Scribd
Upload
197 views24 pages

Ch-3 Mode Choise

The multi-nomial logit model is used to estimate modal split for 5000 commuters with three travel options: private car, bus, and light rail. The utility functions for each mode are given based on cost and travel time. With costs and times provided for each mode, the model can calculate choice probabilities for each mode based on their relative utilities. Reducing the bus fare and recalculating the probabilities allows estimating how fare changes may impact ridership.

Uploaded by

Ķđ Gutema Alemayehu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
197 views24 pages

Ch-3 Mode Choise

The multi-nomial logit model is used to estimate modal split for 5000 commuters with three travel options: private car, bus, and light rail. The utility functions for each mode are given based on cost and travel time. With costs and times provided for each mode, the model can calculate choice probabilities for each mode based on their relative utilities. Reducing the bus fare and recalculating the probabilities allows estimating how fare changes may impact ridership.

Uploaded by

Ķđ Gutema Alemayehu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 24

Dilla University School of Civil Engineering, 2014 EC.

Transportation Planning and Modeling (CENG2191)


The four-step model
3. Mode Choice Modeling
Objectives

 To provide an introduction to mode choice models from


specification through to application
 Session checklist
Part 1: Introduction
Part 2: Aggregate choice models
Part 3: Disaggregate/discrete choice models
Introduction
Introduction

Factors influencing mode choices


1) Characteristics of the trip maker
 Car availability / car ownership (0, 1, 2+)
 Possession of a driving license
 Household structure
(age, number of employed, couple with children, singles, etc)
 Income
 Need for car for something else (at work, take children to school,
etc)
2) Characteristics of the journey:
 Mode (PT, Private)
 Trip purpose / frequent or rare (f.ex work trips are often easier to
undertake by public transport than other trips)
 Time of day (late trips are difficult to accommodate by public
transport)
Introduction

3) Characteristics of the transport facility


a) Quantitative factors
 Relative travel time; in-vehicle, waiting and walking time by mode
 Relative monetary costs (fare, fuel, parking, toll/road pricing)
 Availability and cost of parking
b) Qualitative factors
 Comfort and convenience
 Reliability and regularity
 Protection, security
Introduction

 Mode choice models


 sometimes developed as a separate (stand-alone) component of
the four stage model
 sometimes combined with distribution models
 sometimes defined to estimate main mode split (private
versus public) with sub-mode split determined in the
public transport assignment
In most countries, mode choice modeling is of great political importance
to improve the public transport system and to make the car drivers switch
to public transport.
Introduction

 mode choice models can be calibrated to aggregate or


disaggregate data
 aggregate models are calibrated to aggregate interzonal market
share data
 disaggregate models are calibrated at the level of the individual
traveler, to choice data based on either revealed preference (RP) or
stated preference (SP) data, or a combination of both
 disaggregate models are more likely to be estimated for a specific
purpose, whereas aggregate models are more likely to use behavioral
parameters transferred from elsewhere.
Aggregate choice models

 Aggregate models calibrated to modal market share data for


travel between zones.
 Based on the concept of generalized cost or generalized time.
 Models generally make use of parameter values (e.g. value of
time) transferred from elsewhere.
 Calibration undertaken to ensure that models have an appropriate
scale and can correctly forecast the base market shares.
Generalized cost

V = a1Tiv + a2Tat + a3Twt + a4Tin + a5F + a6


Where:
V = generalized cost of travel
Tiv = in-vehicle time
Tat = access time
Twt = waiting time9
Tin = interchange time
F = fare
a1 = value of time
a2 = the value of access time
a3 = value of wait time
a4 = value of interchange time
a5 = 1
a6 = a mode specific constant (accounts for unobserved or not
explicitly included factors)
Generalized time

V = a1Tiv + a2Tat + a3Twt + a4Tin + a5F + a6


Where:
V = generalized time of travel
Tiv = in-vehicle time
Tat= access time
Twt = waiting time
Tin = interchange time
F = fare
a1 = 1
a2 = the value of access time relative to in-vehicle time
a3 = value of wait time relative to in-vehicle time
a4 = value of interchange time relative to in-vehicle time
a5 = value of money (1/VOT)
a6 = a mode specific constant (accounts for unobserved
or not explicitly included factors)
modal split modeling
Example 1

 Let the number of trips from zone to zone is 5000, and two modes
are available which has the characteristics given in Table 1.
Compute the trips made by mode bus, and the fare that is collected
from the mode bus. If the fare of the bus is reduced to 6, then find
the fare collected.
Table 1: Trip characterisitcs
Example 1 Solution
Solution

The base case is given below.

Cost of travel by car (Equation)= = 2.08

Cost of travel by bus (Equation)= = 2.18

Probability of choosing mode car (Equation)= = 0.52

Probability of choosing mode bus (Equation)= = 0.475


Example 1 Solution
Proportion of trips by car = = 5000 0.52 = 2600

Proportion of trips by bus = = 5000 0.475 = 2400

Fare collected from bus = = 2400 9 = 21600

When the fare of bus gets reduced to 6,

Cost function for bus= = 1.88

Probability of choosing mode bus (Equation)= = 0.55

Proportion of trips by bus = = 5000 0.55 = 2750

Fare collected from the bus = 2750 6 = 16500


Multinomial logit model

 The binary model can easily be extended to multiple modes. The


equation for such a model can be written as:
Example 2

 Let the number of trips from zone to zone is 5000, and two modes
are available which has the characteristics given in Table 1. Compute
the trips made by mode bus, and the fare that is collected from the
mode bus. If the fare of the bus is reduced to 6, then find the fare
collected.
Disaggregate choice model

 Disaggregate models have their foundations in discrete choice analysis:


 branch of behavioral studies which uses mathematical models to
explain, forecast and evaluate decision makers’ choices.9
 consists of two interrelated tasks: specification of the behavioral
model and the estimation of the parameters of that model.
 The key features of the methodology are
 Focus on the decision-maker
 A discrete dependent variable (0-1, yes-no)
Disaggregate choice model:
The decision-maker
 In discrete choice models the unit of observation is the choice made
by the decision-maker
 decision-maker can be
 an individual
 a household0
 a company or
 a government
 In the context of mode choice models
 it is usually assumed that the decision-maker is the individual
 the decision-maker is faced with a choice between available
modes (each described in terms of its attributes)
Disaggregate choice model:
The logit model
The logit model
Value of time
multi-nomial logit model
Example 3

 Use of multi-nomial logit model for estimation of modal split


Use a logit model to determine the probabilities of a group of 5000 work commuters
choosing between three modes of travel during the morning peak hour:
Private car,
Bus and
Light rail.
 The utility functions for the three modes are estimated using the following
equations:
UC = 2.4 - 0.2C - 0.03T
UB = 0.0 - 0.2C - 0.03T
ULR = 0.4 - 0.2C - 0.03T
where
C = cost (£)
T = travel time (minutes)
For all workers:
The cost of driving is £4.00 with a travel time of 20 minutes
The bus fare is £0.50 with a travel time of 40 minutes
The rail fare is £0.80 with a travel time of 25 minutes.
Example 3 Solution

You might also like