5.35

1508570

4.91

1186969

Total

5552758

6.2.3 City level trip productions

By taking avg. values of all input parameters of all the data points obtained trip rate
is 5.23. Taking avg. HH size 4.49, no. of HHs obtained are 1070353. Therefore total daily
trips produced in the Surat city are 5593331. (Work purpose trips = 1789866 no., Education
purpose trips = 1677999 no. and other purpose trips (Shopping, Recreational, Social and
other trips) = 2125466 no.)
6.3

TRIP DISTRIBUTION MODEL

These models identify, where produced trips go. For the present study, readily
available data from previous surveys (2011) has been utilized since surveys for trip
attraction were not made.
6.3.1 Gravity approach
We already discussed gravity model in chapter 2. It is synthetic model based on the
gravity concept propounded by Newton. Accordingly, the interchange of trips between
zones in an area is considered to be direct to the relative attraction between the zones and
the inversely proportional to the spatial separation between them as measured by an
appropriate function of distance popularly known as travel impedance.

71

6.3.2 Model structure

For the present study the model form adopted is as follows.

Where,
Tij = Trips produced in zone i and attracted to zone j,
Pi = Trips produced in zone i,
Aj = Trips attracted in zone j,
Fij = Impedance between zones i and j, and
Kij = Adjustment factor for trips produced in i and attracted to j.
6.3.3 Inputs
From the above discussion, we can see that inputs required here are total no. of
produced and attracted trips from each zone and an impedance matrix. Here impedance
factor is taken in form of zone to zone centroidal distance. For the sack of simplicity
distance taken is aerial distance rather than actual road distance. Tables below show PA
matrix and friction factor matrix respectively.
Table 6.4 Zone wise trip productions and attractions
ZONE NO.

ZONE NAME

PRODUCTIONS ATTRACTIONS

Central

382638

466033

North

376038

374570

East

1035126

949905

South

454571

439084

South west

229104

575532

West

388817

234552

South east

335417

162035

Table 6.5 Friction factor matrix

ORIGIN
ZONE
1
2
3
4
5
6
7

1
7.5
8
9
11.7
8
7.8

2
7.5
6.5
15.5
17
6
8

DESTINATION ZONE
3
4
5
8
9
11.7
6.5
15.5
17
10
18
10
7
18
7
9.5
16
13.6
7.7
6
10

6
8
6
9.5
16
13.6
14.2

7
7.8
8
7.7
6
10
14.2
-

Distance
(km)

72

6.3.4 Developing gravity model in TransCAD

Flow chart below explains how to build a gravity model in TransCAD and get flows
in a matrix form. Steps are self explanatory.

Prepare OD matrix in
MS Excel

Import into
TransCAD

Save dataview file

Prepare friction
factor matrix in MS
Excel

Convert friction
factor dataview into
matrix

Planning ->
Balancing

Select appropriate
balancing method

Planning -> Trip

distribution ->
Gravity application

Select purpose,
method and no. of
iterations

Save model file

View flows in a new

matrix file

View report in
browser

Figure 6.5 Procedure to build gravity model in TransCAD

73

6.3.5 Output
TransCAD generates output report in HTML format and creates a new matrix file
showing output.
Procedure Gravity Application on July 02, 2012 (09:29 PM)
Inputs
View

PA DATAVIEW

Number of
Purposes

Purpose

Purpose 1

Production Field

PRODUCTIONS

Attraction Field

ATTRACTIONS

Constraint

Singly to Productions (Rows)

Friction Factor
Type

Gamma = a.pow(t,-b).exp(-c.t)
a = 28507.000000
b = 0.020000
c = 0.123000

Outputs
Output matrix file C:\Users\Admin\Desktop\cgrav.mtx
Results

No models have failed.

Purpose 1

Succeeded.
K-Factor not applied

--------------------------There are a variety of alternative functional forms for calculating impedances.

Popular choices are the exponential and inverse power functions typically used in entropy
models, and the gamma function recommended in US planning practice. K-Factors are used
to adjust the flows predicted by the gravity model. K matrix is an extra matrix that is
multiplied with the friction factors to adjust and fine-tune the friction factors. Suggested
values for parameters for different purposes are as below.
Trip Purpose

HBW
HBO
NHB

28507
139173
219113

0.020
1.285
1.332

0.123
-0.094
0.010

74

Output in TransCAD, exported in MS Excel, is shown below. Zone to zone matrix is

shown in table 6.6. Here, model was constrained to production and no. of iterations selected
was 50. Method employed was weighted sum.
Row ID's
1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
4
4
4
4
4
4
5
5
5
5
5
5
6
6
6
6
6
6
7
7
7
7
7
7

Col ID's
2
3
4
5
6
7
1
3
4
5
6
7
1
2
4
5
6
7
1
2
3
5
6
7
1
2
3
4
6
7
1
2
3
4
5
7
1
2
3
4
5
6

Purpose 1
63097
141116
57544
58741
34844
27295
71213
186887
24701
25253
46146
21836
252454
296235
185156
89662
112089
99530
82897
31529
149097
131587
17436
42025
50074
19074
42724
77865
19640
19727
75579
88686
135902
26252
49974
12423
57107
40479
116399
61034
48416
11983

75

Table 6.5 shows output in a matrix format and fig. 6.6 shows desire line diagram. It
represents zone to zone daily traffic flows.
Table 6.6 Zonal traffic flows
ORIGIN
ZONE

DESTINATION ZONE
1

63097

141116

57544

58741

34844

27295

71213

186887

24701

25253

46146

21836

252454

296235

185156

89662

112089

99530

82897

31529

149097

131587

17436

42025

50074

19074

42724

77865

19640

19727

75579

88686

135902

26252

49974

12423

57107

40479

116399

61034

48416

11983

Trips
per
day

76

Figure 6.6 Desire line diagram showing zone to zone traffic flows
77

7. CONCLUSIONS
In this dissertation attempt has been made to model travel demand of the people of
the Surat city and to understand their travel behavior. Secondary data were obtained through
SMC. Comprehensive home interview surveys were carried out for primary data. Travel
behavior comparisons were made for the year 2001, 2011 and 2012. Trip generation and
distribution models were developed for better understanding of the travel demand.
Modeling has been done through transportation planning software package TransCAD. Due
to time constraints sample size has been kept limited to 705 (15 samples from each election
ward) samples.
1.

2.

3.

4.

From HIS, at city level, it has found that:

o Average household size is 4.49. Ward wise it varies from 3.70 to 5.40 and zone
wise 4.30 to 4.70.
o Average monthly HH income is about 26284 Rs. Ward wise it varies from 22000
to 33000 and zone wise 23000 to 28000 Rs./ month.
o Working members per HH are 1.39.
o Learners per HH are 1.30.
o Average vehicle ownership per HH is 1.93.
o Average trip rate for work trips per HH per day is 1.64.
o Average trip rate for education trips per HH per day is 1.62.
o Average trip rate for other trips (e.g. trips for shopping, social purpose,
recreation and for other purposes) per HH per day is 1.97.
o Average trip rate per HH per day is 5.24. Ward wise trip rate varies from 4.20 to
6 and zone wise 4.70 to 5.40.
Purpose wise trip behavior analysis suggests that 33% of daily trip productions are
for work purpose, 30% are for education purpose and 37% are for other purpose
trips. This scenario is not in accordance with generality. In general trip proportions
should be 40-50% work trips, 30-40% education trips and remaining other trips.
Correlation analysis suggested that positive, moderately strong correlation exists
between dependent and independent variables. Correlation coefficients for observed
trips to the input parameters are 0.48, 0.5, 0.34, 0.56 and 0.49 respectively. Strong
correlation (R = 0.70) exists between input parameters household income and
vehicle ownership.
Aggregate trip generation model for the city (for the year 2012) is:
Y = 0.282883 + 0.133425 x1 + 0.078979 x2 + 0.612475 x3 +
0.866420 x4 + 0.153660 x5 (R2 = 0.67)

5.

R2 (coefficient of determination) for the equation is 0.67, which is comparatively

low but acceptable due to the large scope of the present study. All the input
parameters have additive effect on trip rate. School/ college going members have the
highest effect on trip generation.

78

6.
7.

8.

9.

T statistic values for all the independent variables are greater than 2, which means
all the input parameters have significant effect on dependent variable.
Droppings of any of the inputs are not improving R2 value in any combination.
Therefore step wise regression is not working in this case. The R2 for observed vs.
estimated daily trip rate for whole sample is 0.56, while the value is 0.59 for avg.
values at ward level.
From the above equation it has found that daily 5593331 trips have been produced in
the city. East zone shows highest trip rate of 5.54 total trips per HH per day while,
West zone produces highest daily trips of about 1500000.
Disaggregate models for trip generations are as below.
Work purpose trip equation:
Y = 0.221266 + 0.015999 x1 + 0.000747 x2 + 0.863750 x3 0.008038 x4 + 0.065601 x5 (R2 = 0.53)
Educational purpose trip equation:
Y = 0.099116 - 0.000736 x1 + 0.004114 x2 - 0.069892 x3 +
1.064492 x4 + 0.061400 x5 (R2 = 0.82)
Other purpose trip equation:
Y = 0.642661 - 0.001846 x1 + 0.031906 x2 - 0.347368 x3 +
0.098838 x4 + 0.174325 x5 (R2 = 0.52)

10.

11.

12.

From the above equations, the proportions of work, education and other purpose
trips are 32, 30 and 38 percentages respectively at city level per day. Therefore out
of 5593331 total trips per day, around 1789866 trips are work trips, 1677999 are
education trips and 2125466 are other purpose trips.
From fitting HH characteristics onto trip production curves, trip rate for a particular
HH can be estimated. Depending upon trends of the curves, they may be
extrapolated.
Trip distribution model at zonal level has been developed for the year 2011 through
gravity model (constrained to productions). The results are shown in desire line
diagram. Highest flow is between zones 3 (East) to 2 (North) with a value of 298235
trips per day. The least flow we get is in between zone 7 (South East) to 6 (West)
with a value of 11983. Here friction factor taken is aerial distance between zone
centroids because of convenience.

79

8. FUTURE SCOPE AND RECOMMENDATIONS

The present study focuses on understanding and estimating travel behavior characteristics of
the people of Surat city. The above developed models can be used for the estimation of trip
generation and distribution for the future years. These models are aggregate in nature. In a
same manner disaggregate models can also be developed. To ensure model predicts future
travel behavior as accurate as possible, traffic volume count surveys can be done to validate
and calibrate the above developed model, since the regression model makes assumptions
about underlying relationships in the data. In gravity model, travel time taken or actual road
distance or travel cost as a friction factor yields better results rather than taking aerial
distance. By carrying out OD surveys gravity model should be validated and calibrated. Trip
distribution can be made more accurate by considering through, external to internal and vice
versa trips. Similarly disaggregate gravity models can also be developed. It will give better
picture of travel behavior of the people. By carrying out detailed HISs, further, modal split
and traffic assignment models can be developed at ward, zone, city and regional level using
this software. By increasing sample size, efficiency of models can be increased.

80

9.

REFERENCES
Books

Bernstein D. (M.I.T.), Chapter-58: Transportation Planning, The Civil Engineering

Handbook (2nd ed.), CRC press, 2003.
Hensher D.A., Button K.J., Handbook of Transport Modeling, Pergamon, Amsterdam,
2000.
Kadiyali L.R.,Section-13: Transport Planning, Traffic Engineering and Transport
Planning (5th reprint), Khanna publishers, 2010.
Khisty C.J., Lall B.K., Urban Transportation Planning, Pg. 485-549, Transportation
Engineering - An Introduction (3rd ed.), PHI, 2008
Mathew T.V. and Rao K.V.K., Chapter-5: Travel demand modeling, Introduction to
Transportation Engineering, NPTEL May 3, 2007.
Mcdaniel D.E., Transportation Forecasting: A review, Technological Forecasting and
Social change, Pg. 367-389., 1972.
Mcnally M.G., Chapter-3: The Four Step Model, Handbook of Transportation Modeling
(2nd ed.), Pergamon, 2007.
Ortuzar J. DeD., Willumsen L.G., Modeling Transport (3rd ed.), Wiley, Chichester, U.K.,
2001.
Papacostas

C.S.,

Prevedouros

P.D.,

Chapter-8:

Travel

Demand

Forecasting,

Transportation Engineering & Planning (3rd ed.), PHI, 2001.

Journal papers
Aggarwal V., Varma A., Trip-end Models for the City of Thane A Comparative Study,
Indian Journal of Transport Management, pp. 96-110, 2010.
Datta P., Urbanization in India-abstract, Regional and Sub-Regional Population Dynamic
Process in Urban Areas European Population Conference, June 21-24, 2006.
Golob T. F., A simultaneous model of household activity participation and trip chain
generation, Transportation research part B 34, Pergamon, Pg. 355-376, 2000.
Kawano M., Yoshitake T., Chishaki T., An approach to trip generation analysis
introducing a concept of zonal trip production, Proceedings of the Eastern Asia
Society for Transportation Studies, Vol.4, Pg. 844-859, October, 2003.
Liya Y., Hai Y., Trip Generation Model Based on Destination Attractiveness, Tsinghua
Science and Technology, Vol. 13(5), Pg. 632-635, Oct-2008.

81

Modi K.B., Zala L.B., Umrigar F.S., Desai T.A., Transportation planning models: A
review, Paper presented at National Conference on Recent Trends in Engineering
& Technology, BVM Engineering college, V.Vidyanagar, 13-14 May, 2011.
Ravin M Tailor, G.J.Joshi, B.K.Katti, Trip Rate Intensity Characterisitics in Metropolitan
Context: A Case Study of Surat, National conference on Urban Transport and the
Environment, 2005
Ravinder K., Velmurugan S., Nataraju J., Gangopadhyay S., Development of Quick
Response Techniques (QRT) for travel demand estimation of small and medium
sized cities in India, proceeding of conference on Urban Mobility 2008 by IUT
(India), Pg. 18-29, 3-5 December, 2008.
Tamin O. Z., Sulistyorini R., Public transport demand estimation by calibrating the
combined Trip Distribution - Mode Choice (TDMC) model from passenger counts,
World Academy of Science, Engineering and Technology, Vol. 54, 2009.
Wong S.C., Tong C.O., Wong K.I., Lam W.H.K., Lo H.K., Yang H., Lo H.P., Estimation
of multiclass Origin-Destination matrices from traffic counts, Journal of Urban
planning and development, ASCE, Pg. 19-29, March-2005.
Dissertations
Mwakalonge J. L., Econometric modeling of total urban travel demand using data
collected in single and repeated cross-sectional surveys, Ph.D. dissertation,
Tennessee technological university, 2010.
Siddhpuria S. R., A Study on Trip Rates and Trip Predictions of Households in newly
Extended Area of Surat City, M. Tech. dissertation, S.V.N.I.T., 2008.
Vaghani A.,

Travel behavior Analysis in Changing Land use Pattern of Surat city,

M.Tech. dissertation, S.V.N.I.T., 2011.

Other sources
A guide to building a travel demand model in TransCAD by Rhett Fussell & Lita
Huntsinger, Model Research & Development Unit, NC Department of
Transportation Statewide Planning Branch
Appendix IV: Modeling, Dillon consulting limited, Kingston Transportation Master Plan Final Report, July 2004.
Chapter-6: Development of travel demand model, Comprehensive mobility plan for Pune
city by PMC in technical collaboration with Wilbur Smith associates and IL&FS.
Draft final report, Comprehensive Traffic and Transportation Study by SMC & prepared by
CRRI, 2006.
82

Model Validation and Reasonableness Checking Manual, Travel Model Improvement

Program, U.S. Department of Transportation, Federal Highway Administration,
June, 2001.
Multiple linear regression analysis notes by Michael L. Orlov, Chemistry department,
Oregon state University, 1996.
Selected TransCAD notes of Murtaza Haider, Ted Rogers school of management, Ryeson
University.
Surat city development plan (revised) (2008-13) by SMC and SUDA in technical
collaboration with CEPT University.
Travel demand modeling with TRANSCAD User manual 5.0 (Caliper Corporation)
Wine Country Travel Demand Model - Model Development Report by Parsons
Brinckerhoff, February, 2011.
Web resources
www.suratmunicipal.gov.in
www.censusindia.gov
www.people.hoftra.edu
www.ocw.mit.edu

83

ANNEXURE A
Break up of study area

Zone
no.

Zone
name

Election
ward no.

4
1

Census
ward
no.

Census ward
Name

Population

Nanpura

58329

30

Nanpura

9506

Sagrampura

9693

Salabatpura

59731

Begampura

46476

Saiyadpura

62329

Gopipura

20917

Wadifaliya

8995

10

Sonifaliya

23275

Haripura

12650

11

Nanavat

15025

12

Shahpor

27781

29

Rampura
laldarwaja

11380

31

Katargam
gotalawadi

70395

38

Tunki

53456

39

Singanpor

35972

67835

9693

106207

62329

CENTRAL
5

Election
ward
population

53187

55456

135231

NORTH

65484

33

174115

Dabholi

22428

41

Ved

7085

42

Katargam

365335

70

Chhapara Bhatha

52075

71

Kosad

98162

73

Utran

23878

34

31703

72

Amroli

31703

10

68716

32

Ashvanikumar
Navagam

68716

11

215045

43

Fulpada

215045

28

Nanavarachha
Water Works

193

44

Kapadra

97

45

Nanavarachha

98

36

Umarwada

66548

46

Karanj

229110

47

Umarwada (Part)

347

48

Magob (Part)

64561

74

Motavarachha

36325

12

365335

40

387

EAST
13

14

35

66548

294019

36325

36

SOUTH

Sarthana

28605

76

Simada

26176

54781

37

385856

77

Puna

385856

15

64882

34

Majura
Khatodara

64882

19

276202

56

Udhana

276202

20

116317

57

Bamroli (Part)

116317

53

Bhedvad

14088

55

Pandesar

81089

66

Vadod

99

54

Bhestan

58911

87

Vadod

45700

88

Bamroli (Part)

81529

85

Budiya

2712

86

Jiyav

6623

82

Unn

68485

83

Sonari

5628

84

Gabheni

19067

21

75

23

39

40

41

95276

58911

127230

9335

93180

24

25

37

Majura

26982

58

Majura

13893

59

Bhatar

50738

64631

26

59933

60

Althan

59933

27

60147

61

Umara

60147

13

Athwa

8122

33

AthwaUmara

34037

89

Bhimrad

2690

90

Bharthana
Vesu

6020

94

Vesu

29343

91

Sarsana

1090

92

Khajod

1930

93

Abhava

3616

95

Rundh

4847

96

Magdalla

6793

97

Gaviyar

2877

99

Dumas

8041

100

Sultanabad

4072

101

Bhimpor

8748

28

42

26982

42159

38053

SOUTH
WEST

43

44

45

6635

14516

20861

29

219070

27

Adajan

219070

30

127520

14-26

RANDER

127520

31

30946

63

Jahangirabad

30946

64

Jahangirpura

2408

65

Pisad

4515

67

Pal

40184

68

Palanpor

26144

69

Variyav

21952

35

Anjana

116898

50

Anjana (Part)

8044

32
6

WEST
46

47

16

6923

SOUTH
EAST

66328

21952

124942

17

66655

49

Dumbhal

66655

18

133973

51

Limbayat

133973

22

167712

52

Dindoli (Part)

167712

78

Magob

23157

79

Parvat

60779

80

Godadara

123576

81

Dindoli

83094

38

290606

ANNEXURE B
Home interview survey format

P. G. Center in Transportation Engineering and Planning

Department of Civil Engineering
S. V. National Institute of Technology, Surat -395007
Name of Surveyor: ___________________________________
Zone Name: ______________________ Zone No: __________

Date: _______________
Time: ____________

Ward Name: ____________________ Ward no.: ______ Sample: ________________

Home Interview Survey

(You can help Surat prepare for its future transportation needs!)
Part 1: Socio Economic Characteristics
House hold Characteristics
1

Other

Monthly H.H.
Income

<5 k

5-15 k

15-25 k

25-35 k

>35 k

Employee
Status (1,2,3)

Business

Govt./ Pvt.
job

Student

Self Employed

Worker
/H.W.

No. of owned
Vehicles

2W

3W

Car

Cycle

Other

House hold size

Part 2 Trip Characteristics

Trip Characteristics
Trip Purposes

Work

School/
college

Recreational

Shopping

Social

Trip Destination
Trip Duration (min)
Mode of Travel
Trip Length (km)
No of Trips
(day/week)

All your information will be kept strictly confidential. It will be combined with that from other respondents
for analysis and your name will not be associated with your responses.

ANNEXURE C
Household and travel characteristics data collected from HIS

Ward name

Election
Ward
no.

Sample
no.

Observed
Trips/HH
(Y)

HH
size
(X1)

HH income/
month (k) (Rs.)
(X2)

Working
members
(X3)

School/college
going members
(X4)

Owned
vehicles
(x5)

Work
trips/day

School/
college
trips/day

Other
trips/day

Rander

25

15

25

15

21

30

27

33

35

10

18

11

15

12

19

13

10

50

14

15

15

20

20

17

30

45

11

60

33

20

Kosad

Varachcha

Kapadara

25

10

10

25

11

15

12

13

13

20

14

18

15

22

27

30

25

40

25

20

25

15

15

10

25

11

35

12

30

13

28

14

16

15

25

30

35

21

Fulpada

15

22

20

30

25

17

10

25

11

30

12

33

13

25

14

43

15

40

20

20

35

15

20

40

26

30

20

10

15

11

20

12

20

13

36

14

31

Ashvanikumar

Katargam

15

30

15

12

20

25

35

25

30

29

28

10

20

11

19

12

25

13

32

14

18

15

20

25

33

34

24

27

25

35

20

21

10

23

Vadinath
Chowk

Ved Road

11

25

12

35

13

29

14

25

15

40

23

25

23

25

35

75

25

20

28

10

29

11

30

12

23

13

19

14

35

15

28

25

25

30

26

32

Rander Gorat
road

10

23

25

19

10

10

42

11

25

12

19

13

24

14

33

15

16

35

25

29

34

40

50

41

19

23

10

17

11

28

12

18

13

29

14

22

15

25

Adajan west

Adajan east

11

12

40

35

50

30

45

35

30

25

33

10

30

11

20

12

30

13

55

14

31

15

21

16

25

20

30

20

28

25

25

40

10

37

11

29

Mugalisara

Kubernagar

13

14

12

22

13

40

14

27

15

25

70

30

35

22

15

25

20

17

30

10

26

11

29

12

26

13

31

14

20

15

29

35

11

34

45

25

30

20

Mahidharpura

Trikamnagar

15

16

20

17

10

19

11

25

12

23

13

30

14

25

15

23

23

17

23

29

45

33

32

24

27

10

24

11

33

12

35

13

45

14

35

15

30

24

27

25

Labheshwar
chowk

17

15

21

25

29

15

40

10

16

11

22

12

18

13

28

14

32

15

30

32

22

27

30

50

15

10

27

19

10

24

11

23

12

30

13

28

Karanj

Puna

18

19

14

10

15

13

27

30

20

33

26

30

25

27

16

10

20

11

27

12

30

13

15

14

18

15

20

18

26

35

35

33

30

24

27

25

Magob

Navapura
Haripura

20

21

10

15

11

21

12

27

13

31

14

40

15

26

50

20

43

20

25

30

28

14

20

10

25

11

29

12

20

13

25

14

35

15

18

20

19

15

11

Sonifalia
Gopipura

22

21

30

45

23

10

10

25

11

20

12

20

13

25

14

10

40

15

26

15

18

20

30

35

25

23

25

10

22

11

27

12

23

13

14

14

31

Athwa
Nanpura

Sagrampura

23

24

15

26

25

33

27

21

35

20

15

15

20

10

34

11

25

12

17

13

27

14

20

15

15

35

36

26

20

30

20

25

40

26

Salabatpura

Anjana

25

26

10

25

11

24

12

30

13

27

14

15

15

20

12

23

26

32

18

22

35

30

35

10

22

11

27

12

35

13

42

14

22

15

30

20

30

27

23

55

Dumbhal

Parvat

27

28

30

19

30

35

10

25

11

22

12

23

13

29

14

24

15

55

26

30

28

10

45

10

28

24

18

35

10

20

11

30

12

29

13

22

14

40

15

16

20

Godadra

Limbayat

29

10

13

15

20

20

14

30

25

10

36

11

20

12

29

13

25

14

25

15

16

30

21

25

25

20

15

20

31

30

10

15

11

35

Udhna (north)

Majura
Khatodara

30

31

12

37

13

21

14

32

15

21

26

25

29

10

17

19

23

10

18

10

24

11

16

12

20

13

30

14

35

15

20

30

25

31

22

10

38

27

Ichhanath
Dumas road

Althan Bhatar

32

33

35

15

35

10

40

11

30

12

45

13

30

14

25

15

22

45

32

43

12

40

16

26

30

29

10

30

11

25

12

25

13

30

14

27

15

42

21

Bamroli

34

31

33

22

20

25

25

22

40

10

16

11

25

12

30

13

20

14

20

15

22

36

20

20

20

23

22

28

30

10

10

14

11

17

12

20

Udhna (south)

Pandesara

35

36

13

24

14

14

15

19

25

35

11

50

30

26

20

30

20

30

10

33

11

33

12

20

13

35

14

40

15

40

24

28

25

16

21

19

35

31

Dindoli

Bhestan

37

38

23

10

17

11

21

12

19

13

33

14

23

15

27

23

21

18

20

30

25

20

15

22

10

30

11

32

12

35

13

22

14

25

15

40

22

12

25

32

39

11

15

14

10

30

30

10

17

11

20

12

16

13

20

14

27

15

20

40

25

20

13

15

35

45

20

22

10

35

11

30

12

30

13

38

14

18

15

27

40

41

20

12

28

11

30

23

20

23

10

10

12

11

37

12

10

40

13

14

14

55

15

20

24

30

30

26

34

24

40

25

10

25

11

24

42

43

12

20

13

45

14

30

15

29

33

11

50

21

25

12

45

28

40

31

10

36

11

28

12

33

13

35

14

35

15

20

30

55

45

30

17

30

28

44

45

15

37

10

30

11

23

12

22

13

17

14

32

15

50

14

24

30

25

19

31

30

35

42

10

30

11

10

12

40

13

15

14

20

15

21

30

27

41

46

15

30

20

15

34

25

10

33

11

31

12

20

13

29

14

15

12

27

10

40

14

25

20

12

30

10

25

10

26

11

30

12

32

13

11

55

14

30

47

15

25

14

48

25

30

33

39

28

40

45

10

11

11

27

12

15

13

13

14

18

15

30

ANNEXURE D
Working with TransCAD

Dataview creation

Multiple linear regression process

Results summary

Gravity application