CHAPTER 3

Transportation Modeling
 2022/23
School of Civil and Environmental Engineering 2nd semester

CHAPTER 3

Transportation Modeling

3.1 Transport modeling

Modeling principles

Models are a simplified representation of a part of reality. Their function is to give insight
into complex interrelationships in the real world and to enable statements about what
(most probably) will happen if changes occur or put in that (part of) reality.
Models are invaluable in offering a common ground for discussing policy and examining
the inevitable compromises required in practice with a minimum of objectivity. During
their formulation, calibration and use, planners can also learn much about the behavior and
internal workings of the system under scrutiny.

However, a model is only realistic from a particular perspective. Its appropriateness is

highly dependent on the context where it will be used i.e. its value is limited to a range of
problems under specific conditions. Therefore, extreme care should be taken when
choosing and adapting models for a particular context.

Transport models study of the behavior of individuals in making decisions regarding the
provision and use of transport. Models in transportation planning are abstract
mathematical models, put into the form of (systems of) mathematical equations in which
the behavior of a dependent variable Y (e.g. the number of daily bus passengers in the
Addis Ababa) can be derived from one or more explaining or independent variables X (e.g.
number of transit lines, bus fares, etc) and related parameters a. Generally, they take the
form;

Y = f (a, X)

Where, the parameters describe the sensitivity of Y to a unit change in X.

Model development starts from the formation of a hypothesis to explain the given
phenomenon or system from a particular point of view. These sets of hypotheses will form
the theory. The translation of such a theory into a quantitative model with quantifiable
variables and associated parameters we call a conceptual model. The Estimation/
Calibration process determines the numerical values of the associated parameters from the
given data. The model then needs further testing and verification on another set of data
that has not been used during the calibration process. It is when the model is finally
verified and approved that it can be applied to the transport system. The process of the
model development is shown in the figure below.

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 1
 2022/23
School of Civil and Environmental Engineering 2nd semester

Figure 2.3 Model formulation process

The mathematical models have several purposes for transport planning, some of which are:
 To gain a modern structural analysis of the complex transport system
 To find out which factors play an important role, and how sensitive the transport
system is to changes in the different factors
 To analyze the effect of alternative traffic projects and contribute towards their
economic appraisal
 To help transport planners make reliable predictions and forecasts of future
changes in usage of traffic facilities for sake of facility design, control and operation.
 To enable quantified calculations of expected effects in the transportation system
when changes (policy measures or interventions) are put in the system
 To find design parameters that lead to an optimal performance of the modeled
system

Thus, transport modeling contributes greatly to improved decision making and planning in
the transport field. However, transport modeling is only one element in transport planning:
administrative practices, an institutional framework, skilled professionals and good level of
communication with decision makers, the media and the public are some of the other
requisites for an effective planning system.

(For more detail on the basics of modeling, please refer O’Flaherty 1997 chapter 5)

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 2
 2022/23
School of Civil and Environmental Engineering 2nd semester

Prerequisite for transport modeling

Before embarking on the modeling of transport systems in the context of this chapter, the
modeler should be familiar with the basic terminologies, grasp the fundamental
characteristics of transport problems, gather the necessary data and understand basic
regression analysis. These four prerequisites are treated in more detail below:

Characteristics of transport problems

Both the demand for travel and the supply for transport facilities exhibit distinct
characteristics that the modeler should bear in mind. These characters include:

 The demand for transport services is highly qualitative and differentiated. There is a
whole range of specific demands for transport which are differentiated by time of
day, day of week, journey purpose, type of cargo, and so on.
 The demand for transport is derived; it is not an end by itself. People travel in order
to satisfy a need (work, leisure, health) at their destination.
 Transport demand takes place over space. It is the distribution of activities over
space that makes for transport demand.
 Both transport demand and supply have very strong dynamic elements. A good deal
of the demand for transport is concentrated on a few hours of the day.
 Transport is a service and not a good. Therefore, it is not possible to stock it.
 The transport system requires fixed assets and the mobile units. It is often the case
that the infrastructure and the vehicles are not owned nor operated by the same
entity.
 Transport infrastructure is lumpy; one cannot provide half a runway or one-third of
a railway station. They take a long time to carry out and significant expenses.
 Transport investment has an important political role.
 Transport services come with side effects: accidents, pollution and environmental
degradation.

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 3
 2022/23
School of Civil and Environmental Engineering 2nd semester

Basic terms and definitions

Activity An activity occurs at the end of every trip and reflects the trip purpose,
that is, what the trip maker did at the trip destination. Activities are
categorized according to schemes such as
subsistence/maintenance/discretionary or work/shop/social/etc.

Calibration Calibration is the fitting of a model to observed data, primarily by

adjusting model parameters. In travel forecasting, the terms estimation
and calibration often have been used interchangeably.

Centroid A defined point within a TAZ from which all trips are assumed to start or
end. It should be located to reflect the center of activity in a TAZ and not
necessarily the geographic center.

Centroid Abstract links that connect centroids to network nodes and represent
Connector general access from a TAZ to the formal transportation network.

Corridor A linear study area along one or more transportation facilities for which
estimates of travel demand and system performance are desired.

Cordon Line An abstract line defining the external boundary of a study area.

Estimation Estimation, part of the model specification process, is the formal

computation of the parameters for a hypothesized model, to maximize the
likelihood of fitting observed data.

External A "door" on the cordon line of a study area corresponding to major entry
Station and exit points for external trips through, into, and out of the study area.

Home-based A classification for trips that either begin and/or end at a trip maker's
home regardless of the origin and destination, the home location
is always the production for home-based trips; the other end is always
the attraction.

Home-based A trip with one end at home and the other end at a non-work location.
Other

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 4
 2022/23
School of Civil and Environmental Engineering 2nd semester

Home-based A trip with one end at work and the other end at home.
Work

Household A fundamental sampling unit for travel surveys, the household (HH) is a
behavioral unit of one or more individuals who share a dwelling unit,
resources (e.g., income, automobiles), and travel
responsibilities. Interaction among HH members influences travel and
activity behavior.

Impedance Impedance, a computed measure of the disincentive to travel due to

spatial separation, is a composite function of travel time (often split
by access, in-vehicle, and egress time), travel cost, and/or distance.

Incidence An efficient means of storing network topology relationships, including

Matrix node-link, link-path, and path-OD incidence matrices.

External Trip A trip with either its origin or its destination located outside of the study
area. The external trip end is assigned to an external station.

Internal Trip A trip with both its origin and its destination located inside of the study
area. If both trip ends are outside the study area, it is a through trip.

Inter-zonal A trip between two different analysis zones.

Trip

Intra-zonal A trip with both origin and destination in the same zone. In trip
Trip assignment, intra-zonal trips are not loaded on the network.

Land Use The primary activity for which a parcel of land is used (residential,
commercial, industrial, open space, undeveloped, etc.).

Link A fundamental element of a transportation network defined by a starting

and an ending node and having attributes such as length, travel time
and/or speed, and capacity.

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 5
 2022/23
School of Civil and Environmental Engineering 2nd semester

Mode A means of conveyance between origins and destinations, modes are

motorized (cars and other private vehicles, buses, rail transit) and non-
motorized (walking, bikes).

Network A graphical and/or mathematical representation of a region's

transportation infrastructure and services,
comprising links and nodes and their corresponding characteristics.

Node A point joining two or more links in a transportation network, and having
attributes such as spatial coordinates, turn penalties and prohibitions.
Non-Home- A classification for trips which neither begin nor end at a trip maker's
Based residence (home). The origin of a NHB trip is also the production;
the destination of a NHB trip is also the attraction.

OD Matrix A trip table in origin / destination format

Origin The location or zone where a trip begins

Parameter A constant describing a population property utilized in a model

development process.

Path A sequence of links and nodes connecting an origin to a destination in a

network. Used interchangeably with route.

Peak Hour For a transportation facility or network, the hour of the day during which
the maximum traffic volume occurs.

Person-Trip A single trip by a single person. The output of trip generation is measured
in person-trips. On the other hand, a vehicle-trip is a single trip by a
vehicle, regardless of the number of occupants in the vehicle.

Prediction A general term for the estimation of the value of a variable of interest in

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 6
 2022/23
School of Civil and Environmental Engineering 2nd semester

an unknown situation.

Production The location or zone responsible for a trip occurring but also used for
the produced trip itself: a zone is a production for N trip productions).
Home-based trips, by definition, have their production in the zone
containing the household, regardless of the origin and destination of the
trip. The productions for non-home-based trips must be allocated to the
NHB trip's origin zone.

Route A path through a network; a series of links and nodes connecting an origin
and a destination. Used interchangeably with path.

Study Area A defined region within which estimates of travel demand and system
performance are desired.

Traffic A defined zone for travel forecasting and traffic simulation studies,
Analysis Zone represented in the network by a centroid.

Transportation The Transportation System constitutes the networks (modes, links, nodes,
System etc.) connecting a region's Activity System.

Validation An independent test of a calibrated model's predictive capabilities using

data that was not used in estimating or calibrating the model. The
objective is to verify that the model can replicate observed system
performance.

(For exhaustive terminologies, visit [Link] courtesy

of Professor M.G. McNally)

3.1.1 Forecasting travel demand and its Approach

 Travel demand is expressed as the number of persons or vehicles per unit time that can be
expressed to travel on a given segment of transportation system under a set of given land use,
socio-economic, and environmental conditions.

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 7
 2022/23
School of Civil and Environmental Engineering 2nd semester

 Forecasts of travel demand are used to establish the loads on future or modified transportation
system alternatives.
 The methods of forecasting travel demand can range from simple extrapolation of observed
trends to a sophisticated computerized process involving extensive data gathering and
mathematical modeling.
 Travel demand forecasting is as much an art as it is a science because judgments are required
concerning the various parameters-that is, population growth, car ownership, and so forth-that
provide the basis for a travel forecast. The method used in forecasting demand will depend on
the availability of data and specific constraints on the project, such as availability of funds and
project schedules.

Demand forecasting approaches

There are two travel demand forecasting situations in transport planning:

i) Travel demand studies for urban areas

ii) Intercity travel demand forecasting
i) Travel demand studies for urban area

This approach requires extensive databases be prepared using home interview and/or roadside
interview surveys. The information gathered provides useful insight concerning the characteristics
of the trip maker, such as age, sex, income, auto ownership, and so forth, the land use at each end of
the trip, and the mode of travel. These and other data then gathered by zone and/or be used at
more disaggregate level to formulate relationship between variables and calibrate models.

ii) Intercity travel demand forecasting

Data that are used in this situation are generally aggregated to a greater extent than for urban
travel forecasting such as city population, average city income, and travel time or cost between city
pairs.

Travel demand forecasting in urban areas

Three major factors affecting the demand for urban travels are:

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 8
 2022/23
School of Civil and Environmental Engineering 2nd semester

a) Land use characteristics: - The amount of traffic generated specific area of land depends
on how the land is used. For example, shopping centers, residential complexes, industrial
areas, recreational places and office buildings produce different traffic patterns.
b) Socioeconomic characteristics: - Lifestyles and other social factors affect how people
decide to use their recourses for transportation.
c) The availability of transport facilities and services: - Travelers are sensitive to the level
of service provided by alternative transportation modes. When deciding to travel at all or
which mode to use they consider attributes such as travel time, travel cost, convenience,
comfort and safety.

Basic steps in Travel Demand Forecast

a) Defining the study area

Defining the study area and subdividing it into traffic zones is the primary task to be accomplished
prior to collecting and summarizing the data.

The selection of zones is based on

 Homogenous socio economic characteristics

 Minimum intra zone trips
 Physical, political and historical boundaries
 Zones should not be defined within other zones
 The zone should generate and attract approximately equal trips, households, population
or area.
b) Data collection

Data is an essential input to the effective planning and design of transport systems, either directly
by describing the current state of the system, or indirectly by allowing the calibration of models
which yield insights into the process at work in the system or help to predict how the system is
likely to perform in the future with or without policy intervention. Data collection provides

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 9
 2022/23
School of Civil and Environmental Engineering 2nd semester

information about the land use as well as socio economic factors and the transport system which
will serve as the basis for developing travel demand estimates.

There are various methods that are used for transport planning. Some of these methods are

i) Road use study

Road use studies determine the relative use of various parts of a highway system. Information may
be obtained by personal interviews with representative sample of the registered motor vehicle
operators. Drivers are asked to state the total mileage and the proportion of that mileage driven on
a particular street or highway type.

ii) Transport facility and traffic studies

Transport facility and traffic studies involve in inventory of transport facilities and traffic studies.
Such studies are only necessary where there is no well-established data base containing
information on transport facilities and traffic. It involves the determination of spot speed, traffic
volume, travel time and delay, parking facilities, parking habits etc...

iii) Travel survey

Travel survey identifies

 Where and when trips begin and end (Origin and destination survey/O-D Survey)
 The trip purpose
 The mode of travel
 Socio economic characteristics of the trip makers
 The type of land use
 Automobile occupancy
 For freight transport, the type of commodity transported etc
Travel surveys serve as a fundamental source of data for transport studies. Travel survey enables
transport planners to obtain a sample of all travel data on a typical time within a specified area. By
expanding the data it is possible to produce an estimate of the average travel demand on transport
system at the time the survey was conducted.

There are four general classification of travel survey:

1. House hold travel surveys

Transport Planning and Modeling
Lecture Note: Chapter 3 Page 10
 2022/23
School of Civil and Environmental Engineering 2nd semester

2. Road side surveys

3. Modal surveys
4. Good measurement surveys
1. House hold travel surveys

House hold travel survey has long been a measure source of information on the characteristics,
behavior and attitude of travelers.

House hold travel surveys can be used in determination of

 The number and characteristics of person trips or auto drive trips made by residents
with specified study area.
 Socio economic characteristics of the house hold.
This can be accomplished by

 Home interviews – more preferable than other survey methods for its high response rates
but are more expensive.
 Telephone interviews and
 Mail surveys.
2. Road side surveys

This method consists of stopping vehicles and asking drivers for the information on trip origin, trip
destination, trip purposes etc. Vehicles passing the survey stations are counted but only sample of
drivers are interviewed.

Other alternative ways of doing road side surveys include

 Handing out voluntary return post card questionnaires to drivers.

 Recording license no. and sending questionnaires to the address of the owner.
Road side surveys are especially useful in recording relatively long and infrequent trips including
trips made by persons outside the survey area.

3. Modal Survey

Travel surveying of the number of trips made by each mode in transport system such as using
private car, bus, rail, air transport, walking etc. Typical modal survey is mostly undertaken by
performing an interview at passenger terminals.

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 11
 2022/23
School of Civil and Environmental Engineering 2nd semester

4. Good measurement Survey

It consists of interviewing drivers about origin-destination for trucks, load type and weight and
detail information on the transport commodities.

Transport System Modeling

The modification of highway network either by new road construction or operating improvements
on existing roads (e.g., use or retiming of traffic signalization) must be predicted upon some
expectation or forecast of traffic volumes. For new road construction, forecasts of traffic are needed
to determine the appropriate pavement and geometric design that will provide an acceptable level
of service. For operational improvements, traffic forecast are needed to estimate the effectiveness
of alternate improvement options.

In forecasting traffic volumes, two interrelated elements must be considered:

1. Overall regional traffic growth/decline

2. Traffic diversion

Specifying requirement for modeling

When selecting a model for a particular purpose the analyst should begin by clarifying the
requirement of the exercise and the resources available. The requirement should be expressed in
terms of:

 What input variables are required, and at what level of detail?

 What output variables are required, at what level of detail and in what form?

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 12
 2022/23
School of Civil and Environmental Engineering 2nd semester

 What response should be allowed for?

The challenge to transport planners is to make reliable forecast of traffic demand that reflect the
effect of changes in population, social and economical conditions as well as changes in transport
networks. Unless reliable forecasts of future traffic are made, there is a risk of building facilities that
will receive little use or prematurely overloaded.

The components of future demand includes

1) Existing traffic: the traffic that currently use the existing facility
2) Normal traffic growth: anticipated traffic in regional population or economy.
3) Diverted traffic: a traffic that is diverted to the improved road due to some factors.
4) Converted traffic: a traffic that results from change of mode due to convenience.
5) Change of destination: traffic that has changed to different destination where such changes
is attributable to the attractiveness of the road and not to changes in land use.
6) Development traffic: it is due to the development near by the road which causes change in
the transport system of the road in addition to the development that would have been taken
place had the new or improved transportation facility not constructed.
7) Induced traffic: traffic that did not previously exist in any form but result when new or
improved transport facilities are provided.
Basic Concept in transport planning

Transport models are based on some fundamental assumptions and approaches. The most
important ones include:

i. Trip making is a function of land use.

ii. Trips are made for different purposes.
iii. Trips are made at different times of the day.
iv. Travelers often have different options available for them (in terms of route and mode of
transport)
v. Trips are made to minimize the level of inconvenience associated with reaching a
destination.
vi. Transport network and traffic analysis zone are the basis for system modeling.
The use of transport models

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 13
 2022/23
School of Civil and Environmental Engineering 2nd semester

 Providing traffic data for the future; thus understanding the challenges to come;
congestion, environmental conditions, future transport costs for people and for
business/industry, accessibility and mobility, etc
 Analyzing the effect of alternative plans/projects providing traffic data for benefit
cost analysis.
 Providing traffic data for calculating the development of traffic accidents and
environmental conditions.
Organizing the transport model
The structure of transport model will be according to “the four step procedure”, but there is some
preparation to be done before the start of the modeling work.

 Zoning and land use data

 Transport network coding
 Data acquisition (inventories, travel survey)
Zoning and land use data

We have some basic principle of how a zone should be organized/located, but most important is the
availability of land use data for the zones.

o Population: when we start to look population into trip production models we discover that
we may wish to have population data by “household”, “age group”, “income group”, “sex”
and “car ownership” etc
o Jobs: the number of jobs in each zone, split in different type of job; the simplest
stratification is two groups of jobs
 Ordinary jobs which do not attract customers or visitors, the only trips made
are the journey to and from work.(and may be some freight transport)
 Jobs that attract customers(public attractive jobs)
Other land use influencing on trip production; shopping malls, cinemas, sport area, schools,
universities, etc.

Data requirements

Since transport models are applied to large systems, they require information about
travelers of the area influenced by the system. Here the data requirement is very high, and
can be broadly categorized into four:

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 14
 2022/23
School of Civil and Environmental Engineering 2nd semester

 Socio-economic data: Information regarding the socio-economic characteristics of

the study area. Important ones include income, vehicle ownership, family size, etc.
 Travel surveys: Origin-destination travel survey at households and traffic data
from cordon lines. Former data include the number of trips made by each member
of the household, the direction of travel, destination, the cost of the travel, etc. The
latter include the traffic flow, speed, and travel time measurements.
 Land use inventory: This includes data on the housing density at residential zones,
establishments at commercial and industrial zones.
 Network data: This includes data on the transport network and existing
inventories. Transport network data includes road network, traffic signals, junctions
etc. The service inventories include data on public and private transport networks.

The data required for modeling is primarily collected through surveys. These surveys
include:

 Household survey
 External cordon and Intercept surveys
 Travel Diary
 O-D survey
 Questionnaire
 In-house and Roadside Interviews

Designing the data collection survey for the transportation projects requires considerable
experience, skill, and a sound understanding of the study area. It is also important to know
the purpose of the study and details of the modeling approaches. Further, many practical
considerations like availability of time and money also has a strong bearing on the survey
design.
(For more on data collection and sampling, please refer chapter 3 of Ortuzar & Willumsen
2001)

Mathematical background

For this particular chapter, the modeling student needs to revise:

 Multiple regression analysis

 Elementary statistics

(For more on the basic mathematical requirements, please refer chapter 2 of Ortuzar and
Willumsen 2001)

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 15
 2022/23
School of Civil and Environmental Engineering 2nd semester

The Four step model

The most popular of the transport modeling approaches is the classic Four-Step Model
(FSM). The FSM aims to establish the spatial distribution of travel explicitly by means of an
appropriate system of zones. It is presented as a sequence of four mathematical sub
models:
1. Trip generation- forecasts the number of trips that will be made.
2. Trip distribution- determines where the trips will go.
3. Mode usage- predicts how the trips will be divided among the available modes of
travel.
4. Trip assignment- predicts the routes that the trips will take, resulting in traffic
forecasts for the highway system and rider-ship forecasts for the transit system.

In a nutshell, the FSM aims at explaining where the trips come from and where they go, and
what modes and which routes are used. The four sequential models result with the volume
of traffic on the road network, the level of service and travel attributes on each link, the
split between the private vehicles and the public transit, etc. They can also project the level
of traffic and the challenges to be faced in the future.

1. Trip Generation

Some basic Definition

 Home-Based (HB) Trip:- one where the home of the trip maker is either the origin
or the destination of the journey.
 Non-Home- Based Trip:- one where neither end of the trip is the home of the
traveler.
 Trip Production:- is the home end of an HB trip or the origin of an NHB trip.
 Trip Attraction:-is the non-home end of an HB trip or the destination of an NHB
trip.

Example 1

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 16
 2022/23
School of Civil and Environmental Engineering 2nd semester

Trip Generation is Define the magnitude of total daily travel in the model system, at
the household and zonal level, for various trip purposes (activities).
The objective of this first stage of the FSM process is to define the magnitude of total
daily travel in the model system, at the household and zonal level, for various trip
purposes (activities). This first stage also explicitly translates the FSM from activity-
based to trip-based, and simultaneously separates each trip into a production and an
attraction. It aims at predicting the total number of trips produced in the zone and
attracted by it respectively for each TAZ of the study area. It has two basic functions:
 To develop a relationship between trip production or attraction and land use,
and
 To use the relationship developed to estimate the number of trips generated
at some future date under a new set of land-use conditions.

Trip generation (both production and attraction) depends on the nature and
characteristics of the activity system. In production models, estimates are primarily
based on the demographics of the population within a zone. For attraction models, the
variables that have been found to have the best explanatory power are those based on
characteristics of the land use, such as office and retail space or the employment levels
of various sectors.
We get better trip production and attraction models (and thus better transport
models if different trips are modeled separately:

Compulsory or mandatory trips

 Home to work
 Home to school or college
Discretionary or optional trips

 Shopping trips
 Social or recreational trips
 Other trips

Some factors which have been found to have a considerable impact on the trip
producing capacity of a TAZ are:
 Income
 Car ownership
 Household structure
 Family size
 Value of land
 Residential density
 Accessibility

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 17
 2022/23
School of Civil and Environmental Engineering 2nd semester

While the following factors are widely used with the affinity of a zone to attract trips:
 Employment
 Sales
 Space available for industrial, commercial or other services
 School, Universities
 Entertainment…
Trips can be modeled at the zonal, household, or personal level, with household level
models most common for trip productions and zonal level models most common for
trip attractions. Furthermore, it have been found in practice that better trip generation
models can be obtained if trips by different purposes are identified and modeled
separately. In the case of home-based (HB) trips, five categories have been usually
employed:
 trips to work
 trips to school
 shopping trips
 social and recreational trips
 other trips

It is also important to classify trips into peak and off-peak periods as the proportion of
journeys vary greatly with the time of the day. It is also wise to differentiate trips into
personal and freight as the two types have significant difference in nature and
characteristics.

Several modeling approaches are available for trip generation including growth factor,
regression, discrete choice and category classification. In this course, only growth factor
modeling and regression analysis will be discussed.

Growth Factor Models

Growth factor model tries to predict the number of trips produced or attracted by a
house hold or a zone as a linear function of explanatory variables. The model has the
following basic equation:

Where is the number of future trips in the zone and is the number of current trips

in that zone and is the growth factor.

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 18
 2022/23
School of Civil and Environmental Engineering 2nd semester

The growth factor depends on the explanatory variable such as population (P) of the
zone, average house hold income (I), average vehicle ownership (V). The simplest form
of is represented as follows:

Where the subscript "d" denotes the design year and the subscript "c" denotes the
current year.

The growth factor method delivers a simple and easy to understand formulation.
However, the method has usually resulted in an over-estimated number of trips (Check
examples). Such an error at the early stage of the FSM will be carried down to the
subsequent stages and will grossly mislead planners and decision makers.

Therefore, the growth factor method is only used in practice to predict the future
number of external trips to an area. This is because they are not too many in the first
place (so errors cannot be too large) and also because there are no simple ways to
predict them.

Example 2

Regression analysis models

Regression methods can be used to establish a statistical relationship between the

number of trips produced and the characteristics of the individuals, the zone, and the
transportation network.

The most common form of trip generation model is a linear multiple regression function of the form:

Where, are explanatory variables such as income, car ownership, population etc. and

is generated trip. are parameters determined through calibration process.

Model parameters and variables vary from one study area to another and are
established by using base-year information. Once the equations are calibrated, they are
used to estimate future travel for a target year. In developing regression equations the
following is assumed:
Transport Planning and Modeling
Lecture Note: Chapter 3 Page 19
 2022/23
School of Civil and Environmental Engineering 2nd semester

1. All the independent variables are independent of each other.

2. All the independent variables are normally distributed.
3. The independent variables are continuous.

Two types of regression models are commonly used. The first uses data aggregated at
the zonal level, with average number of trips per household in the zone as the
dependent variable and average zonal characteristics as the independent (explanatory)
variable. The second uses disaggregated data at the household or individual level, with
the number of trips made by a household or individual as the dependent variable and
the household and personal characteristics as the independent variables.

The zonal-based regression is a premier method for modeling trip attractions where as
the household-based regression is primarily used with trip production.

(For more on trip generation models, refer chapter 4 of Ortuzar and Willumsen 2001 or
chapter 4 of Bovy [Link] 2006)

Example 3

2. Trip Distribution

The trip-generation analysis provides the planner with the numbers of trip productions
and trip attraction that each zone will have. But where do the attractions in the zones
come from and where do the productions go? What are the zone-to-zone travel
volumes?

Trip-distribution procedures determine where the trips produced in each zone will go-
how they will be divided among all other zones in the study area. The decision on where
the trips go is represented by comparing the relative attractiveness and accessibility of
all zones in the area.

The major product of trip distribution models is an O-D matrix that shows the number
of trips originated in the study zone and where these trips are destined to. This is a two
dimensional array of cells where rows and columns represent each of the zones in the
study area. is the number of trips between origin and destination . is the total

number of trips originating in zone and is the total number of trips attracted to

zone .

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 20
 2022/23
School of Civil and Environmental Engineering 2nd semester

Two basic categories of aggregate trip distribution methods predominate in urban

transportation planning:

 The Growth Factor methods- These involve scaling an existing matrix (called
base matrix) by applying multiplicative factors (often derived from predicted
productions and/or attractions) to matrix cells.
 The Gravity Model- This explicitly relates flows between zones to inter-zonal
impedance to travel. For gravity models, typical inputs include one or more
flow matrices, an impedance matrix reflecting the distance, time, or cost of
travel between zones, and estimates of future levels of productions and
attractions.

The Growth Factor Methods

In this approach a base year matrix is needed. Each cell of this matrix is multiplied by a
growth factor. Growth factors may be computed in a number of ways, e.g. as the output
of an economic model, a trend model, etc. However, in these course notes, we only
discuss methods of computing growth factors based on trip generation modeling.

The base year matrix contains an estimate of the trips being made in the base year.
Depending on these estimates, we may be able to use different growth-factor methods
in our estimation of future trip patterns.

Uniform Growth Factor

If the only information available is about a general growth rate for the whole of the
study area, then we can only assume that it will apply to each cell in the matrix, which is
a uniform growth rate. The equation can be written as:

Where is the uniform growth factor, is the previous total number of trips and is

the expected total number of trips.

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 21
 2022/23
School of Civil and Environmental Engineering 2nd semester

Advantages are that they are simple to understand, and they are useful for short-term
planning. Limitation is that the same growth factor is assumed for all zones, where as in
most cases differential growth for different parts of the study area is expected.

Example 4

Singly Constrained Growth-Factor

If information is available on the expected growth of either trips originating or trips

attracted to each zone, it will result in origin-specific and destination-specific
growth factors respectively. In this case, the expected number of trips between origin-
destination pairs is written as:

for origin-specific factors

for destination-specific factors

Example 5

Doubly Constrained Growth Factor

When information is available on the growth in the number of trips originating and
terminating in each zone, we know that there will be different growth rates for trips in
and out of each zone and consequently having two sets of growth factors for each zone.
This implies that there are two constraints for that model and such a model is called
doubly constrained growth factor model. Historically a number of iterative methods
have been proposed to obtain an estimated trip matrix which satisfies both sets of trip-
end constraints, or the two sets of growth factors.

The best known of these methods is due to Furness (1965), who introduced balancing
factors’ and as follows:

Or incorporating the growth rates into new variables and :

With and

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 22
 2022/23
School of Civil and Environmental Engineering 2nd semester

The factors and must be calculated so that the constraints are satisfied. The

procedure is:
1. Set =1

2. With = 1, solve for to satisfy trip generation constraint ( .

3. With solve for to satisfy trip attraction constraint ( .

4. Update matrix and check for errors.
5. Repeat steps 2 and 3 till convergence.

The advantages of Growth Factor method are:

1. Simple to understand.
2. Preserve observed trip pattern.
3. Useful in short term-planning.

The limitations are:

1. Depends heavily on the observed trip pattern.
2. It cannot explain unobserved trips.
3. Do not consider changes in travel cost.
4. Not suitable for policy studies like introduction of a mode.

Example 6

The Gravity model

The gravity model derives its base from Newton’s law of gravity, which states that the
attractive force between any two bodies is directly related to their masses and inversely
related to the distance between them. Similarly, in the gravity model, the number of
trips between two zones is directly related to activities in the two zones, and inversely
related to the separation between the zones as a function of the generalized cost. A
more general term used to represent the generalized cost (for the separation between
zones) is impedance or deterrence function.

In its simplest formulation, the model has the following functional form:

Where is the proportionality factor and

is a generalized function of the travel costs with one or more parameters for

calibration.
Transport Planning and Modeling
Lecture Note: Chapter 3 Page 23
 2022/23
School of Civil and Environmental Engineering 2nd semester

The need to satisfy the constraints ( requires replacing the

single proportionality factor by two sets of balancing factors and as in the

Furness model, yielding:

In the case of the doubly constrained model, the values of the balancing factors are:

The calculation of and , thus, requires an iterative process analogous to Furness’s

as shown in the Growth Factor modeling.

The deterrence function is the essence of the gravity model. It takes several forms

based on the decision of the modeler such as (the last one being empirical and more
common):

Example 7-1

Other approach

The Gravity model is a synthetic model as opposed to the growth factor model. It estimates
trips for each cell of the O-D matrix without directly using observed trip. This is the most
widely used trip distribution model which states that the number of trips between two
zones is directly proportional to the number of trip attraction generated by the zone of

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 24
 2022/23
School of Civil and Environmental Engineering 2nd semester

destination and inversely proportional to a function of time of travel between the two
zones.
Mathematically, the gravity model is expressed as

Where

Tij = number of trips that are produced in zone i and attracted to zone j

Pi = the total number of trips produced in zone i

Aj = number of trips attracted to zone j

Fij = a value which has an inverse function of travel time

Kij = socioeconomic adjustment factor for interchange ij

The values of Pi and Aj have been determined in the trip generation process. The sum of the
trip production for all zones must be equal to the sum of trip attraction for all zones.
Pattern i.e. it doesn’t require a base year matrix to forecast future trip patterns.
Note that the sum of productions in each zones is equal the number of production in the
statement. However, the number of attractions estimated in the trip distribution phase
differs from the number of attractions given. And it considered a singly constrained
gravity model.
Hence, calculate the adjusted attraction factor according to the following formula

Where
Ajk = adjusted attraction factor for attraction zone (column) j, iteration k
Ajk = Aj when k=1
Cjk = actual attraction (column) total for zone j, iteration k
Aj = desired attraction total for attraction zone (column) j
j = attraction zone number, j=1, 2, 3…..n
n = number of zones
k = iteration number, k=1, 2, 3….n

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 25
 2022/23
School of Civil and Environmental Engineering 2nd semester

m= number of iteration
To produce a mathematically correct result, repeat the trip distribution computation using
the modified attraction values, so that the number of attracted will be increased or reduced
as required. Therefore, the new attraction factors are adjusted downward by multiplying
the original attraction value by the ratio of the original to estimated attraction values.
Example 7-2

3. Modal Choice

In this phase of travel-demand forecasting, we analyze people’s decisions regarding

mode of travel; auto, bus, train, and so on. Before we can predict how travel will be split
among the modes available to the travelers, we must analyze the factors that affect the
choices that people make. Three broad categories of factors are considered in mode
usage:

1. The characteristics of the trip maker (e.g. family income, number of autos
available, family size, residential density)
2. The characteristics of the trip (e.g. trip distance, time of day)
3. The characteristics of the transportation system (e.g. riding time, excess time)

Mode usage analysis can be done at various points in the forecasting process. Mode
usage analyses are sometimes done within trip-generation analyses. However, the most
common point is after trip distribution, because the information on where trips are
going allows the mode usage relationship to compare the alternative transportation
services competing for users. Mode choice models can also be done on both aggregate
(Zonal) and disaggregate (Household or individual) levels. In this course, we will
concentrate on aggregate post-distribution models.

The most common of these aggregate post-distribution models is the family of the logit
models (binary logit, multinomial logit, nested logit etc.). A logit model is choice model
that assumes an individual maximizes utility in choosing between available
alternatives. The logit model's utility function comprises a deterministic component
(which is a function of measurable characteristics of the individual and of the
alternatives in the individual's choice set) and a stochastic component (or error term)
assumed to have an extreme value distribution

The functional form of the logit model for k number of alternative modes is:

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 26
 2022/23
School of Civil and Environmental Engineering 2nd semester

Where, is the proportion of trips travelling from to via mode 1. is the

generalized cost of mode 1 and is a calibrated parameter. The logit function results in
an S-shaped curve.

Figure 2.4- S-shaped logit mode choice curve

Some time instead of the generalized cost we use the utility of different mode in this
case the proportion for one mode is given by:-

Where P(A) = the probability of choosing mode A

Ui = utility of any mode i
UA = utility of mode A
Example 8 and 9
4. Trip Assignment

Traffic assignment is the step in traffic analysis in which inter-zonal trips are assigned
to the network. The traffic demand, as described in the origin-destination (OD) tables
per trip purpose and per travel mode (and sometimes per period), is confronted with
the infrastructure supply, which is a network of links and nodes having characteristics
as capacity, maximum travel speed, one-way streets, tolls and other factors of
resistance.

Traffic assignment involves computing one or more optimal (usually shortest) routes
between each origin and destination and distributing travel demand over these routes.
The sum of all trips along these routes over all OD pairs results in a traffic load on all
links and nodes. Usually, there is a separate assignment for each mode, since the
Transport Planning and Modeling
Lecture Note: Chapter 3 Page 27
 2022/23
School of Civil and Environmental Engineering 2nd semester

networks for each of the modes is very different. For the sake of simplicity, we restrict
ourselves to assignments of individual road traffic (car, bike); the more complex
assignments on public transportation networks will not be discussed here.

The major aims of traffic assignment procedures are:

 To estimate the volume of traffic on the links of the network and obtain
aggregate network measures.
 To analyze the travel pattern of each origin to destination (O-D) pair.
 To identify congested links and to collect traffic data useful for the design of
future junctions

Necessary input for the assignment:

 an O-D table of trips between the zones, usually all trip purposes combined;
 a (computer)representation of the network;
 characteristics of the network elements (links and nodes);
 a route choice model.

Direct output of the assignment computation:

 the routes (consecutive series of adjacent links and nodes);
 the route characteristics (travel times, distances, costs);
 route loads: the number of trips per route;
 link and node loads: the number of trips per unit time (flow) on each link and
each turn at junctions.
Technically, there are two broad assignment models: the minimum path assignment
and the congested assignment. The minimum path assignment models assume that the
capacity and travel cost of the links is unaffected by the volume of traffic and all the
traffic will choose to travel on the shortest path. Whereas, the congested assignment
models address the fact that the travel time and cost on a link increases as the volume
of traffic on the link increases.

The all-or-nothing (AON) assignment is the basic form of the minimum path assignment
models while incremental assignment, capacity restraint assignment, user equilibrium
assignment (UE), stochastic user equilibrium assignment (SUE), system optimum
assignment (SO), etc are some forms of the congested assignment models.

All-or-Nothing Assignment

In an All-Or-Nothing (AON) assignment, all traffic between an O-D pair is assigned to

just one path (usually the shortest path) connecting the origin and destination. This
model is unrealistic in that only one path between every O-D pair is utilized even if
there is another path with the same or nearly the same travel time. Also, traffic is

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 28
 2022/23
School of Civil and Environmental Engineering 2nd semester

assigned to links without consideration of whether or not there is adequate capacity or

heavy congestion; travel time is taken as a fixed input and does not vary depending on
the congestion on a link.

However, this model may be reasonable in sparse and uncongested networks where
there are few alternative routes and they have a large difference in travel cost. This
model may also be used to identify the desired path: the path which the drivers would
like to travel in the absence of congestion. In fact, this model's most important practical
application is that it acts as a building block for other types of assignment techniques. It
has a limitation that it ignores the fact that link travel time is a function of link volume
and when there is congestion or that multiple paths are used to carry traffic.

One form of the AON is the shortest path all-or-nothing assignment. This is an
assignment in which for each OD pair the corresponding flow is assigned to a single
path that, according to a fixed set of link costs, has minimum path costs (congestion
effects are not taken into account).

Finding the minimum path in the transportation network is an optimization problem.

Several numerical formulations such as Moore’s and Dijksta’s are available to solve this
minimum path problem. But this is outside the scope of this course.

Example 10

User Equilibrium Assignment

The user equilibrium assignment is based on Wardrop's first principle, which states
that:

Under equilibrium conditions traffic arranges itself in congested networks in such a

way that no individual trip maker can reduce his path costs by switching routes.

This means, in the congested network, all the used routes between an O-D pair have
equal and minimum costs while all unused routes have greater or equal costs.

The user equilibrium assignment assumes that:

 The user has perfect knowledge of the path cost.

 Travel time on a given link is a function of the flow on that link only.
 Travel time functions are positive and increasing.
Example 11

System Optimum Assignment (SO)

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 29
 2022/23
School of Civil and Environmental Engineering 2nd semester

The system optimum assignment is based on Wardrop's second principle, which states
that:

Under social equilibrium conditions, traffic should be arranged in congested

networks in such a way that the average (or total) travel cost is minimized.
This assignment can be thought of as a model in which congestion is minimised when
drivers are told which routes to use. Obviously, this is not a behaviourally realistic
model, but it can be useful to transport planners and engineers, trying to manage the
traffic to minimise travel costs and therefore achieve an optimum social equilibrium.

In general the flows resulting from the two principles are not the same but one can only
expect, in practice, to arrange itself following an approximation to Wardrop’s first
principle, i.e. selfish or users’ equilibrium.

The indicator is often used to measure how close a solution is to Wardrop’s equilibrium

Convergence criteria

Where: (Cijr – Cij*) excess cost between i & j relative to the minimum (Cij*)

Incremental Loading

 Parts of the matrix are assigned in each iteration. E.g.: 4 iterations: 40%, 30%, 20%
or 10%, alternatively 25% at each iteration

 Does not necessarily converge to Wardrop’s equilibrium!!

 Suffers from the limitation that once a flow has been assigned to a link it is not
removed and loaded onto another one

 In other words, if one of the initial iterations assigns too much flow on a link for
Wardrop’s equilibrium to be met, then the algorithm will not converge to the
correct solution.

Step for Incremental Loading

1. Select initial set of link costs (free flow/warm start)

 set all flows to 0

 partition trip matrix

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 30
 2022/23
School of Civil and Environmental Engineering 2nd semester

2. build set of minimum cost trees for all origins

3. load first (next) matrix segment using AON

4. calculate new set of link costs

5. Whole matrix assigned?

 if yes, END

 if no, go to (3)

Example 12
Successive Average

1. Select initial set of link costs (free flow/warm start)

 get set of link costs (most likely free flow)

 set all link volumes to 0,

 set iteration number (n = 0)

2. Set n = n + 1

3. Update the minimum cost trees

4. All-or-nothing assignment, Fan

5. Update the link flows

6. Update link costs given Van

7. Check for convergence; if no convergence then goes to step 2.

Example 13

(For detail explanation of traffic assignment models, please refer to Ortuzar and Willumsen
Chapter 10)

References

C.A. O’Flaherty [Link], Transport Planning and Traffic Engineering, Elsevier, 1997

C. J. Khisty and B.K. Lall, Transportation Engineering: An Introduction, 3rd Edition, Prentice
Hall of India, 2006

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 31
 2022/23
School of Civil and Environmental Engineering 2nd semester

J.D. Ortuzar & L.G. Willumsen, Modeling Transport, 3rd edition, Wiley, 2001

P.H.L Bovy, M.C.J Blemier & R. van Nes, Transportation Modeling: CT4801 Course Notes,
Delft University of Technology, Faculty of Civil Engineering, 2006

Transport Planning and Modeling

Lecture Note: Chapter 3 Page 32