Transportation, Land Use and Integration: Perspectives for Developing Countries 3

TRANSPORTATION MODELLING IN PRACTICE:

CONNECTING BASIC THEORY TO PRACTICE
WERNER HEYNS1 & SIMON VAN JAARSVELD2
1
Aurecon Pty Ltd, Pretoria, South Africa
2
Arup Pty Ltd, Johannesburg, South Africa

ABSTRACT
Transportation planning and modelling can appear to be a complex process when first encountered.
Transportation planning is a process that develops information to help make decisions on the future
development and management of transportation systems. It involves the determination of the need for
new or expanded highways, public transport systems, freight facilities, transportation terminals, their
location, their capacity and the management of their demand. Typically, transportation planning
involves a forecast of travel patterns and demand 15 to 25 years into the future with an aim to develop
a future transportation system that will support spatial growth at that time. Travel demand occurs as a
result of thousands of individual travellers making individual decisions on how, where and when to
travel. These decisions are affected by many factors such as family situations, characteristics of the
person making the trip, and the choices (destination, route and mode) available for the trip.
Transportation modelling is used to forecast travel patterns and demand and includes numerous
mathematical equations to simulate or represent how people travel.
This chapter explains how the process works, the assumptions made and the steps taken. It also
provides an insight into how transportation modelling is applied in practice in demonstrating its role in
the inter-relationship between spatial development and transport provision and planning.
Keywords: traffic, modelling, land use, transport integration, four step process, trip generation, trip
distribution, mode split, trip assignment, land use, spatial development.

1 INTRODUCTION
The implementation of transport infrastructure and services are derived from transportation
planning and modelling outputs. Transportation modelling and simulation as interface in the
planning process plays an important role in enabling spatial development.
The models developed by transport planning professionals are simplified representations
of reality which can be used to explore the consequences of particular policies, strategies and
spatial changes. The models forecast transport demand based on expected spatial
development, population and economic growth, which translates in the creation of additional
trips on the road network requiring expansion or the provision of infrastructure for a variety
of modes.
The travel forecasting process is at the heart of transportation planning (Goulias [1]). The
process is used to estimate the number of trips that will be made on a transportation systems
alternative at some future date. It involves a series of mathematical models that attempt to
simulate human behaviour while traveling (Tri-County Regional Planning Commission [2]).
The models include a sequence of steps that answer a series of questions about traveller
decisions. Attempts are made to simulate all choices that travellers make in response to a
given system of highways, public transport and policies. Many assumptions need to be made
about how people make decisions, the factors they consider and how they react to a particular
transportation alternative.
This chapter is broadly divided into two sections. The first 8 sub-sections provide basic
theory and an overview of transportation modelling processes in general, considering
the what, why, who and how of modelling and actual models used to forecast future travel.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
doi:10.2495/978-1-78466-233-2/001
4 Transportation, Land Use and Integration: Perspectives for Developing Countries

The second component of the chapter defines the practical application of modelling software,
how it relates land use changes to infrastructure provision, the type of problems they seek to
address, the decisions planning professionals have to make in deciding about the appropriate
model in terms of applicability, costs and data requirements. Fig. 1 illustrates the components
covered in this Chapter.

2 LAND-USE AND TRANSPORT INTEGRATION

The control and planning of land use is to a large extent the key to the control of both the
demand for transport and its impact upon the environment. Although the inter-relationship
between land use and transport is complex, it is well established. An inherent interaction or
feedback process takes place between land use and transport, where the one influences and
depends on the other. (American Association of State Highway and Transportation Officials
[3], Toth [4]). Wegener [5] refers to the interaction as the land use transport feedback cycle,
illustrated by Fig. 2.

3. Overview of the
2. Land‐use & 4. What is
steps involved in 6. Selecting a
transport transportation 5. Why is it used?
transportation model
integration modelling?
planning

8. Four‐step
7. Types of 9. Model 10. Transport
transport
transport calibration and modelling in
modelling
models validation practice
process

Figure 1: Components of transportation modelling examined.

Figure 2: Land use transport feedback cycle.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 5

Starting at the bottom of the cycle (Land Use), the locations of population and
employment determine the origins and destinations of most trips in travel models (Activities).
Modelling the transportation system allows calculating accessibilities, which describe how
well accessible a zone is to all other zones. Accessibility shapes land use, as both households
and businesses search for locations that are – among other location factors – accessible.
Because of this interaction, it is important for transport and land use planning professionals
to understand the likely inter-relationships and possible negative side effects when planning
is not executed within an appropriate framework. For example, a land use policy aimed at
encouraging urban sprawl, does little to promote shorter work-home trips, reduced travel cost
and the financial requirements of operating and funding public transport. Counter intuitive
and counterproductive land use and transport planning has serious long lasting consequences.
In South African cities, the legacy of Apartheid is a good example demonstrating the
transport inefficiencies resulting from inappropriate land policies, including:
 Misguided investment in transportation systems that prioritize high speed mobility over
local accessibility (This places all other modes of travel at a disadvantage);
 Urban sprawl;
 Under provision of Non-Motorised Transport Infrastructure;
 Commuter public transport services focused on operating mainly during the morning and
afternoon peak hours, with limited inter peak services;
 High travel cost (private and public transport) – mobility and accessibility is expensive
and has a high economic and social cost;
 Rural communities often isolated and inaccessible, contributing to the inability to access
opportunities and jobs resulting in perpetuated poverty.
Transportation modelling is a strong and helpful tool in the transport planner’s tool kit that
supports decisions making. It will not eradicate poor decision making or controversial land
use policies, but some major advantages of using models in transport planning in supporting
integration is that:
 it estimates likely land use transport interactions more quickly and at lower cost and risk
than would be through implementation and monitoring or trial by error;
 it helps planners optimise the balance between land use provision and appropriate
transport infrastructure and relevant modes to enable sustained economic growth;
 it defines the appropriate level of infrastructure required to support land use and spatial
development;
 it defines the appropriate modes of transport to connect land use;
 it weighs up benefits/dis-benefits of various development scenarios; and
 it informs investment decisions by providing input into financial, economic,
environmental and other relevant forms of appraisal.

3 OVERVIEW OF THE STEPS INVOLVED IN TRANSPORTATION PLANNING

Before exploring what transport modelling is, it is necessary to understand the broad steps
involved in transportation planning as to understand how transportation modelling fits into
the overall planning process.
Transportation planning goes through a basic sequence of steps. Several can take place
simultaneously and it is not unusual to repeat some of the steps (United States Department
of Transportation Federal Highway Authority of Transportation [6], [7]). Fig. 3 illustrates the
broad overall planning process.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
6 Transportation, Land Use and Integration: Perspectives for Developing Countries

Goals,
Problem Data Develop Implementa‐
objectives Forecasting Evaluation
definition collection alternatives tion
and criteria

Figure 3: Basic steps in the transportation planning process.

 Problem definition: What is the key transportation, economic and land use issues and
problems facing the community? This step may also involve definition of the size of an
area to be studied, determination of the scope of the study and the establishment of a
steering committee structure or forum to oversee the planning process.
 Define goals, objectives and criteria: A consensus should be developed by elected
officials and citizens about the future of the community and its transportation system.
Goals are developed for the quality of transportation service, environmental impacts and
costs and are typically in conflict. A good planning effort will identify the trade-offs
between these factors among alternatives in a clear, concise way to help make decisions.
 Data collection: Data must be compiled about the present status of the transportation
system and its use. This could include traffic data, public transport ridership statistics,
census information and interviews of households about their travel patterns. Data are
also gathered on land use, development trends, environmental factors, and financial
resources. This will help in problem definition and in developing methods to forecast
future travel patterns. Good data are essential to the planning process. The statement
‘garbage in/garbage out’ applies in transportation planning. Without good data, the
results of the planning process have little real meaning and can lead to wrong project
selection and an inappropriate developmental direction for the region.
 Forecasts: Data from existing travel is used to make forecasts of future travel using
travel demand models. This requires forecasts of future population, land use and
economic conditions as well as understanding of how people make travel choices.
Forecasting requires large amounts of data and is carried out under many assumptions
which is based on benchmarking, sensitivity checks and follow up surveys to identify
trends.
 Develop alternatives: Forecasts are used to determine the performance of alternative
future land use and transportation systems. Alternatives normally include different land
use and transportation patterns with different mixtures of highway and transit services
and facilities. Since land use affects travel and travel affects land use, both must be
considered.
 Evaluation: Results of forecasts are used to compare the performance of alternatives in
meeting goals, objectives and criteria. This information may be extensively discussed by
interested citizens, elected officials, different government agencies and the private
sector. Ultimately decisions are made by appropriate elected or appointed officials and
groups.
 Implementation of plans: Once decisions are made, plans should be further developed
and refined for implementation. This may include more detailed analysis for design and
evaluation following a similar process as above.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 7

4 WHAT IS TRANSPORTATION MODELLING?

Whilst many definitions exist defining transportation modelling it is primarily a mathematical
tool using computer software to represent an actual transport system (the real world) to
forecast travel patterns and flows between origins and destinations in geographic space by
different modes (Victoria Transport Policy Institute [8]). The model output informs decisions
about the type of infrastructure and transport modes to be implemented or operated in support
of changing land use and travel patterns.
For example, when a new road is to be built, a transport model is usually developed to see
how much traffic the road will carry, how wide it should be, where to include road
intersections and defining the appropriate type of intersection control etc. After all, if you are
able to save the cost of a slip road or the implementation of an additional lane on a highway
the cost of developing and using the model is only a fraction. Because the investment
requirement of transport infrastructure is very high and most often a burden on the fiscus
clear advice is required to government or developers to provide value for money solutions.
Transport models also sometimes cover public transport, walking and cycling and at the other
end of the scale they cover air, sea and freight transport. These models may also include fares
where the higher the fare the less people will use the infrastructure facility or mode of
transport. This gives rise to the notion of elasticity in demand (Litman [9]), which is a
relationship between a fare and the number of people who use a transport service or
infrastructure.

5 APPLICATION OF TRANSPORTATION MODELS

Transportation models informs decisions about the type of infrastructure and transport modes
to be implemented or operated in support of changing land use and travel patterns
(Chandrasekaran [10]). Modelling helps to gain insight into complex situations by
understanding simpler situations resembling them (Florian [11]). Professionals in transport
and land use development also apply transportation models for some of the following
reasons:
 Forecasting traffic flow;
 Assessing the outcomes of alternative land uses, infrastructure network and policy
alternatives:
 For smaller areas – models are good for considering alternative designs
(e.g. different sized roundabouts, intersection layouts etc.);
 For larger areas – to answer strategic questions such as whether or not to build new
roads or implement urban pricing;
 Corridor studies;
 Supporting development control policy testing;
 Land-use transport interaction;
 Understanding travel behaviour;
 Establishing how to reduce traffic congestion;
 Establishing how to reduce fuel consumption;
 Establishing how to improve air quality;
 Optimizing the use of resources in building or maintaining transport systems; and
 Providing a platform for testing competing ideas and use in negotiations and decision-
making.
Transportation models are used in a sequence of steps to answer a series of questions about
future travel patterns that supports forecasting demand and ultimately planning decisions.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
8 Transportation, Land Use and Integration: Perspectives for Developing Countries

The following questions follow a logical sequence and inform the steps in the modelling
process:
1. What will our community/city/region look like in the future?
a. How many people will live there? (population forecasts)
b. What will they be doing? (economic forecasts)
c. Where will activities take place? (land use)
2. What are the travel patterns in the future?
a. How many trips will be made? (trip generation)
b. Where will the trips be? (trip distribution)
c. What modes will be used? (mode split)
d. What routes will be used? (traffic assignment)
e. What will be the effects of this travel? (impact analysis)
Ultimately transportation models should help the decision maker save money, define the
appropriate level of transport investment required in support of spatial changes and
implement value for money transport projects, systems and operations. Most modelling work
is concerned with appraisal, in the broadest sense of the word, of alternative options or
designs.

6 SELECTING A MODEL
When a range of models is available, the choice is influenced by the level of complexity of
the problem being sought to be addressed. Models range in complexity from simple equations
representing relationships which can be calculated on the back of an envelope, on a calculator
in a spreadsheet or using powerful off the shelf (commercial) software programmes. The
latter involving a large volume of coding to perform sophisticated mathematical functions or
detailed simulations.
When selecting a model, the user should begin by defining the requirements of the
exercise to be undertaken and the resources required. The requirements that should be defined
include:
 What input variables are required, and to what level of detail?
 What output variables are required, at what level of detail and in what form (e.g. on
screen results or graphics and results to “come alive” where the movement of cars are
observed on screen)?
 Geographical coverage (will a single intersection be simulated, a number of intersections
or an entire suburb?)
 Constraints of time and money;
 Is it possible to buy new commercially available software or should new software or
processes be developed? and
 Determine whether new data can be collected.
The objective of defining the requirements for analysis to determine up front the most
appropriate model for the task at hand, to save money and time and derive the right answer.

7 TYPES OF TRANSPORTATION MODELS

A variety of types of models is available to the analyst. The different traffic models each have
its own pros and cons. The type of network under study and the traffic characteristics strictly
influence the choice of the traffic model used for analysis. There is no one single model that
can be used effectively for modelling traffic in all kinds of networks (Chandrasekaran [10]).
For simplicity, two main groups exist:

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 9

 Abstract mathematic models; and

 Simulation models.
The review provided below is not exhaustive and highlights the most common models within
each group.

7.1 Abstract mathematic models

A number of abstract mathematical models exist including the following:

 Simple formulae: An equation of the general form y = f(x) can be used to represent an
empirically observed relationship between one variable (x), which can be taken as given,
and another (y) which is to be predicted.
 Time series models: Many aspects of transport demand vary over time. For example,
car ownership and total vehicle- kilometres have been rising almost continually over the
past few decades, except for blips coinciding with fuel crises or economic recession. By
plotting the relevant data over time, it becomes possible to deduce and underlying trend
and or relationship. A prediction of future levels of travel demand can be made by
extrapolating the trend into the future. Trend based forecasting is very popular because
of its simplicity and because it is based on past evidence (Brockwell et al. [12]).
 Averaging and smoothing models: When no apparent trend data is available to support
demand forecasting, simple arithmetic can be used to calculate an average or an
exponentially smoothed prediction. Average forecasts are simply but very crude and may
give a distorted view, exponential smoothing seek to overcome the crudeness, by giving
weights to the most recent and trustworthy data.
 Regression analysis: This is the process of identifying the mathematical function
(relationship) which best fits the observed data. The most common applications are in
the prediction of car ownership, trip ends (numbers of origins and destinations at a given
location) and trip volumes (in an area between a given pair of zones or along a link)
(Jha et al. [13]). It is also used to predict waiting times at intersections, numbers of
accidents or levels of pollution.
 Matrix estimation models: The cells of a demand matrix (known as an origin –
destination matrix) indicate the number of trips between each origin and destination pair,
and the row and column totals indicate the total number of origins and destinations
respectively in each zone. The matrix provides the overall picture of travel demand in a
study area and is a key component in transport analyses.
 Elasticity models: When short term planning is undertaken, such that some influences
can be considered constant, it is possible to specify a model which is solely concerned
with the sensitivity of the forecast to key policy variables – known as elasticity modelling
The elasticity of demand with respect to a certain variable is defined as the rate of change
of demand with respect to that variable, normalized by the current levels of demand and
the variable in question (Meyer and Miller [14]). Elasticity is thus a measure of
sensitivity to change in system conditions (He and Zhao [15]). Its general form is:

Yt = Yt-1 {1+E(Xt – Xt-1)/Xt-1},

where, Yt = quantity demand in year t, Xt = value of a supply variable (e.g. price or

journey time) in year t, and E is the elasticity coefficient for Y with respect to X.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
10 Transportation, Land Use and Integration: Perspectives for Developing Countries

The elasticity coefficient should be calibrated on past data containing evidence of the
marginal effect on Y of a marginal change in X.
 Demand allocation – modelling the choice between alternatives: Trips are made as a
result of choices made by travellers; choices between alternative modes of travel,
alternative times of travel and alternative routes. Some of these choices are heavily
constrained (e.g. if the traveller has no car available or if a public transport timetable
provides only one service per day) while others may be made from a very extensive
choice set. Different groups in the population have different constraints and those who
have a very restricted choice set may be described as captives of a particular option.
Even if they are not captive to one option some groups may have a strong preference for
one option over the others; for example, high income travellers are likely to have a high
value of time which causes them to prefer a fast and expensive option over a slower and
cheaper one.
There is a wide and growing range of models which seek to predict travellers’ choices
among available options. They vary in sophistication from models which simply aim to
reproduce an observed pattern to others which seek to replicate the underlying choice
processes. An important concept underlying many of these models is market
segmentation whereby separate forecasts are made for subgroups within the total
population who, by virtue of their particular characteristics of car ownership, income,
journey purpose etc., may be expected to respond to the choices available differently
from other subgroups. Having made a forecast, for each subgroup, of the proportion of
group members selecting each option, the overall forecast is simply achieved by
weighting the proportions by the expected size of each subgroup in the forecast year’s
population.

7.2 Simulation models

Simulation models differ from abstract mathematical models in that they attempt to represent
the dynamic evolution of some aspect of the transport system through an explicit
representation of the behaviour of actors within it. Therefore, at a detailed scale, the
performance of an intersection might be simulated by representing individual cars passing
through it and the performance of a bus route might be simulated by representing the boarding
and alighting of individual passengers at each stop. At a less detailed level, the development
of a bus network might be predicted as the result of competition between different operating
companies.
Another major difference between simulation and abstract modelling is that simulation
modelling allows the analyst to develop a logic model of a very complex system by putting
together a number of components which are themselves fairly well understood. It therefore
becomes possible to develop a model of a complex system without having recourse to
overarching theories about system dynamics or concepts such as equilibrium. When
simulation is very detailed (zooming into fine detail analysis), it is sometimes referred to as
micro-simulation, but the term is sometimes used interchangeably. When applied in traffic
engineering the units are vehicles. When applied in travel behaviour the units are persons and
households (Goulias [1]).
With cheaper, readily available commercial software programmes and ease of use and
understanding opposed to abstract mathematical models, simulation modelling is
undoubtedly the fastest growing area in modelling and are more often applied in solving
transport and traffic problems.
Boxill and Yu [16] have classified simulation models into three categories:

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 11

 Macroscopic;
 Mesoscopic models; or
 Microscopic models;
Simulation models can also be classified according to the nature of the network that they
can be applied to, e.g. signalized networks, highway networks, and integrated networks or
for specific purposes (Variable road pricing, etc.). Tables 1 to 3 illustrate a range of
simulation modelling software programmes according to the classification by Boxill and Yu
[16].

Table 1: Macroscopic models.

Model Classification
ARTWORK Highway networks
AUTOS Highway networks
CORFLO Integrated Networks
FREFLO Highway networks
FREQ Highway networks
KRONOS Unknown
METACOR Unknown
METANET Highway networks
NETFLO Signalized networks
NETFLO 2 Signalized networks
NETVACI Unknown
PASSER-II Signalized networks
PASSER-IV Signalized networks
TRANSYT-7F Signalized networks
TRANSYT/10 Signalized networks
TEXAS Signalized networks
Source: Boxill and Yu [16].

Table 2: Mesoscopic models.

Model Classification
DYNAMIT Integrated Networks
DYNEMO Integrated Networks
DYNASMART Integrated Networks
Source: Boxill and Yu [16].

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
12 Transportation, Land Use and Integration: Perspectives for Developing Countries

Table 3: Microscopic models.

Model Classification Model Classification

AIMSUN Integrated Networks PLANSIM-T Highway networks
ANATOLL Tolls ROADSIM Rural road network
ARCADY Signalized networks SATURN Integrated Networks
AUTOBAHN Highway networks SHIVA Other
AVENUE Signalized networks SIGSIM Signalized networks
CARSIM Signalized networks SIMCO2 Highway networks
CASIMIR Signalized networks SIMDAC Highway networks
CONTRAM Integrated Networks SIMNET Highway networks
CORSIM Integrated Networks SIMTRAFFIC Signalized networks
DRACULA Integrated Networks SISTM Highway networks
FLEXYT II Signalized networks SITRA-B+ Integrated Networks
FOSIM Highway networks SIDRA Integrated Networks
FREEVU Highway networks SITRAS Integrated Networks
FRESIM Highway networks SMARTAHS Unknown
HIPERTRANS Signalized networks SMARTPATH Highway networks
HUTSIM Signalized networks SOUND Integrated Networks
ICARUS Integrated Networks SPEACS Highway networks
INTEGRATION Integrated Networks STEER Highway networks
INTRAS Signalized networks STREETSIM Signalized networks
JAM No Available Data TEXAS Signalized networks
MELROSE Integrated Networks TEXSIM Signalized networks
METROPOLIS Highway networks TRAFFICQ Signalized networks
MICROSIM Signalized networks TRANSIMS Highway networks
MICSTRAN Signalized networks TRGMSM LRT
MIMIC Safety TRITRAM Highway networks
MITSIM Integrated Networks THOREAU Integrated Networks
MIXIC Highway networks UTSS Integrated Networks
NEMIS Integrated Networks VEDENS Signalized networks
NETSIM Integrated Networks VISSIM Integrated Networks
OLSIM Highway networks VISSUM Integrated Networks
PADSIM Highway networks WATSIM Integrated Networks
PARAMICS Integrated Networks WEAVSIM Integrated Networks
PELOPS Integrated Networks
Source: Boxill and Yu [16].

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 13

8 FOUR-STEP TRANSPORT MODELLING PROCESS

Travel demand modelling was first developed in the late 1950s as a means to assist in
highway planning. As the need to look at other problems and issues arose, the transportation
modelling process was modified subsequently to add additional techniques to deal with these
problems.
As a starting point one can ask, how is the city or region represented for computer
analysis? Travel simulation require that an urban area or region be represented as a series of
small geographic areas referred to as travel analysis zones (TAZs). Fig. 4 illustrates the travel
analysis zones and links concept. Zones are characterized by their population, employment
and other factors and are the places where trip making decisions are made (trip producers)
and the trip need is met (trip attractors). Trip making is assumed to begin at the centre of
activity in a zone (zone centroid).
Trips that are very short, that begin and end in a single zone (intrazonal trips) are usually
not directly included in the forecasts. This limits the analysis of pedestrian and bicycle trips
in the process. Zones can be as small as a single block but typically are 1/4 to one mile square
in area. A planning study can easily use 500–2000 zones. A large number of zones will
increase the accuracy of the forecasts but require more data and computer processing time.
Zones tend to be small in areas of high population and larger in areas of low density and
population. Internal zones are those within the study area while external zones are those
outside of the study area. The study area should be large enough so that nearly all (over 90%)
of the trips begin and end within the study area. The trips that do not begin or end within the
study area but travel through it is considered to be through traffic.

TAZ

Node Link (e.g. road)

Centroid

Centroid connector

Figure 4: Travel analysis zones.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
14 Transportation, Land Use and Integration: Perspectives for Developing Countries

The highway system and transit systems are represented as networks for computer
analysis. Networks consist of links to represent segments of highways or public transport
lines and nodes to represent intersections and other points on the network. Data for links
includes travel times on the link, average speeds, capacity, and direction. Node data is more
limited to information on which links connect to the node and the location of the node
(coordinates). Node data could also include data on intersections to help calculate delay
encountered at intersections.
The travel simulation process follows trips as they begin at a trip generation zone, move
through a network of links and nodes and end at a trip attracting zone. The simulation process
is known as the four-step process and includes (i) trip generation, (ii) trip distribution,
(iii) mode split and (iv) traffic assignment (Tri-County Regional Planning Commission [2],
Victoria Transport Policy Institute [8], Florian [11]).

8.1 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 trips will be made to and
from each zone. This is carried out separately by trip purpose. Some of the trip purposes that
could be used are: 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 or end at home), truck trips and taxi trips. Trips are calculated based on the
characteristics of the zones. Trip productions are based on household characteristics such as
the number of people in the household and the number of cars available. For example,
a household with four people and two cars may be assumed to produce 3.00 work trips per
day. Trip attractions are typically based on the level of employment in a zone. For example,
a zone could be assumed to attract 1.32 home based work trips for every person employed in
that zone. Trip generation uses trip rates that are averages for large segments of the study
area. In South Africa, the following sources are used providing empirical estimates of trip
generation per trip purpose or land use type:
 South African Trip Generation Rates Manual [17];
 Technical Methods for Highways (TMH) 17, Trip Data Manual [18]; and
 Institute of Transport Engineers (ITE), Trip Generation Manual [19].

8.2 Trip distribution

Trip generation only finds the number of trips that begin or end at a particular zone. The
process of trip distribution links the trip ends to form an origin–destination pattern. Trip
distribution is used to represent the process of destination choice, e.g. “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 1200 zone study area would have 1,440,000 possible trip combinations in its
O–D table for each trip purpose. The most commonly used procedure for predict trip
distribution is gravity modelling. 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 trip
attractions) and on the basis of the distance to other zones (see Fig. 5).
A zone with a large number of trip attractions (say a large shopping centre) will receive a
greater number of distributed trips than one with a small trip attraction (a small shopping
centre).

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 15

To
From 1 2 3 4
1
2
3
4

Intra-zonal trips are

not assigned to the
road network that
connects inter-
zonal traffic

Separate tables are

used for each trip
purpose

Figure 5: Origin–destination table.

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 the destination (it is inversely
proportional). For example, you would expect more trips to a nearby shopping centre than
one further away. The distance effect is found through a calibration process which gives
travel times to destinations from the model similar to that found from field data. “Distance”
can be measured in several ways. The simplest way this is carried out is to use car travel
times between zones is as the measurement of distance.
Other ways might be to use a combination of car travel time and costs such as tolls as the
measurement of distance. Still another way is to use a combination of public transport and
car times and costs (composite cost). This method involves using a percentage of the car time
and cost and a percentage of the public transport time/cost. Because of calculation
procedures, the model must be used/run a number of times (iterated) in order to balance the
trip numbers to match the initial values.

8.3 Mode split

Mode split or mode choice is one of the most critical parts of the demand modelling process.
It is the step where trips between a given origin and destination are split into trips using public
transport, trips by car pool or as car passengers and trips by car drivers – see Fig. 6.
Calculations are conducted that compare the attractiveness of travel by different modes to
determine their relative usage.
All proposals to improve public transport or to change the ease of using the cars are passed
through the mode split/car occupancy process as part of their assessment and 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.
Mode split is carried out by 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 an origin and a destination.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
16 Transportation, Land Use and Integration: Perspectives for Developing Countries

Figure 6: Mode split.

The disutility equations do not normally recognize differences within travel modes. A bus
system and a rail system with the same time and cost characteristics will have the same
disutility values. It is possible that mode bias factors could be different for different
technologies, to represent a preference for light rail over bus, but this must be specifically
included.
Once disutilities are known for the various choices between an origin and a destination,
the trips are split among various modes based on the relative differences between disutilities.
A large advantage will mean a high percentage for that mode. Splits are calculated to match
splits found from actual traveller data. Sometimes a fixed percentage is used for the minimum
transit use (percent captive users) to represent travellers who have no transport available or
are unable to use a car for their trip.
Mode split and car occupancy analysis can be two separate steps or can be combined into
a single step, depending on how a forecasting process is set up. In the simplest application, a
highway/public transport split is made first which is followed by a split of car trips into car
driver and car passenger trips. More complex analysis splits trips into multiple categories
(single occupant auto (SOV), two-person car pool, 3–5-person car pool (HOV), van pool,
local bus, express bus, etc.). Car occupancy analysis is often a highly simplified process
which uses fixed car occupancy rates for a given trip purpose or for given household size and
car ownership categories. This means that the forecasts of car-pooling are insensitive to
changes in the cost of travel, the cost of parking, the presence of special programs to promote
car-pooling, etc.

8.4 Traffic assignment

Once trips have been split into highway and public transport trips, the specific path that they
use to travel from their origin to their destination must be found (Chiu et al. [20]). These trips
are then assigned to that path in the step referred to as traffic assignment. Traffic assignment
is the most time consuming and data intensive step in the process and is performed differently
for highway trips and public transport trips. The process first involves the calculation of the

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 17

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 larger travel time on that link. When speeds and travel times are changed, the
shortest path may change. Hence the whole process must be repeated many times (iterated)
until there is an equilibrium between travel demand and travel supply.
Trips on congested links will be shifted to uncongested links until this equilibrium,
condition occurs. There are a variety of ways in which the calculations are carried out to
reach network equilibrium, in order to keep the computer time to a minimum. One way
to get a feel for the accuracy of the models is to look at the resulting speeds on the network.
These should be realistic after equilibrium.
Public transport trip assignment is carried out in a similar way except that public transport
headways are adjusted rather than travel times. Public transport headways (minutes between
consecutive vehicles) affect the capacity of the public transport route. Low headways mean
that there is more frequent service and a greater number of vehicles. Public transport supply
and demand are also recalculated to reach equilibrium between supply and demand.
It is important to understand the concept of equilibrium. If a highway or public transport
route is congested during the peak hour, its excess trips will be shifted to alternative routes.
If the alternative routes are also congested the final results will show congestion over a wide
part of the network. In the real world, this congestion will eventually dissipate over time.
Another important step in assignment is the time of day analysis. Daily trip patterns need
to be converted to peak time period traffic. A key assumption needed is the portion of daily
travel that occurs during the peak period. This is normally used as a constant and conventional
travel models have very limited capability to describe how travellers will shift their trips to
less congested times of the day.
Some of the assumptions in traffic assignment are as follows:

 Delay occurs on links: Most traffic assignment procedures assume that delay occurs
on the links rather than at intersections.
 Travel only occurs on the network: It is assumed that all trips begin and end at a single
point in a zone (the centroids) and occurs only on the links included in the network. Not
all roads streets are included in the network nor all possible trip beginning and end
points included. The zone/network system is a simplification of reality.
 Capacities are simplified: To determine the capacity of roads and public transport
systems requires a complex process of calculations that consider many factors. In most
travel forecasts this is greatly simplified.
 Time of day variations: Traffic varies considerably throughout the day and during the
week. The travel demand forecasts are made on a daily basis for a typical weekday and
then converted to peak hour conditions. Daily trips are multiplied by a “hour adjustment
factor”, for example 10%, to convert them to peak hour trips. The number assumed for
this factor is very critical. A small variation, say plus or minus one percent, will make
a large difference in the level of congestion that would be forecast on a network. Most
models are unable to represent how travellers often cope with congestion by changing
the time they make their trips.
 Emphasis on peak hour travel: As described above, forecasts are carried out for the
peak hour. A forecast for the peak hour of the day does not provide any information on
what is happening the other 23 hours of the day. The duration of congestion beyond the

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
18 Transportation, Land Use and Integration: Perspectives for Developing Countries

peak hour is not determined. In addition, travel forecasts are made for an “average or
typical weekday”. Variations in travel by time of year or day of the week are usually
not considered.

9 MODEL CALIBRATION AND VALIDATION

Transportation models are aimed at reflecting an open system – the real world populated by
people who are responding to influences that are constantly changing; do not always make
rational decisions, and whose responses to influences affecting travel are not always the
same. Since transportation models (and travel modellers) cannot be omniscient, there will
always be missed information and abstractions resulting in less than perfect models.
Therefore, once the model is estimated and or set up using the four-step process, small
adjustments are made to it until the model accurately replicates observed patterns and
behaviours – referred to as calibration. Then the model is validated to a different set of
observed data (for example, traffic counts) than was used for model estimation or calibration
(United States Department of Transportation Federal Highway Administration [21],
Wegmann and Everett [22]).
 Model calibration is the adjustment of constants and other model parameters in
estimated or asserted models in an effort to make the models replicate observed data for
a base year or otherwise produce more reasonable results;
 Model validation is the application of the calibrated models and comparison of the
results against observed data; ideally, the observed data are not the same data used for
model estimation or calibration.
Model calibration and validation generally occur in an iterative fashion. Model validation
may reveal the need to return to the model estimation or model calibration steps. The
application of the model using future year conditions requires that the model forecasts are
reasonable and consistent with expectations and also might reveal a need to return to the
model estimation or calibration step.

10 TRANSPORTATION MODELLING IN PRACTICE

Once a model has been developed, calibrated and validated it is ready for application,
assessing policy decisions, testing the outcome of different scenarios or alternatives. This
section outlines a number of examples where transportation modelling has been used to
explore and solve “real life” problems – joining theory and practice. The examples
demonstrate the use of models on a variety of levels. From the local level site specific at the
one end to a regional level at the other end. To protect client confidentiality, street names
and features that may identify buildings or land marks have been removed from the figures
and images in this section.
The examples provided are presented in the form of seven case studies outlining the type
of problem that was encountered, the type of modelling software used to address the problem
and why and the outcome and or benefits of the modelling exercise.

Case study 1

The first case study relates to the local level – a site specific development, where a practical
problem is considered. That is: “what is the sizing/vehicle capacity requirement of the site’s
vehicle entrance(s) to accommodate the forecasted vehicle demand?” When determining the
number of vehicle entrances, there are two factors that determine ingress and egress capacity:

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 19

(i) the number of lanes provided; and

(ii) the distance between the car park barrier/gates and the public road (also referred to as
vehicle stacking distance).
If there are too little gate/lane capacity or the vehicle stacking distance is too short there
is a risk that vehicles will spill over into the public road or stack on the public road causing
congestion, road safety issues and traffic on the public road being negatively impacted (see
Fig. 7 as a general illustration).
This is a very common problem and question when faced with new site developments.
The success of the development depends on having sufficient access and the accesses
operating optimally. If it does not the land owner/shop keeper may lose business as shoppers
may consider going elsewhere if it is simply too difficult to access.
However, there is a very simple solution to solving the problem in isolation. A vehicle
queuing model, determining the probability of vehicles that may stack, based on expected
vehicle volumes and density, gate service levels and number of gates provided can be used
as a suitable modelling tool to determine the optimum access requirements. It is inexpensive
and data input requirements are limited.

Case study 2

The problem discussed in case study 1 can be expanded to a more complex problem where
for example a development site has multiple accesses over which traffic is distributed into a
large office building (as example). The question that arises is: “how do we design the car
park optimally given a number of accesses and the inter-operability between the levels of
structured or basement parking.’’ This problem is compounded when the parking facility is
a multi-storey (see Fig. 8) or underground car park with many parking levels accommodating
a few thousands parking bays. The attraction of vehicles to each access is based on an
‘attraction’ factor, determined by the perceived access to parking bays. Whilst the distribution
of vehicles around the site was determined during the trip distribution step in the four-step
model, it did not address how the interaction of vehicles will occur once within the building
and drivers are hunting parking spaces.

Figure 7: Determining vehicle access lane capacity.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
20 Transportation, Land Use and Integration: Perspectives for Developing Countries

Figure 8: Structured parking.

Although using a spreadsheet and queuing analysis model in isolation will provide some
clues to solve the problem it is incapable of solving the problem entirely due to the dynamic
nature of the problem. Why? Because simple queuing analysis models consider one access
problem at a time and not the inter relationship between different accesses along the same
street or from different streets to the same site or building. The problem is further exacerbated
where traffic streams at different accesses meet within the basement or structure parking as
a result of random internal basement route choice.
To deal with this problem and the inter-related driver choices and behaviour dynamics,
micro-simulation modelling with vehicle assignment functionality has proven to be a suitable
modelling tool. Fig. 9 illustrates a screen shot of a micro simulation model being used to
determine access requirements to a large-scale development.

Figure 9: Micro-simulation modelling and parking design.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 21

One of the benefits of micro-simulation modelling is the visual presentation of the

modelling results. Non-traffic engineers/transport planners are able to observe the ‘logic’ of
how traffic patterns occur in and around the site/building and therefore reduce scepticism
about whether the solution may be effective or not. The model is often used to demonstrate
to clients and local authorities that a specific access strategy works optimally or that some
other solution may be required to deal effectively with forecasted traffic volumes.

Case study 3

The previous two case studies examined access problems at the local or site specific level.
However, traffic engineers and transport planners often have to assess the operation of
intersections/interchanges on public roads, which are off site. Typically, such assessments
are necessary as part of Traffic Impact Assessment studies or Environmental Impact
Assessments. The challenges/problems faced here include some of the following:
 What is the most suitable intersection control type where two roads cross?
 What is the most appropriate configuration for an intersection?
 How much lane capacity is adequate?
 What geometry is suitable (e.g. grade separate or at grade)?
 How will a particular intersection operate in terms of Level of Service (LOS) and delay
in a future scenario given a certain volume of traffic?
When intersections are considered in isolation (where interaction between intersections
are not important for the purpose of the exercise), deterministic models such as SIDRA,
Oscady, Picady and Arcady are used. These models are reasonable inexpensive to operate
and data input requirements are limited. Existing and expected traffic volumes at an
intersection can be modelled assuming different intersection types and lanes or
reconfigurations can be added until a geometrical configuration is obtained that adheres to
acceptable design and operational standards. The model output in the form of LOS, vehicle
delay and queue lengths will inform how to conceptually design an intersection to a
geometrical optimal layout. A variety of Traffic Engineering Manuals and guidelines exist
defining the appropriate and or acceptable requirements and specification to render the
intersections’ operating level acceptable.
Typical requirements and specification include limited vehicle delay, limited vehicle
queuing and a Volume/Capacity ratio that is less than 0.95 (see Fig. 10).
Finding the appropriate intersection control type is an iterative process. The process that
would be typically followed is to test the most simplistic intersection type first given a
specific volume of traffic. A “build up” process is then followed. For example, the following
intersection types could be assessed sequentially for optimal suitability until an intersection
type is identified that satisfies guidelines, specifications and requirements:
 Give-way
 Three or four-way stop
 Roundabout/traffic circle
 Signalisation
 Grade separation
As each intersection type is considered, intersection specification is changed until an optimal
layout/geometry is determined. For example, lanes can be added to accommodate additional
through or turning movements. Traffic circles or roundabouts however presents issues of its

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
22 Transportation, Land Use and Integration: Perspectives for Developing Countries

Figure 10: Requirements and specification for different intersection control types.

own including large land or space requirements and pedestrians having difficulty crossing
from one side of the road to the other. Another factor to be considered is the cost of
implementing the intersections. The more complex the intersection the higher the cost.
Therefore, the decision about which intersection is the most appropriate depends on a variety
of factors that cannot be fully addressed by transportation modelling alone. The most suitable
intersection type is often a result of balancing different needs, consideration and stakeholder
requirements.

Case study 4
The previous case study focused on defining the most appropriate intersection control type
and geometrical layout of an intersection. However, intersections form part of a road network
comprising a variety of roads intersecting with one another and their impact/influence on
each other. Traffic flow at one intersection influences traffic flow at the next intersection
(particularly in urban areas) and therefore it may not be realistic to design or assess
intersections in isolation (particularly when the objective is to define the types of intersection
control for a large area). Signal controlled intersections provides the opportunity to
synchronise the operation of a series of intersections with one another allowing a platoon of
vehicles arriving at a certain speed to be able to reach each intersection during the green
phase and therefore not having to stop at any of the intersections. This is made possible by
including an offset time equal to the anticipated time that it takes to travel from one
intersection to another in the entire traffic phase.
The modelling software used for this operation would therefore need to be able to link one
intersection to another and deterministic models that views intersections in isolation would
not be able to provide this function. Fig. 11 illustrates a progression analysis output diagram
for a series of three intersections from Transyt software. Transyt is particularly useful in
signal optimisation and is reasonably inexpensive and data input requirements are limited.
Typical data input requirements include intersections count data, queue length data and signal
phasing plans.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 23

Figure 11: Transyt output.

Case study 5

In nodes or business districts where significant development growth occurs on multiple sites,
a network based approach to transport planning is considered. The transportation model used
should be able to assign traffic to the road network based on an estimated travel cost over a
sufficiently wide enough area. The development of such a model includes an iterative process
requiring multiple computations, which is too time consuming and repetitive to be calculated
on paper or by using a spreadsheet.
For large-scale development or when having to present a region or urban node by a
transportation model consisting of a vast network of road links and intersections affecting
being interconnected and affecting one another a microscopic model is suitable,
e.g. SATURN. These model types are expensive to develop, are traffic data hungry and take
a lot of time to calibrate and validate.
The model would be developed as described in section 8. For example, the area would be
divided into travel analysis zones and for each the number of trips ending in it or generated
by it would be identified. Road links and intersection capacity will be coded and act as the
potential pathways for trips to find their destination. The model will assign trips to the route
that shows the least resistance or in other words takes the least time for a trip to be made
between two points or nodes. The images below shows assigned traffic volumes for modelled
areas in Johannesburg using SATURN software (see Fig. 12).

Figure 12: SATURN output.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
24 Transportation, Land Use and Integration: Perspectives for Developing Countries

Case study 6

The traffic engineer or transport planner often has to decide what transportation infrastructure
would be most beneficial within a road network given specific traffic volumes and
characterises. Large scale infrastructure such as new interchanges, major roads and or bridges
has far reaching traffic impacts and costs. For instance, drivers living up to say 15 to 20 km
from a major road network might decide to change their travel behaviour due to new road
infrastructure provided elsewhere or the lack of it to improve their travel mobility.
Fig. 13 illustrates large scale infrastructure proposed at an existing grade separated
interchange. Again, microscopic modelling was carried out using SATURN software to
determine which of a number of geometrical options would produce the optimal level of
benefit to the surrounding area and whilst maintaining the mobility function of the road
network. In Fig. 13 below it was considered that the provision of the southern interchange
provides a new route option, while the northern interchange only increases existing capacity
on an established route. The new route option dramatically changes travel patterns and
relieves road infrastructure currently operating under tremendous pressure. This
infrastructure was considered to be the most beneficial and was recommended for
implementation.

Figure 13: Large-scale infrastructure assessment.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 25

Case study 7

At a strategic planning level decisions have to be made on regional, provincial and national
levels to determine the impact of long distance routes and other transport infrastructure and
the interaction between them. The type of route infrastructure and possible public transport
systems involved has huge capital costs and the benefits of its provision need to be
understood well by decision-makers. To support the decision-making process travel demand
models are often used to demonstrate the future expected utilisation of proposed
infrastructure. Such benefits are converted into social benefit i.e. travel time savings, new
employment opportunities and development opportunities to help decide among different
options.
Fig. 14 represents a snapshot of an EMME model, a microscopic travel demand model.
Whilst SATURN reviews the road network in more detail by considering for example number
of lanes on road links and exact intersection configurations, EMME considers each road only
as a route option with a generalised capacity. It therefore determines long distance travel
desire lines within a region and allocates traffic accordingly but also considering the routes
of least resistance and is therefore employed given its strategic geographical nature
advantage.

Figure 14: Regional transport model.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
26 Transportation, Land Use and Integration: Perspectives for Developing Countries

11 CONCLUDING COMMENTS
Transportation modelling and simulation as interface in the planning process plays an
important role in enabling spatial development. The models developed by transport planning
professionals are simplified representations of reality.
Transportation modelling is a young and developing science that has generally developed
as part of the evolution of advancing computer technology and was originally developed to
solve highway capacity problems. Many mistakes have been made in the past where
inappropriate solutions were implemented – with hind sight. A classic mistake includes
where new infrastructure provision is considered purely on the basis of the “predict and
provide” approach resulting in infrastructure being implemented completely surplus to
requirements and unnecessary. The consequence is an unpractical and unaffordable cost
to government.
However, although mistakes have been made in the past, models have also proven their
worth in solving highly complex problems and the current diverse range of modelling
software available means that models are becoming more bespoke in how they support traffic
engineers and transportation modellers addressing specific problems.
Transportation models should be seen for what they are: a support tool that helps in
decision making. They will always require humans to contextualise and interpret outputs,
considering the results in context of specific environments with unique characteristics,
features, balancing needs, different stakeholder opinions and humans occupying and living
in space.
Over time the relationship between transportation and land use has been better understood.
Transport design more often than not determines the success of urban environments and can
either be an enabling factor or a hindrance. Transportation modelling should therefore always
be considered as a tool that supports decisions and not the panacea to every traffic – transport
– land use problem we face or wish to solve. It is expected that transportation modelling will
diversify and develop significantly more sophisticated techniques to simulate human
behaviour and hence transport flow in the decades to come to better support decision making
in planning.

REFERENCES
[1] Goulias, K.G., Travel behavior and demand analysis and prediction, University of
California, 65 pp., 2007.
[2] Tri-County Regional Planning Commission, Travel forecast model development,
14 pp., 2009.
[3] American Association of State Highway and Transportation Officials, Handbook on
integrating land use considerations into transportation projects to address induced
growth, 52 pp., 2005.
[4] Toth, G., Integrating land use and transportation planning through placemaking.
Projects for public space, Online. [Link]
transportation-planning-through-placemaking/
[5] Wegener, M., Overview of land use transport models. Handbook of Transport
Geography and Spatial Systems, Vol. 5, Handbooks in Transport, eds D.A. Hensher &
K.J. Button, Elsevier: Amsterdam, pp. 127–146, 2004.
[6] United States Department of Transportation Federal Highway Authority of
Transportation, Planning for transportation in rural areas, Online.
[Link]
2012.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)
 Transportation, Land Use and Integration: Perspectives for Developing Countries 27

[7] United States Department of Transportation Federal Highway Administration, The

transportation planning process: key issues, 77 pp., 2007, Online.
[Link]
[8] Victoria Transport Policy Institute, TDM Encyclopaedia. Transport model
improvements improving methods for evaluating the effects and value of
transportation system changes, 2014, Online. [Link]
[9] Litman, T., Understanding transport demands and elasticities: How prices and other
factors affect travel behaviour, Victoria Transport Policy Institute, 76 pp., 2013.
[10] Chandrasekaran, B., Survey of network traffic models, Online.
[Link]
[11] Florian, M., Models and software for urban and regional transportation planning,
Centre for Research on Transportation, 2008.
[12] Brockwell, A.E., Chan, N.H. & Lee, P.K., A class of models for aggregated traffic
volume time series, 17 pp., 2008.
[13] Jha, K., Sinha, N., Arkatkar, S.S. & Sarkar, A.K., Modelling growth trend and
forecasting techniques for vehicular population in India. International Journal for
Traffic and Transport Engineering, 3(2), pp. 139–158, 2013.
[14] Meyer, D.M. & Miller, E.J., Urban Transportation Planning: A Decision-Oriented
Approach, McGraw-Hill: USA, 2001.
[15] He, N. & Zhao, S., Discussion of basic elasticity models applied to induced traffic in
China. Proceedings of the Eastern Asia Society for Transportation Studies, 9, 2013.
[16] Boxill, S.A. & Yu, L., An evaluation of traffic simulation models for supporting ITS
development, Center for Transportation Training and Research, Texas Southern
University, Houston, 116 pp., 2000.
[17] Department of Transport, Revised South African Trip Generation Rates Manual, 2007.
[18] Committee of Transport Officials, Technical Methods for Highways (TMH) 17,
Vol. 1, Trip Data Manual, 84 pp., 2012.
[19] Institute of Transport Engineers (ITE), Trip Generation Manual, 9th ed., 1216 pp.,
2012.
[20] Chiu, Y., Bottom, J., Mahut, M., Paz, A., Balakrishna, R., Waller, T. & Hicks, J.,
Dynamic Traffic Assignment, Transportation Network Modelling Committee,
Transportation Research Board, 62 pp., 2011.
[21] United States Department of Transportation Federal Highway Administration,
Calibration and Validation, 36 pp., 2009.
[22] Wegmann, F. & Everett, J., Minimum travel demand model calibration and validation
guidelines for the State of Tennessee, University of Tennessee, Center for
Transportation Research, 25 pp., 2009.

WIT Transactions on State of the Art in Science and Engineering, Vol 100, © 2017 WIT Press
[Link], ISSN 1755-8336 (on-line)