TRANSPORT PLANNING & MODELING (CENG

2191)

CHAPTER -III

Transportation Planning and Modeling

May 13, 2025

1
Lecture Overview

 Transportation Modeling

 Modeling principles

 Transport Modeling

 Prerequisite for transport modeling

 The Four step model

2
Transport Modeling
 Modeling principles
• Models are a simplified representation of a part of
reality.
• Is only realistic from a particular perspective.
• During their formulation, calibration and use,
planners can also learn much about the behavior
and internal workings of the system under scrutiny.
• Their function is to give insight into complex
interrelationships in the real world and to enable
3
Transport Modeling Cont…
 Transport Models

• Study of the behavior of individuals in making

decisions regarding the provision and use of

transport.

• Are abstract mathematical models

4
Transport Modeling Cont…

5
Transport Modeling Cont…
 Purpose of mathematical Models
• To gain a more 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

6
Transport Modeling Cont…
 Purpose of mathematical Models
• 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
7
Transport Modeling Cont…

 Prerequisite for transport Modeling

• Fundamental characteristics of transport

problems,

• Basic terms & Definitions

• Gather the necessary data and

• Understand basic regression analysis

8
Transport Modeling Cont…

 Fundamental characteristics of transport

problems
• Transport services come with side effects
• The demand for transport is derived; it is not an end by
itself.
• Transport demand takes place over space.
• Both transport demand & supply have very strong
dynamic elements.
• Transport is a service and not a good.
• The transport system requires fixed assets and the mobile
9
Transport Modeling Cont…

 Data requirements

• Socio-economic data

• Travel surveys

• Network data

• Land use inventory

10
Transport Modeling Cont…

 Data required for modeling is primarily

collected through surveys;
• Household survey
• External cordon and Intercept surveys
• Travel Diary
• O-D survey
• Questionnaire
• In-house and Roadside Interviews
11
Transport Modeling Cont…

 Mathematical background;

 Multiple regression analysis

 Elementary statistics

12
The Four Step Model
1. Trip-Generation
 Forecast number of
trips
 Predict O-D
frequency
 Estimate zonal trip
2. Trip-Distribution ends
 Predict O-D matrix 3. Modal-Usage
 Estimate zonal trip  Predict mode of
interchange transport
 Determine where the trips  Determine share of trip
to go

4. Trip-Assignment
 predicts the routes
choice Link volumes on
street network.
13
 Transport Modeling Cont…
Trip
Trip
To Zone
Distribution Modal Split
Generation
Trip Production 1 2 3 Pi
(P)Zone -- 1 10 18 19 47 Mode 1 15
From zone

1 47 2 30 32 4 66 Mode 2 25
2 66 3 5 40 65 110
3 110 Aj 45 90 88 223
Trip
Trip Attraction Route 1 5
Assignment
Zone
(A) -- Route 2 17
1 45 Route 3 3
2 90
3 88
14
Trip Generation

15
 Trip Generation

• Objective: to understand the reasons behind the trip making behavior,

and produce mathematical relation ship.
Trips generated by residential zones
Trip
Trip where trips may have home as origin or
Production
Generatio destination
n Trip Trips generated by activities at non-
Attraction home end of a home-based trip

Origi Destinatio Trip generation (Ti) is

n n sum of trip produced
(Pi) & trip attraction
(Aj)
Ti=Pi+ Aj

16
Trip Generation

 Journey:- is a one-way movement from a point of origin to a point

of destination.
 Home-Based (HB) Trip:- a trip which has one end of the trip at
Home Work
home. Ex. Trip to work, shopping, school.

 Non-Home- Based Trip:- A trip which does not end or start from
Work Shop
home. Ex. Trip from work for shopping.

 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. 17
Factors Affecting Trip Generation &
Attraction Rates
• Factors
1. affecting trip Production
Socio-economic
factors
 Household 2. Location factors:
Land use characteristics
structure
 Family size such as
 Value of land
 Income
 Residential
 Car ownership
 Industrial
 Family size
 Commercial etc.
 Composition of
distance from
household
civil area
structure and
[Link]
Public Transport
accessibility ( factors)
 Accessiblitu\y
 Location of Bus
stop, local train,
Plane
 Routes 18
Trip Generation Cont…

• Classification of trips
By trip Purpose:- In the By time of Day
 Peak Period
case Home-Based(HB)  Off-Peak
trips
 Period
Trip to School or Collage
 Trip to work
 Shopping trips
 Social and recreational and
 Other trips
By Person type
 Income Level
 Car Ownership
 Household size
and Structure

19
Basic Approaches for Trip-
Generation Analysis

1. Trip rate analysis

Uniform Growth
factor
2. Growth factor
analysis Average Growth
factor

Zonal-based
3. Regression Multiple regression
analysis
Household Based
(Category) 20
1. Trip Rates Based on Activity
unit

 Trips are attracted to zones for purposes such

as work, shopping, visiting friends and medical

trips.

 These types of “activity units” can be described

by measures such as square meter, number of

employees, etc.
21
Example Using Trip rate Analysis

 Determine the number of trips per day attracted to a zones

where 220 retail and 650 nonretail workers are employed.
Attraction Attraction per Attraction per Downtown Attraction per Other
per nonretail retail Employee Retail Employee
Household employee
HBW --- 1.7 1.7 1.7
HBO 1.0 2.0 5.0 10.0
NHB 1.0 1.0 3.0 5.0

 HBW: (220*1.7)+(650*1.7)= 1479

 HBO:(220*5)+(650*2)= 2400 N.B: HBO= Non-work trip
 NHB: (220*3)+(650*1)=1310
 Total= 5,189 trips per day 22
2. Growth factor Modeling

 Tries to predict the number of trips produced or

attracted by a house hold or a zone as a linear
function of explanatory variables.
• What Trips to be Considered
• What is the minimum age to be included in the
analysis

23
Growth factor Modeling Cont…

 The growth factor fi depends on

• Population (P) of the zone,
• Average house hold Income (I),
• Average vehicle ownership (V).
d d
f(Pi , I i , Vi )
fi  c c c
f(Pi , Ii , Vi )
d design year
c current year

• Merits= simple and easy to understand,

• Demerit= Over-estimated number of trips

24
 Example Using Growth Factor

 The distribution of present trips among the zones

A,B, & C are given in O-D matrix. The future trips
generated Ti are also given in the last column.
Distribute the future trips among the zones using
O\D A B C Ti
uniform
A
growth
60
factor
100
method
200 360
B 100 20 300 1260
C 200 300 20 3120

Ti j  t ij * F

 Solution: 25
 Example Using Growth Factor

O\D A B C ti Ti O\D A B C ti (C) Ti

A 60 100 200 360 360 A 218 365 729 1312 360
B 100 20 300 420 1260 B 365 73 1042 1532 1260
C 200 300 20 520 3120 C 729 1094 73 1896 3120
Total 1300 4740 Total 4740 4740

4740 Tij
 Solution: 1
F  3.646 
t ij 1300
 Multiply each cell in above matrix by uniform growth factor, F=3.646.
means, 60*3.646=218
 Future distributed trips are same as the total future trips given in data.
Tij 4740(calaulated)
F2   3.646
t ij calculated
 F1=F2, and Ti 1300
and Ti are also equal
26
 Example Average Growth Factor

 The distribution of present trips among the zones A,B, & C are
given in O-D matrix. The future trips generated Ti are also
given in the last column. Distribute the future trips among the
O\D A B C Ti
zones using average growth factor method
A 60 100 200 360
B 100 20 300 1260
C 200 300 20 3120
 Solution:  Fi  Fj 
Ti j  t ij * 
 2 

 Fi=growth factor for zone i=Pi/pi, Vertical values

 Fj=growth factor for zone j=Aj/aj, Horizontal values
27
 Example Average Growth Factor

 First Iteration O\D A B C pi Pi F=Pi/pi

Tba=200 A 60 100 200 360 360 1
Tbb=60 B 100 20 300 420 126 3
Tbc=1350 0
C 200 300 20 520 312 6
Tca=700
0
Tcb=1350
Tcc=120 aj 360 420 520 130
0
 Fi AjF  360 126 312 474
j
Ti j  t ij *   0 0 0
 2 
F=Aj/ 1 3 6
 Fa1  Fa1   1  1
Ta a  t aa *  aj  , T  60 *  60
 2 
aa
 2  
F  Fb2   1  3
Ta b  t ab *  a1  , T  100 *   200
2 
ab
 2   
F  Fc3  1 6
Ta c  t ac *  a1  , T  200 *  700 28
2 
ac
 2   
 Example Average Growth Factor

 Second IterationO\D A B C pi Pi F=Pi/pi

Tba=116 A 60 200 700 960 360 0.375
Tbb=47 B 200 60 135 161 126 0.783
Tbc=1499 0 0 0
C 700 135 120 217 312 1.438
Tca=634
0 0 0
Tcb=1499
Tcc=173 aj 960 161 217 474
0 0 0
 Fi  Fj  Aj 360 126 312 474
Ti j  t ij *   0 0 0
 2 
F=Aj/ 0.37 0.78 1.43
 Fa1 
F  0.375  0.375 
Ta a  t aa *  a1 aj  , T
5a a 60
3 *8   22
 2   2 
F  Fb2   0.375  0.783 
Ta b  t ab *  a1  , Tab  200 *   116
 2   2 
F  Fc3   0.375  1.438 
Ta c  t ac *  a1  , T ac  700 *   29
634
 2   2 
 Example Average Growth Factor

 Second Iteration

O\D A B C pi Pi F=Pi/pi
A 22 116 634 772 360 0.466
B 116 47 149 166 126 0.758
9 2 0
C 634 149 173 230 312 1.352
9 6 0
aj 772 166 230 474
2 6 0
Aj 360 126 312 474
0
Here, result obtained 0 from the 0two successive iterations give
approximately
F=Aj/ 0.46 equal
0.75growth
1.35 facto.
aj 6 8 2
30
3. Regression analysis model

 Used to establish a statistical relationship between the

number of trips produced and the characteristics of the

individuals, the zone, and the transportation network:

 Two types of regression models are commonly

used.

 Zonal-Based Multiple Regression

 Household-Based Regression (Category)

31
3.1. Simple Regression analysis
model Cont…

 Linear relations exists between dependent

variable with one independent variable.

 Independent Variable: Planning factors

 Dependent Variable: # of zonal trip ends (Productions

& attractions) by mode & trip purpose.

y a  bx
32
3.2. Multiple Regression analysis
model Cont…
For more than one independent variable:

Y a  b 1 x 1  b 2 x 2

And correlation coefficients are determined by the

equations

33
3.3. Category Analysis

 It is also called cross-classification technique.

 Considered the Household as the fundamental unit of

trip making.

 Number of trips from household is a stable function of

3 parameters.

1. Household income

2. Car ownership
34
Trip Generation Category
Analysis
 Zonal trip productions estimated as.
 h i t p
c c
Ti
 Zonal trip attractions estimated as,
c
T j  b j t a
c c

Where
Ti number of trips produced by zone i
c
h i number of households in zone i in category c
c
t p  trip production rate of category c
T j number of trips attracted to zone j
c
b j number of employment opportunities in category c
c
t a  trip attraction rate of employment category c
35
Category analysis model Cont…

Advantages:
• Simple concept of household trip making behavior.
• The technique categorizes the household according to socio-
economic characteristics and this appears rational.
• No mathematical relationship is derived..
• The computations are simpler
• Saves considerable time, effort and money.

Disadvantages:
• There is no scope for testing the statistical significance of
variables.
• New variables can not be introduced
36
• Large samples are needed to assign trip rates.
4. Balancing trip generation

 Prior to proceeding to Trip Distribution step, it is

important to ensure that the production and

attractions within the study area are equal.

 Productions are generally considered to be more

accurate.

 Attraction are balanced proportionality to match

productions within each zone. 37

Trip Distribution

38
Trip Distribution Cont…

 Is the procedure utilized to distribute the generated

trips from each zone to any other zone.

 Determine where the trips produced in each zone will

go- how they will be divided among all other zones in

the study area.

 Produce O-D matrix that shows the number of trips

originated in the study zone and where these trips are

39
Trip Distribution Cont…

 Trip distribution is a function of:

1. Zonal socio-economic and demographic

characteristics

2. The level of service, type and nature of

transportation facilities.

3. Spatial pattern, distribution and intensity of

land use
40
Trip Distribution Cont…

 Tij is the number of trips between origin i and

destination j.
 Oi is the total number of trips originating in zone i and
41
Trip Distribution Cont…

Trip Distribution Models

Growth Factor
Synthetic Methods
Methods

Gravity Model
Uniform Factor Model Intervening Opportunity
Average Factor Model
Model
Detroit Model
Frater Model Competing Opportunity
Furness Model Model
42
Tanner Model
The Growth Factor Methods

 Involve scaling an existing matrix (called base

matrix) by applying multiplicative factors (often
derived from predicted productions and/or
attractions) to matrix cells.
 The base year matrix contains an estimate of the
trips being made in the base year.
 Uniform Growth Factor
 Singly Constrained Growth-Factor
 Doubly Constrained Growth Factor
43
The Growth Factor Methods
Cont..
Uniform Growth Factor
 A uniform growth rate can be applied If the only
information available is about a general growth
rate for the whole study area.
Tij  t ij * F
F= growth factor
tij = base year (present) number of trips from zone I to zone j.
Tij= horizon year (future) number of trips from zone I to zone
j.

44
The Growth Factor Methods
Cont..

45
The Growth Factor Methods
Cont..
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 τi and destination-specific τj growth
factors respectively.

46
The Growth Factor Methods
Cont..
Case 1: Origin constrained growth factor is used
when estimation of growth in origin is more reliable
than that in destination

Case 2: Destination constrained growth factor is

used when estimation of growth in destination is
more reliable than that in origin

47
The Growth Factor Methods
Cont..
Case 3: average zonal growth factor is used when
estimations of growth and destination are equally
reliable

48
The Growth Factor Methods
Cont..
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.
 Iterative methods are proposed to obtain an estimated
trip matrix which satisfies both sets of trip-end
constraints, or the two sets of growth factors.

49
Doubly Constrained Growth
Factor

Furness Method

Furness suggests updating the initial matrix by

 adjusting alternatively both the constraints to

origins and destinations, until convergence is

reached.

 This technique is iterative since the solution for ai

depends on bj, and vice versa. 50

Growth Factor Cont…

The advantages of Growth Factor method are:

 Simple to understand.
 Preserve observed trip pattern.
 Useful in short term-planning.

The limitations are:

 Depends heavily on the observed trip pattern.
 It cannot explain unobserved trips.
 Do not consider changes in travel cost.
 Not suitable for policy studies like introduction of a
mode. 51
The Gravity model

 Analogous to Newton’s Law of Gravity.

m1m 2
F G 2
r
 The number of trips between two zones is
 Directly proportional to the number of trip
attractions generated by the destination zone.
 Inversely proportional to travel time between
the two zones.
52
The Gravity model

Comments on Gravity Model

• Most widely used trip distribution model
• Simple process with computer oriented technique
• Consider land use types and different trip purposes.
• Best spatial deference measurement between zones
• Over-estimated of density populated areas and under
estimated of lightly populated areas
• Uses aggregate data than individual data
• Cannot accommodate changes between travel modes
• Cannot define the trip making behavior change
53
• Calibration process is very difficult
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.

Where: α is the proportionality factor and

f(cij) is a generalized function of the travel
costs with one or more parameters for
54
The Gravity model Cont…

The need to satisfy the constraints (ΣTij=Oi and ΣTij=Dj)

requires replacing the single proportionality factor α by two
sets of balancing factors Ai and Bj as in the Furness model,

yielding:

The deterrence function f(cij) is the essence of the

gravity model.

55
The Gravity model Other
approach
Singly constrained gravity model

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 destination and

inversely proportional to a function of time of travel

56
between the two zones.
The Gravity model Other
approach

A j Fij K ij
Tij Pi *
A j Fij K ij
1
Fij Friction factor (travel time factor)Fij 
t c ij
if k ij 1, c 2
Aj
t 2 ij
Tij Pi *
 Aj 
  t 2 ij 

 

57
The Gravity model Other
approach
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
58
k = iteration number, k=1, 2, 3….n
Example

For study area consisting of three zones, the number of

production, attraction, and average travel times between each
zone is given as shown in the following tables. Compute the
intra zonal trips ( Zone to zone trips) by gravity model. Assume
kij=1 for all.

59
  
 

Example Solution
A jFij K ij
Tij Pi  n

 
  A j Fij K ij 
 j1 

 300 * 52 
 300 * 39  T21 330  188
T11 140   47  300 * 52   270 * 26   
180 * 26 
 300 * 39   270 * 52   180 * 50 
 270 * 26 
 270 * 52  T22 330  85
T12 140   57  300 * 50  270 * 26  180 * 26
 300 * 39   270 * 52   180 * 50 
 180 * 26 
 180 * 50  T23 330  57
T13 140   36  300 * 50   270 * 26   
180 * 26 
 300 * 39   270 * 52   180 * 50 
P 188  85  57 330
P 47  57  36 140

 300 * 50 
T31 280  145
 300 * 50   270 * 26   
180 * 39 
 270 * 26 
T32 280  68
 300 * 50  270 * 26  180 * 39
 180 * 39 
T33 280  68
 300 * 50   270 * 26   
180 * 39 
P 145  68  68 280

60
Example Solution

Table: Zone-to-zone trips: First Iteration, Singly Constrained

Zone 1 2 3 Computed P Given P
1 47 57 36 140 140
2 188 85 57 330 330
3 144 68 68 280 280
Computed A 379 210 161 750 750
Given A 300 270 180 750
 This first iteration is “singly constrained” in that the
computed productions match the given productions.
 However, the attractions do NOT converge at the
zonal level.
 Consequently, “new” attraction values should be 61
Example Solution
300
A 11 300 * 237
379
270
A 22 270 * 347
210
180
A 33 180 * 201
The process is then repeated: 161

Zone 1 2 3
T. production 140 330 280 750
T. Attraction 237 347 201 785

62
  
 

Example Solution
A jFij K ij
Tij Pi  n

 
  A j Fij K ij 
 j1 

 237 * 52 
 237 * 39  T21 330  153
T11 140   34  237 * 52   347 * 26   201 * 26 
 237 * 39   347 * 52   201 * 50 
 347 * 26 
 347 * 52  T22 330  112
T12 140   68  237 * 50  347 * 26  201* 26
 237 * 39   347 * 52   201 * 50 
 201* 26 
 201 * 50  T23 330  65
T13 140   38  237 * 50   347 * 26   201 * 26 
 237 * 39   347 * 52   201 * 50 
P 153  112  65 330
P 34  68  38 140

 237 * 50 
T31 280  116
 237 * 50   347 * 26   201 * 39 
 347 * 26 
T32 280  88
 237 * 50  347 * 26  201 * 39
 201 * 39 
T33 280  76
 237 * 50   347 * 26   201 * 39 
P 116  88  76 280

63
Example Solution

 The second iteration, which is now “doubly

constrained” results in the computed productions
and attractions converging (reasonably close) to the
Table: Zone-to-zone trips: Second Iteration, Doubly Constrained
given values.
Zone 1 2 3 Computed P Given P

1 34 68 38 140 140

2 153 112 65 330 330

3 116 88 76 280 280

Computed A 303 268 179 750 750

Given A 300 270 180 750

64
Exercise

 A self contained city having four residential zones A,B,C and D.

two industrial estate X and Y, the generation equation shows
that trips from home to work from each residential area are
given below during 24 hours per day. There are 3690 jobs in X
zone and 4495 jobs in Y zone. It is also known that attraction
b/w zones is inversely proportional to square of journey times
b/w zones. The journey time is mentioned in table.
 Calculate the inter zonal trips for home to work by gravity
Zones X Y
model.
A 14 19
 A=1000, B=2245, C=1750, D=3190
B 16 11
C 9 11
D 14 21 65
Exercise: Solution
A j Fij K ij
Tij Pi *
A j Fij K ij
1
Fij Friction factor (travel time factor)Fij 
t c ij
if k ij 1, c 2
Aj
t 2 ij
Tij Pi *
 Aj 
  t 2 ij 
 

A=1000 Journey time

B=2245 Zone X Y
s
C=1750 A 14 19
D=3190 B 16 11
Total Jobs X C 9 11
zone=3690 66
Exercise: Solution
3690 4495
Tax 1000 * 14 2 602, Tay 1000 * 19 2 398,
 3690 4495   3690 4495 
  14 2  19 2    14 2  19 2 
3690 4495
2 2
Tbx 2245 * 16 628, Tby 2245 * 11 1617
 3690 4495   3690 4495 
  16 2  112    16 2  112 
3690 3690
2 2
Tcx 1750 * 9 964, Tcy 1750 * 11 786
 3690 4495   3690 4495 
  9 2  112    9 2  112 
3690 3690
Tdx 3190 * 14 2 2070, Tdy 3190 * 212 1120
 3690 4495   3690 4495 
  14 2  212    14 2  219 2 
67
Exercise: Solution

O\D X Y Tij for origin zones

A,B,C,D (Total
population)
A 602 398 1000
B 628 1617 2245
C 964 786 1750
D 2070 1120 3190
Total calculate 4264 3921 8185
attraction (Cj)
Total predicted 3690 4495 8185
attraction (Aj)

68
Exercise

 Here, total attractions do not tally with the predicted

attractions.
 Hence, second iteration is required to balance for that,
A j * A1 j
the total A
attractions are first adjusted.
2
 j
1
C j

A 2 j adjusted attraction after iteration 1

A j desired attraction
A 1 j predicted attraction at iteration 1
C1 j Calculated attraction at iteration 1
For second iteration
3690 * 3690
A 2 j for zone X  3193
4264
4495 * 4495
A 2 j for zone Y  5153
3921

69
Exercise: Solution

A=1000 Journey time

B=2245 Zones X Y

C=1750 A 14 19

D=3190 B 16 11

Total Jobs X zone=3193 C 9 11

Total Jobs Y zone=5153 D 14 21

70
Exercise: Solution
3193 5153
Tax 1000 * 14 2 553, Tay 1000 * 19 2 467,
 3193 5153   3193 5153 
  14 2  19 2    14 2  19 2 
3193 5153
2 2
Tbx 2245 * 16 508, Tby 2245 * 11 1736
 3193 5153   3193 5153 
  16 2  112    16 2  112 
3193 5153
2 2
Tcx 1750 * 9 842, Tcy 1750 * 11 909
 3193 5153   3193 5153 
  9 2  112    9 2  112 
3193 5153
Tdx 3190 * 14 2 1858, Tdy 3190 * 212 1332
 3193 5153   3193 5153 
  14 2  212    14 2  219 2 
71
Exercise: Solution

O\D X Y Tij for origin zones

A,B,C,D (Total
population)
A 533 467 1000
B 508 1736 2245
C 842 909 1750
D 1858 1332 3190
Total calculate 3741 4444 8185
attraction (Cj)
Total predicted 3690 4495 8185
attraction (Aj)

72