Mode Split Modelling

1
Mode Split Modelling
1. Trip-End Models 2. Trip-Interchange
Models 3. Explanatory Variables 4. Modes 5.
Decision Structures 6. Logit Models
2
Trip-End Models
It is possible to develop trip-end mode split
models, which are applied prior to trip
distribution (I.e., they split the trip ends
estimated by the trip generation model). Essentia
lly these are mode-specific trip
generation models, and take on the same form as
normal trip generation models (e.g., regression
or cross-classification).
3
Trip-End Models, contd
Trip-End mode split is a function of
socio-economic variables such as income, auto
ownership, etc. Cannot include modal
level-of-service attributes (travel time, cost,
etc.) since do not yet have O-D flows.
Trip-end models are suitable for small urban
areas where transit is basically a social
service, or in developing countries where mode
choice is almost completely determined by income
auto ownership.
4
Trip-End Model Example
5
Trip-Interchange Models
Trip-Interchange mode split models are
applied after trip distribution (I.e., they split
the trip interchanges or O-D flows computed by
the trip distribution model). Since O-D flows
are known, can compute travel times, costs, etc.
for the competing modes.
Trip-interchange models generally applied in
medium to large urban areas where transit
actively competes with auto.
6
Trip-Interchange Models, contd

• Since trip-interchange models are sensitive to
• travel times, etc. they can be used to assess the
• impacts of a broad range of transportation
  policies
• improved transit service (headways, coverage,
• travel times, etc.)
• road pricing, gasoline taxes, etc.
• transit fare policy
• parking supply/cost
• HOV policies
• ...

7
Explanatory Variables

• Variables characterizing each mode available to
• the trip-maker
• travel time
• in-vehicle
• out-of-vehicle
• walk (access, egress)
• wait (initial, transfer)
• out-of-pocket travel cost
• transit fare or in-vehicle auto cost
• parking cost
• other? (reliability, safety, comfort,
  convenience, )

8
Explanatory Variables, contd

• Variables characterizing the trip-maker
• income
• auto availability
• no. of cars in household
• drivers license?
• age
• sex
• occupation
• household composition
• .

9
Modes Modelled
The number and definition of modes
modelled depends on the problem application,
available data, network modelling capabilities,
etc. At a minimum, some representation of
the competition between auto and transit
is usually required.
10
Auto Mode Representation
One method is to distinguish between drivers
and passengers Auto Drive Mode trip-maker
drives a car from origin to destination.
Auto Passenger Mode trip-maker is a passenger
in a car from origin to destination.
11
Auto Mode Representation, contd
Alternatively, can distinguish between
drive-alone and shared-ride Drive-Alone
Mode trip-maker is the driver and sole
occupant of a car from origin to destination
(also referred to as a single-occupancy vehicle
(SOV) trip). Shared-Ride Mode trip-maker
is one of several occupants (driver or
passenger) of a car from origin to
destination (also referred to as a
high-occupancy vehicle (HOV) trip).
12
Shared-Ride Mode

• The shared-ride mode is sometimes further broken
• down by number of occupants, e.g.
• 2 persons
• 3 or more
• Note shared rides can occur among household
• members (e.g., parent drives child to school) or
• people from different households (e.g.,
  co-workers
• car-pool to work).

13
Transit Mode Representation

• Possible ways of categorizing transit include
• local (e.g., TTC) vs. regional (e.g., GO
  Transit)
• surface, shared right-of-way (bus, streetcar)
• vs. dedicated ROW (subway, LRT, busways)
• bus vs. rail
• regular service vs. express or other
  premium
• service

14
Mixed Modes

• Combined auto-transit modes exist, in which auto
• is used to access the transit system
• Park Ride trip-makers drives to a transit
  station
• and parks the car at the station
• Kiss Ride trip-maker is driven as a
  passenger
• to a transit station and is dropped off there

Auto access greatly extends the catchment area
of the transit service.
15
Representation of GO-Rail Trips, GTAModel
Access Sta. 1
Access Sta. 2
Egress Station
auto or local transit access trip link
local transit or walk egress trip link
GO-Rail line-haul trip link
Access Sta. 4 (chosen station for this trip)
Access Sta. 3
16
Other Modes

• Other motorized modes of interest in some
• cities might include
• jitneys
• ferries, water taxis
• taxis
• van pools
• motorcycles
• Non-motorized modes also generally of importance
• walking
• bicycles

17
Decision Structures
Given the set of modes to be included in the
model, one can always represent the mode choice
decision for a given trip as a decision tree, in
which each node of the tree represents an
alternative and relationships among choices can
be indicated through the hierarchical arrangement
of the tree.
18
Mode Definitions Choice Structure GTAModel
19
Possible Alternative ChoiceStructure
20
Logit Mode Choice Models
Current state of practice in modelling
trip-interchange mode split is to use the
multinomial logit model. This model is derived
from basic principles of random utility theory
developed in economics and psychology. Define
the following terms Ct Set of feasible
alternatives available to trip-maker t Uit
Utility of alternative i for trip-maker t
21
Random Utility Theory
It is assumed that people are rational and
so will chose the alternative i which maximizes
their utility for the given trip. I.e., a
trip-makers decision rule is Chose alternative
i element of Ct iff Uit gt Ujt for all j .ne. i
1
22
Random Utility Theory, contd
We do not observe person ts utility Uit. The
best that we can say is that Uit Vit e
it 2 Vit Systematic or observable
utility eit Random utility I.e., utility
is random, and so we cannot say with certainty
which alternative will be chosen.
23
Random Utility Theory, contd
Pit Probability that person t will choose alt.
I Probability that i is the max.
utility alt. for t P(Uit gt Ujt for
all j .ne. i i,j in Ct) 3 The mathematical
expression for Pit depends upon the distribution
of the random error term, eit. The most common
assumption is that the error terms are
identically and independently distributed (iid)
with the Type I Extreme Value distribution.
24
Logit Model
This assumption generates the logit model
exp(Vit) Pit ---------- 4 S j exp(Vjt)
25
Systematic Utility Function
Generally assume Vit bXit 5.1 b0
b1Xit1 b2Xit2 bkXitk 5.2 b
Vector of parameters (utility function
weights) Xit Vector of explanatory variables
26
Example The Ottawa-Carleton Work Trip Mode
Choice Model
The Ottawa-Carleton (O-C) Regional
morning peak-period work trip mode choice model
is a three-mode logit model. Modes modelled d
-- auto drive, allway p -- auto passenger,
allway t -- transit allway
27
O-C Work Trip Mode Choice Model, contd
Variable definitions Vm utility for mode
m (m d, drive p, passenger t,
transit) COSTm out-of-pocket travel cost (),
mode m IVTTm in-vehicle travel time for mode m
(min.) OVTTm out-of-vehicle travel time for mode
m (min.) NVEH avg. no. of vehicles per
household in home zone TWY 1 if emp. zone
is located within the catchment
area of a Transitway station outside the CBD
0 otherwise REGION1 if the home
zone is located in the Outaouais
0 otherwise
28
O-C Work Trip Mode Choice Model, contd
Systematic utility functions Vd -0.5472 -
0.5691COSTd - 0.0161IVTTd
0.7520NVEH 6.1 Vp -2.282 - 0.5691COSTp
- 0.0161IVTTp - 0.0261OVTTp
0.4529NVEH 6.2 Vt -
0.5691COSTt - 0.0161IVTTt -
0.0261OVTTt 1.0746TWY -
0.9784REGION 6.3
29
O-C Model, contd
Sample calculations using the O-C work trip
mode choice model.
30
Logit Model Parameter Estimation
Commercial software exists to estimate
model parameters using the method of
maximum likelihood estimation. Examples
include ALOGIT LIMDEP SPSS Results are
interpreted very similarly to regression (t-statis
tics, goodness-of-fit measures, etc.).
31
Forecasting
In forecasting, aggregate, zone-to-zone flows
are the required output. --gt Must aggregate
(sum up) over individual trip makers
(either explicitly or implicitly) to
generate required aggregate results. See text,
pgs. 298-302 for detailed discussion
of alternative methods. Most common practical
method is so-called classification with naive
aggregation.
32
Forecasting Example GTAModel Procedure
The GTAModel work trip mode split logit
model includes the no. of hh. vehicles (NVEH) and
whether a worker has a drivers licence (DLIC) as
variables. The aggregation/forecasting procedure
used is to 1. Divide workers into 5 different
NVEH,DLIC categories. 2. Estimate the
percentage of workers in each category for
each O-D pair. 3. Compute mode choice
probabilities for each category for each O-D
pair. 4. Compute weighted average mode splits
for each O-D pair.
33
GTAModel Example, contd
Define Tijm Predicted trips from I to j by
mode m Tij Total trips from I to j wijk
Fraction of workers living in I, working in
j who belong in NVEH-DLIC category k Pijkm
Logit probability of a worker of type k, living
in i, working in j. using mode
m Then Tijm TijS k wijkPijkm 7
34
Independence of Irrelevant Alternatives (IIA)
Property
Note that for two modes in the choice set, say I
and k (Pit/Pkt) exp(Vit)/exp(Vkt)
exp(Vit-Vkt) 8 I.e., relative probability
of choosing mode i vs.mode j depends only on the
utilities of i and j, independent of what other
alternatives are in the choice set. Problems
with this?
35
IIA, contd
If the IIA assumption is not valid, can go back
to Eqn. 3 and make another distributional
assumption for the error term, which will
generate a new model with different
characteristics. The two most common
alternatives to the logit model are 1. Nested
logit (can handle complex decision
structures). 2. Probit (normal error terms very
general model, but difficult to work with).