This document describes transportation modeling and planning using the four-step model. It discusses data collection and characterization for the study area of North Cyprus. The four steps of the model are then explained: trip generation calculates the number of trips produced and attracted in each zone based on household surveys; trip distribution develops an origin-destination matrix to show trip flows between zones; modal split determines the proportion of trips using each transportation mode; and trip assignment allocates the trips to the transportation network. The document provides examples of applying each step of the four-step model to the study area.

1. TRANSPORTATION MODELING
AND PLANNING
Report
The Four-Step Model
(An Example)
North Cyprus

3. 1 | P a g e
Contents
1. Introduction: .................................................................................................................... 2
2. Purpose of Study.............................................................................................................. 2
3. Methodology:................................................................................................................... 3
3.1. Data collection ......................................................................................................... 3
3.2. Data Characterization .............................................................................................. 3
3.3. Data preparation...................................................................................................... 3
3.1. Growth factor modeling........................................................................................... 4
3.2. Defining Study Area, And Zoning ............................................................................. 4
4. The Four-Stage Model...................................................................................................... 6
4.1. Trip Generation:....................................................................................................... 6
4.2. Trip Distribution ....................................................................................................... 8
4.3. Modal split ............................................................................................................. 10
4.4. Trip assignment...................................................................................................... 12
5. Conclusion:..................................................................................................................... 14
6. Reference:...................................................................................................................... 15

4. 2 | P a g e
1. Introduction:
The demand for transport is derived, it is not an end in itself. With the possible
exception of sightseeing, people travel in order to satisfy a need (work, leisure,
health) undertaking an activity at particular locations (Ortuzar and Willumsen,
2002).
Travel Demand Forecasting is a key component of the transportation engineer's
technical repertoire. It allows the engineer to predict the volume of traffic that will
use a given transportation element in the future, whether that element is an existing
highway or a potential light-rail route. The history of demand modeling for person
travel has been dominated by the approach that has come to be referred to as the
four-step model (FSM) (Hensher, 2008). The FSM is the primary tool for forecasting
future demand and performance of a transportation system, typically defined at a
regional or sub-regional scale (smaller scales often apply simplified models)
(Knoflacher et. al., 2017).
The Four-Step Model is four separate sub-models that are performed sequentially.
The four steps are trip generation, trip distribution, modal split, and trip assignment.
Each step in the model requires information from the previous step (s). The end
product of the Four-Step Model is the demand ( the number of forecasted trips on the
network) on each link of the network (Meyer and Miller 2001).
Figure 1.1 Components of the four-step model ( Mintesnot and G. Woldeamanuel, 2016).
2. Purpose of Study
The objectives of the course-work are: To understand how to collect data and
analysis it by using household survey for FSM. To gain knowledge of the needs for
transport demand model in transport projects evaluation. To gain knowledge of the
various techniques applied in transport demand modeling.

5. 3 | P a g e
3. Methodology:
3.1. Data collection
According to HTM (2005), the Minimum sample size [%] is 10% for population
in analyzed area below 50,000. In this study 13.1% has been used. The study area
consists of 320 apartments and 83 single-family detached home. This survey has
been conducted in Hasplaot region, has collected data on more than 53 households.
Data are divided as follows; 35 apartments, 12 houses, 3 with no response and 3
classified as unrealistic data.
3.2. Data Characterization
The arithmetic mean of a population in flats is 3.15 and in houses are 3.67
persons. 61% are men and 39% are women. The average auto occupancy is 1.6 in
single-family detached homes and 0.66 in flats. 45% of population is student, during
survey it is noticed that most population at flats are students where there are only 16
families in whole campus. Full-Time status is about 23% and 8% is retired, if it is
compared with present of people more than 60 years, it is reasonable and indicates
the extent of homogeneity of data.
3.3. Data preparation
Sample expansion
Once the data have been collected it is necessary to expand them in order to
represent the total population. The following expression is fairly general in this
sense:
= = 8.76
Where A is the total number of addresses (Household) in the original population
list, B is the total number of addresses selected as the original sample, and D is the
number of samples where no response was obtained.
34%
42%
15%
9%
Age Ranges
<20
20-40
40-60
>60
23%
5%
45%
12%
7%
8%
Working Status
Full work
Part work
Full study
Pre school
unemployed
Retired

6. 4 | P a g e
Validation of results
The current research study has been used simple method based on site checks of
the completeness and coherence of the data. For example, one of the household when
he asked for mode of trip, he said by using Airplane, so it is classified as non-
eligible. The other factor used in this study to check the reality is manual count. The
MC has conducted between 8:30 a.m. and 9:00 a.m.
3.1. Growth factor modeling
Growth factor modes tries to predict the number of trips produced or attracted by
a house hold or zone as a linear function of explanatory variables. The models have
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 a growth factor. 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. In this report assuming that in the future, auto ownership will Increase
80% to meet forecast demand, assuming that the population and income remains
constant.
3.2. Defining Study Area, And Zoning
Study area: Once the nature of the study is identified, the study area should be
defined such that majority of trips have their origin and destination in the study area
and should be bigger than the area-of-interest covering the transportation project. In
this study Nicosia city are selected as a study area based on the significant of the
town and needs of traffic studies. Fig 1.2 have shown the location of study area.
Zoning: Once the study area have selected, the next step is defining the origin
and destination of passenger and freight. In order to reduce the number of origin
and destination of travellers, the travellers' locations are grouped into zones such as
city blocks. In current study, the zone defined based on shape, main roads, and
nature border of the area. The zone shape is a triangle. It is bordered on the north
corner by Haspolat roundabout, the south leg by Dr. Fazıl Küçük Cd, from the east
leg by Bülent Ecevit Cd. Street, and on the west side by Zafer CD. The zone consist
of two types of residential buildings, about 83 houses and 20 building (each building
has 16 flats). Fig 1.2 presents the selected zone and future work zoning.

7. 5 | P a g e
Figure 1.2 shows the selected zone and future work zoning.

8. 6 | P a g e
4. The Four-Stage Model
4.1. Trip Generation:
The first step in four step model is forecasting trip generation. The final product
of this step is two forecast numbers for each zone: a number of trips produced and a
number of trips attracted. One of the methods for calculating these numbers is cross-
classification.
Table 1.1 have shown trip rates.
Time
# of trips Trip rate hour / Home
in out In out Total
6:30-7:30 0.00 2.00 0.00 0.04 0.04
7:30-8:30 0.00 22.00 0.00 0.48 0.48
8:30-9:30 1.00 10.00 0.02 0.22 0.24
9:30-10:30 0.00 8.00 0.00 0.17 0.17
10:30-11.30 0.00 5.00 0.00 0.11 0.11
11.30-12:30 3.00 1.00 0.07 0.02 0.09
12:30-13:30 1.00 1.00 0.02 0.02 0.04
13:30-14:30 7.00 1.00 0.15 0.02 0.17
14:30-15:30 1.00 1.00 0.02 0.02 0.04
15:30-16:30 5.00 0.00 0.11 0.00 0.11
16:30-17:30 13.00 1.00 0.28 0.02 0.30
17:30-18:30 7.00 0.00 0.15 0.00 0.15
18:30-19:30 3.00 0.00 0.07 0.00 0.07
19:30-20:30 2.00 1.00 0.04 0.02 0.07
20:30-21:30 2.00 0.00 0.04 0.00 0.04
21:30-22:30 4.00 0.00 0.09 0.00 0.09
22:30-23:30 1.00 0.00 0.02 0.00 0.02
23:30-00:30 1.00 0.00 0.02 0.00 0.02
00:30-01:30 0.00 0.00 0.00 0.00 0.00
01:30-02:30 0.00 0.00 0.00 0.00 0.00
02:30-03:30 1.00 0.00 0.02 0.00 0.02
∑ 52.00 53.00 1.130 1.152 2.28
From table above the trip rate for the zone = 2.30 trip per home, and the total
number of trip is 105 for the sample. The approximately numbers of trip of trip can
be estimated as following:
Total trips (produce) per day = Number of trips (sample) x expansion factor.
= 105x8.76=919.8 trips/day

9. 7 | P a g e
Figure 1.2 shows peak Spreading.
a) Characterization of Journeys:
By Purpose
It has been found in practice that a better understanding of travel and trip
generation models can be obtained if journeys by different purposes are identified
and modeled separately (See table 1.2). The work and education trips are usually
called compulsory (or mandatory) trips and all the others are called discretionary (or
optional) trips.
By Time of Day
Trips are sometimes classified into peak and off-peak period trips; the proportion
of journeys by different purposes usually varies greatly with time of day. Table 1.2
presents characterization of Journeys by time of day, the morning (AM) peak period
occurred between 7:00- 9:00 (the evening peak period is sometimes assumed to be its
mirror image) and the representative off-peak period was taken between 10:00 -
12:00.
Table 1.2 Characterisation of Journeys By Time of Day
Purpose
AM Peak (7:00-
9:00)
AM Off Peak
(10:00-12:00)
No % No %
Home/Hotel 1 4% 0.022 1 13%
Place of work 8 31% 0.174 ‒ 0%
Shopping 1 4% 0.022 ‒ 0%
Education 16 62% 0.348 7 88%
Leisure/Religious ‒ 0% ‒ ‒ 0%
Visit family/friends ‒ 0% ‒ ‒ 0%
Business ‒ 0% ‒ ‒ 0%
0.00
0.10
0.20
0.30
0.40
0.50
0.60
Triprate
Time

10. 8 | P a g e
One comment are in order with respect to this table 1.2. The numbers of trip
generation for peak period are 26 trips. Secondly, note that although the vast majority
(92%) of trips in the AM peak are compulsory (i.e. either to work or education), this
is not the case in the off-peak period.
b) Trip generation forecasting example:
Table 1.3 Base year number of households and trips by HH size and auto
ownership.
Auto ownership
Hosehould
(HH) size
0 car 1 car 2+ car
# of
HH's
# of
trips
# of
HH's
# of
trips
# of
HH's
# of
trips
1 person 1 2 1 2 0 0
2 Persons 9 18 5 12 0 0
3+ Persons 9 20 10 24 12 30
Total trips (produce) per day = Number of trips (sample) x expansion factor.
= 105x8.76=919.8 trips/day
Table 1.4 Base year trip rates by HH size and auto ownership.
Hosehould
(HH) size
Auto ownership
0 car 1 car 2 car
1 person 2 2 0
2 Persons 2 2.4 0
3+ Persons 2.22 2.4 2.5
The trip rate for the zone = 2.30. For total number of trips expected to be
generated in the feature, the current trips must be multiply by the growth factor = 1.8
Total trips (produce) in future = Number of trips (current) x growth factor.
= 919.8x1.8=1576.8 trips/day
4.2. Trip Distribution
The second step is modeling the microscopic flow of forecasted trips from each
zone to every other zone. The trip generation step estimates the number of trips
produced and attracted to each zone. Trip generation is commonly expressed using
an O-D (Origin-Destination) matrix. An O-D matrix operates as an input-output
table. Where trip origins in each zone are inputs and trip destinations to each zone
are outputs, which spatially expresses the interaction of trips throughout the study
area. The O-D matrix is the final product of the trip distribution step (Ben-Akiva
2008).

11. 9 | P a g e
Table 2.1 A general form of a two-dimensional trip matrix (Base year) for sample.
Destination
Z1 Z2 Z3 Z4 Z5 Z6 Z7 Z8 Z9 Z10 Z11 Z12 Z13 Z14
Origins
Z1 ‒ 22 1 3 1 2 4 5 1 2 1 3 1 1
Z2 1 1 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z3 3 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z4 10 ‒ 1 ‒ ‒ ‒ ‒ ‒ ‒ ‒ 1 ‒ ‒ ‒
Z5 1 1 ‒ 2 ‒ ‒ ‒ 1 ‒ ‒ ‒ ‒ ‒ ‒
Z6 19 ‒ 2 ‒ ‒ 1 ‒ 1 ‒ ‒ ‒ 2 ‒ ‒
Z7 1 ‒ ‒ ‒ ‒ ‒ ‒ 1 1 1 ‒ ‒ ‒ ‒
Z8 2 ‒ ‒ ‒ ‒ ‒ ‒ ‒ 1 ‒ ‒ ‒ ‒ ‒
Z9 5 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z10 1 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z11 1 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z12 1 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z13 3 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z14 4 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Table 2.3 O-D matrix zones code.
Code Zone Code Zone
Z1 Haspolat Z8 Terminal
Z2 CIU Z9 Göçmenköy
Z3 Lemar Market Z10 Gönyeli
Z4 Erülkü sup. Z11 Taşkınköy
Z5 Ercan Airport Z12 NEU
Z6 Hamitköy Z13 Girne
Z7 yolu okullar Z14 Famagusta
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,
that is a uniform growth rate ( .
Table 2.3 Future estimated trip matrix with = 1.8
Z1 Z2 Z3 Z4 Z5 Z6 Z7 Z8 Z9 Z10 Z11 Z12 Z13 Z14
Z1 ‒ 347 16 47 16 32 63 79 16 32 16 47 16 15.8 741.1
Z2 157.7 1 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z3 473 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z4 1577 ‒ 1 ‒ ‒ ‒ ‒ ‒ ‒ ‒ 1 ‒ ‒ ‒
Z5 157.7 1 ‒ 2 ‒ ‒ ‒ 1 ‒ ‒ ‒ ‒ ‒ ‒
Z6 2996 ‒ 2 ‒ ‒ 1 ‒ 1 ‒ ‒ ‒ 2 ‒ ‒
Z7 157.7 ‒ ‒ ‒ ‒ ‒ ‒ 1 1 1 ‒ ‒ ‒ ‒
Z8 315.4 ‒ ‒ ‒ ‒ ‒ ‒ ‒ 1 ‒ ‒ ‒ ‒ ‒
Z9 788.4 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z10 157.7 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z11 157.7 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z12 157.7 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z13 473 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
Z14 630.7 ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒ ‒
8199
Destination
Origins

12. 10 | P a g e
4.3. Modal split
The third stage in travel demand modeling is modal split. The modal spilt model
disaggregates the trip distrusted by travel mode (care, bus, train walk…etc.). In other
words, the step estimates the proportion of trips originating and arriving in each zone
that made by passengers using different mode. Table 3.1 presents trip rates by mode
per hour/home.
Table 3.1 trip rates by mode per hour/home.
Time
Trip rates by mode per hour/Home
Vehicle occupation Passenger Bus Pedestrian
6:30-7:30 0.043 0.000 0.000 0.000
7:30-8:30 0.348 0.000 0.000 0.130
8:30-9:30 0.065 0.043 0.065 0.065
9:30-10:30 0.065 0.000 0.022 0.087
10:30-11.30 0.022 0.043 0.000 0.043
11.30-12:30 0.022 0.022 0.043 0.000
12:30-13:30 0.043 0.000 0.000 0.000
13:30-14:30 0.109 0.000 0.022 0.043
14:30-15:30 0.022 0.000 0.000 0.022
15:30-16:30 0.000 0.000 0.000 0.109
16:30-17:30 0.174 0.000 0.000 0.130
17:30-18:30 0.065 0.022 0.022 0.043
18:30-19:30 0.043 0.022 0.000 0.000
19:30-20:30 0.022 0.000 0.000 0.043
20:30-21:30 0.022 0.000 0.022 0.000
21:30-22:30 0.043 0.022 0.000 0.022
22:30-23:30 0.000 0.000 0.000 0.022
23:30-00:30 0.022 0.000 0.000 0.000
00:30-01:30 0.000 0.000 0.000 0.000
01:30-02:30 0.00 0.000 0.000 0.000
02:30-03:30 0.02 0.000 0.000 0.000
Total 1.152 0.174 0.196 0.761
Percentage 50% 8% 9% 33%
Figure 3.1 shows the relative size of modes split of zone. The pie charts (Figure
3.1 a. b) have shown that the pedestrian percentage (53%) increases in peak period
morning because majority of population are students (45%) and most of them
traveling to CIU by walking. In addition, the public transport has increased from
(9%-23%).

13. 11 | P a g e
The bar chart (see Figure 3.1 c.) shows that the car drivers in houses is higher
than in flats, this can be explained by the high rates of average vehicle occupancy in
houses (1.6 car per house and 0.66 per flat) and also it is due to the majority of
population in flats are student as well as economic factors . The second thing to
notice is percentage of walk and bus modes are 34% and 25% respectively and fully
reflected in houses with 0%.
Figure 3.1 Modes split of transport for trip making
58%
9%
33%
(a) Mode split for all day (in+out).
Vehicle occupation
Public transport
Pedestrian
53%
27%
20%
(b) Mode split for peak period
Vehicle occupation
Public transport
Pedestrian
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Car Driver Passenger Bus Walk L.G Vehicle
(c) Mode split for for different type of development.
Flats Single-Family home

14. 12 | P a g e
4.4. Trip assignment
Network assignment or trip assignment is the final stage of the four-step model.
Trip assignment models aim to determine the number of trips on different links (road
sections) of the network given the travel demand between different pairs of nodes
(zones). The trip assignment involves assigning traffic to a transportation network
such as roads and streets or a transit network. Network assignment applies the trip
distribution O-D matrix and mode data to the transportation network. There are many
used methods for trip assignment such as all-or-nothing. In all-or-nothing technic all
trips between an origins and a destination assumed to take shortest way.
An example for as all-or-nothing method
It is unrealistic to produce a plan for traffic assignment without knowing all
information about whole region (all zones) to understand volumes and trips
distribution for these zones. In this example we assumed available information just
for practicing. By using O-D matrix (Table 2.1) to assign the trips to the network
given in Figure 4.1 using all-or-nothing method. The output will be the list of links in
the network and their corresponding traffic volume.

15. 13 | P a g e
As discussed early in this method we used short way. For example, all of the 35
vehicles that travel between nodes Z1 and Z12 will travel via Z7-Z11 (shortest travel
time). The table shown below indicates the route that we selected from Haspolat
zone to other destination.
Table 4.1 choosing the shortest path between Haspolat zone (orgin) and other destination.
From Node To Node Link/road Travel time (min)
1 2 1‐2 3
3 1‐3 3
4 1‐4 10
5 1‐4,4‐5 22
6 1‐6 10
7 1‐6,6‐7 25
8 1‐6,6‐8 24
9 1-6,6-11,11‐9 29
10 1‐6,6‐11,11‐9,9‐10 41
11 1‐6,6‐11 20
12 1‐6,6‐11,11‐12 28

16. 14 | P a g e
5. Conclusion:
 Travel Demand Forecasting is a key component of the transportation
engineer's technical repertoire.
 The four-step model (FSM) is the primary tool for forecasting future demand
and performance of a transportation system, typically defined at a regional or
sub-regional scale
 The trip rate for the zone = 2.30 trip per home, and the total number of trip is
105 for the sample.
 The numbers of trip generation for peak period are 26 trips. Secondly, note
that although the vast majority (92%) of trips in the AM peak are compulsory
(i.e. either to work or education), this is not the case in the off-peak period.
 The pedestrian percentage (53%) increases in peak period morning because
majority of population are students (45%) and most of them traveling to CIU
by walking. In addition, the public transport has increased from (9%-23%).
 The car drivers in houses is higher than in flats, this can be explained by the
high rates of average vehicle occupancy in houses (1.6 car per house and 0.66
per flat) and also it is due to the majority of population in flats are student as
well as economic factors . The second thing to notice is percentage of walk
and bus modes are 34% and 25% respectively and fully reflected in houses
with 0%.

17. 15 | P a g e
6. Reference:
1) Aldian, A. (2006). On the development of evaluation system and transport
demand model for road network planning in developing countries:(a case
study of Indonesia) (Doctoral dissertation, University of South Australia).
2) Butler, M. N. (2012). An assessment tool for the appropriateness of
activity-based travel demand models(Doctoral dissertation, Georgia
Institute of Technology).
3) Hensher, D. A., & Button, K. J. (Eds.). (2007). Handbook of transport
modelling. Emerald Group Publishing Limited.
4) Knoflacher, H., & Ocalir-Akunal, E. V. (Eds.). (2017). Engineering Tools
and Solutions for Sustainable Transportation Planning. IGI
Global.Chicago.
5) Ortuzar, Juan de Dios, and Luis G. Willumsen. Modelling transport. Vol.
3. 2002.
6) United Nations. Statistical Division. (2008). Designing household survey
samples: practical guidelines (Vol. 98). United Nations Publications