Transport planning & modeling 2023 G.C.

Trip generation models

Growth factor,

𝑇 =𝑓 𝑡
𝑖 𝑖 𝑖

𝑓𝑖 = 𝑓(𝑃𝑖𝑑,𝐼𝑖𝑑,𝑉𝑖𝑑)
𝑓(𝑃𝑖𝑐,𝐼𝑖𝑐,𝑉𝑖𝑐)
Regression,

𝑇 =a + a x + a x + … a x … + a x
𝑖 0 1 1 2 2 i i k k

Where:-

Xi: are explanatory variables such as income, Car, ownership, population etc

Ti: is generated trip.

ai: are parameters determined through calibration process.

Trip distribution

The Growth Factor methods

Uniform Growth Factor

Singly Constrained Growth-Factor

Doubly Constrained Growth Factor

Uniform Growth Factor

𝑇 =𝜏𝑡
𝑖𝑗 𝑖𝑗

Where: - τ is the uniform growth factor,

- 𝑡𝑖𝑗 is the previous total number of trips and

- 𝑇𝑖𝑗 is the expected total number of trips.

ASTU UNIVERSITY Page 1

 Transport planning & modeling 2023 G.C.

Singly Constrained Growth-Factor

𝑇 =𝜏 𝑡
𝑖𝑗 j 𝑖𝑗
for origin-specific factors

𝑇 =𝜏 𝑡
𝑖𝑗 i 𝑖𝑗
for destination-specific factors

Doubly Constrained Growth Factor

𝑇 =𝑡 𝜏 𝜏 𝐴 𝐵
𝑖𝑗 𝑖𝑗 𝑖 𝑗 𝑖 𝑗

𝑇 =𝑡 𝑎 𝑏
𝑖𝑗 𝑖𝑗 𝑖 𝑗

The procedure is:

Set 𝑏𝑗 = 1
With 𝑏𝑗 = 1, solve for 𝑎𝑖 to satisfy trip generation constraint (Σ𝑇𝑖𝑗=𝑂𝑖𝑗).
1.

With 𝑎𝑖=1, solve for 𝑏𝑗 to satisfy trip attraction constraint (Σ𝑇𝑖𝑗=𝐷𝑗𝑖).

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

The Gravity model

𝑇𝑖𝑗= 𝛼*𝑂𝑖*𝐷𝑗*𝑓(𝑐𝑖𝑗)
Where: α: is the proportionality factor and
f(cij): is a generalized function of the travel costs with one or more parameters for
calibration.
The need to satisfy the constraints (ΣTij=Oij and ΣTij=Dji) requires replacing the single
proportionality factor α by two sets of balancing factors Ai and Bj as in the Furness model,

𝑇𝑖𝑗= 𝐴𝑖𝐵𝑗𝑂𝑖𝐷𝑗𝑓(𝑐𝑖𝑗)
yielding:

The deterrence function f(cij) is the essence of the gravity model.

Reciprocal (𝑐𝑖𝑗) =m/ 𝑐𝑖𝑗

Exponential (𝑐𝑖𝑗) = exp (−𝛽𝑐𝑖𝑗)
Power (𝑐𝑖𝑗) =𝑐𝑖𝑗^−𝑛

ASTU UNIVERSITY Page 2

 Transport planning & modeling 2023 G.C.

Modal choice

Logit model

Trip Assignment

ASTU UNIVERSITY Page 3

 Transport planning & modeling 2023 G.C.

There are two broad assignment models:

 The all-or-nothing (AON) assignment
 User equilibrium assignment (UE), system
All-or-Nothing Assignment
All traffic between an O-D pair is assigned to just one path (usually the shortest path)
connecting the origin and destination.
Unrealistic in that only one path between every O-D pair is utilized even if there is another path
with the same or nearly the same travel time.
Travel time is taken as a fixed input and does not vary depending on the congestion on a link.
May be reasonable in sparse and uncongested networks where there are few alternative routes
and they have a large difference in travel cost.
Used to identify the desired path
The user equilibrium assignment
The user equilibrium assignment is based on War drop’s first principle, which states that:
Under equilibrium conditions, traffic arranges itself in congested networks in such a way that no
individual trip maker can reduce his path costs by switching routes. This means, in the
congested network, all the used routes between an O-D pair have equal and minimum costs
while all unused routes have greater or equal costs.
The user equilibrium assignment assumes that:
 The user has perfect knowledge of the path cost.
 Travel time on a given link is a function of the flow on that link only.
 Travel time functions are positive and increasing.

Chapter – five (5) Examples

Example 1: Trip Generation Using Growth Factor Modeling

Given that a zone has 250 household with car and 300 household without car and the average
trip generation rates for each groups is respectively 5.0 and 2.0 trips per day. Assuming that in
the future all household will have a car, find the growth factor and future trips from that zone.
Assume that the population and income remains constant (Pd=pc & Id=Ic) and Show that the
growth factor modeling can lead to over estimating the future trip.

Solution:

Current trip rate (ti) =2x1+5x2

Where: x1 is number of household without car

X2 is number of household with car.

ASTU UNIVERSITY Page 4

 Transport planning & modeling 2023 G.C.

Current trip (ti) = 2*250+5*300=2000trips/day.

Growth factor (fi) = 𝑓 𝑃𝑖𝑑,𝐼𝑖𝑑,𝑉𝑖𝑑)

𝑓(𝑃𝑖𝑐,𝐼𝑖𝑐,𝑉𝑖𝑐)
Vd= all households of the zone (250+300=550)
Vc= 250
fi= 550/250=2.2
Future trip (Ti) =fi* ti
=2.2*2000
=4400trips/day

However, the method is obviously very crude. If we use our information about average trip
rates and make the assumption that these will remain constant (which is actually the main
assumption behind one of the most popular forecasting methods, as we will see below), we
could estimate the future number of trips as:

Ti = 550 × 5 = 2750trips/day

Which means that the growth factor method would overestimate the total number of trips by
approximately 60 %( 4400-2750)*100/2750

Example 2: Trip Generation Using Regression Analysis Model

Let the trip rate of a zone is explained by the household size done from the field survey. It was
found that the household sizes are 1, 2, 3 and 4. The trip rates of the corresponding household
are as shown in the table below. Fit a linear equation relating trip rate and household size.

ASTU UNIVERSITY Page 5

 Transport planning & modeling 2023 G.C.

Therefore, Y = 1.3X + 0.33

Example 3: Trip Distribution Using Uniform Growth Factor

Trips originating from zone 1, 2, and 3 of a study area are 78, 92 and 82 respectively and those
terminating at zones 1, 2, and 3 are given as 88, 96 and 78 respectively. If the growth factor is
1.3 and the cost matrix is as shown below, find the expanded origin-constrained growth trip
table.

𝑇 =𝜏𝑡
𝑖𝑗 𝑖𝑗

ASTU UNIVERSITY Page 6

 Transport planning & modeling 2023 G.C.

Example 4: Trip Distribution Using Doubly Constrained Growth Factor

The base year trip matrix for a study area consisting of three zones is given below. The
productions from the zone 1, 2 and 3 for the horizon year are expected to grow to 98, 106, and
122 respectively. The attractions from these zones are expected to increase to 102, 118, and
106 respectively. Compute the trip matrix for the horizon year using doubly constrained
growth factor model.

ASTU UNIVERSITY Page 7

 Transport planning & modeling 2023 G.C.

Example-5

ASTU UNIVERSITY Page 8

 Transport planning & modeling 2023 G.C.

Solution:
4X4 matrix of
Assume kij=1
Step find fij=?
Using travel time putting in matrix form
Eg. For F11 travel time is 5 therefore for travelling time of 5 minute the value of friction factor
is 1.3

Step 2
Determining trip distribution

ASTU UNIVERSITY Page 9

 Transport planning & modeling 2023 G.C.

Eg

A 2∗F 12∗K 12
T12=p1* A 1∗F 11∗K 11+ A 2∗F 12∗K 12+ A 3∗F 13∗K 13+ A 4∗F 14∗K 14
700∗0.95∗1
= 1000* 1000∗1.3∗1+700∗0.95∗1+ 6000∗0.8∗1+500∗0.65∗1

As a attraction values are not as expected we need to conduct column factoring

ASTU UNIVERSITY Page 10

 Transport planning & modeling 2023 G.C.

Here the production is not matched

We have to conduct row factoring (RF)

ASTU UNIVERSITY Page 11

 Transport planning & modeling 2023 G.C.

Example 6: Trip Distribution Using Gravity Model

The productions from zone 1, 2 and 3 are 98, 106, 122 and attractions to zone 1, 2 and 3 are
102, 118, and 106. The function f (cij) is defined as f (cij ) = 1/ Cij 2. The cost matrix is as shown
below. Compute the trip matrix for the horizon year using gravity model.

Solution: The first step is given in table 1. The second step is to find B j. This can be found out as
the following.
Step1: Computation of parameter Ai

Table 1: Computation of parameter Ai

Step2: Computation of parameter Bj

ASTU UNIVERSITY Page 12

 Transport planning & modeling 2023 G.C.

Table 2: Computation of parameter Bj

Step 3: Final Step

+-
Table 3: Final table

Example 7: Trip Distribution using Fratar Method

A study area consists of four zones (A, B, C, and D). An O-D survey indicates the number of trips
between each zone as shown in Table-1. Planning estimates for the area indicate that in five
years then number of trips in each zone will increase by the growth factor shown in Table-2 and
the trip generation will be increased to the amounts shown in the last column of the table.
Determine the number of trips between each zone for future conditions.

ASTU UNIVERSITY Page 13

 Transport planning & modeling 2023 G.C.

Table 1; Present trip between zones

Table 2; Present Trip generation and growth factors between zones

ASTU UNIVERSITY Page 14

 Transport planning & modeling 2023 G.C.

ASTU UNIVERSITY Page 15

 Transport planning & modeling 2023 G.C.

Example 8: Modal Choice Using Logit Model

Use logit model to predict the choice mode. The utility functions for auto and transit are as
follows.

Where T1= total travel time (minutes)

T2= waiting time (minutes)

C= cost (cents)
The travel characteristics between two zones are as follows:

ASTU UNIVERSITY Page 16

 Transport planning & modeling 2023 G.C.

Example9: Modal Choice Using Logit Model

The city decides to spend money to create and improve bike trails so that biking becomes a
viable option, what percent of the trips will be by bike? Assume:

 A bike trip is similar to a transit trip.

 A bike trip takes 5 minutes more than a transit trip but with no waiting time.

 After the initial purchase of the bike, the trip is “free”.

Travel characteristics between two zones

ASTU UNIVERSITY Page 17

 Transport planning & modeling 2023 G.C.

Variable Auto Transit Bike

ak -0.46 -0.07 -0.07
t1 20 30 35
t2 8 6 0
c 320 100 0

Utility functions for auto, bike and transit is

U = ak– 0.35t1 – 0.08t2 – 0.005c

Where,

 ak = mode specific variable

 t1 = total travel time (minutes)

 t2 = waiting time (minutes)

 c = cost (cents)

Solution:

 Utility factors

 Uauto = -0.46 – 0.35(20) – 0.08(8) – 0.005(320) = -9.70

 Utransit = -0.07 – 0.35(30) – 0.08(6) – 0.005(100) = -11.55

 Ubike = -0.07 – 0.35(35) – 0.08(0) – 0.005(0) = -12.32

 Logit Model:

 p(auto) = _____eUa ____ = _______e-9.70 ______ = 0.81

eUa + eUt +eUb e-9.70 + e-11.55 + e-12.32

 p(transit) = _____eUt__ __ = ______e-11.55 ______ = 0.13

eUa + eUt +eUb e-9.70 + e-11.55 + e-12.32

 p(bike) = _____eUb__ __ = ________e-11.55 ______ = 0.06

ASTU UNIVERSITY Page 18

 Transport planning & modeling 2023 G.C.

eUa + eUt +eUb e-9.70 + e-11.55 + e-12.32

Therefore: The trip proportion by mode will be

 80% by auto
 13% by transit
 6% by bike

Example 10: Trip Assignment Using All-or-Nothing Assignment Model

To demonstrate how this assignment works, an example network is considered. This network
has two nodes having two paths as links. Let us suppose a case where travel time is not a
function of flow as shown in the figure below, in other words it is constant. Take q 12 = 12

Solution: The travel time functions for both the links is given by:

t1 = 10

t2 = 15

Example 11: Trip Assignment Using User Equilibrium Assignment Model

Let us suppose a case where travel time is a function of flow as shown in other words it is
constant as shown in the figure below.

ASTU UNIVERSITY Page 19

 Transport planning & modeling 2023 G.C.

Solution:

T1 = 10+3X1 and T2 = 15+2X2

Total flow is X1 + X2 = 12 ………..a

Equilibrium occurs when T1 = T2 . Therefore, 10+3X1 = 15+2X2………b

From equation “a”, X1 = 12- X2

Substitute X1 into equation “b”, 10+3(12- X2) = 15+2X2

10+36-3 X2 = 15+2X2

31=5 X2

X2 = 6.2 trips

Then, X1 = 12- X2 = 12-6.2

X1 = 5.8 trips

ASTU UNIVERSITY Page 20