EMME Users’ Group Meeting
Recent toll patronage forecasting using EMME
27 May 2011
General Modelling Methodology
Base
Vehicle
Demand
Model
* Travel Surveys
* Demographics
* Transport Networks
Toll choice model is typically an “add-on” Traffic
Assignment :

Vehicle demand from other models

More detailed route assessment

One or more toll facilities
* Value of Time
(SP/RP Surveys)

Can be project specific

Projections for average weekday

Demands expanded externally for annual
revenue and patronage

Two general approaches:
–
–
Logit demand segmentation
Distributed VOT MC equilibrium assignment
Vehicle Trips
by Purpose
and Toll Class
Toll Choice Model
Road Network
Performance
Patronage and Revenue
Model
Route combinations
T3
B
T4
Tolled Segment
Access Route (untolled)
Egress Route (untolled)
T3
Full Toll Route
Possible Alternative Routes
T2
T1
A
for Single Time Period
Logit-based demand
segmentation model
Vehicle
Demand
by Class
Route Market Skimming
Toll
Route A
Skims
Toll
Route B
Skims
Toll
Route C
Skims
Toll
Route D
Skims
Toll
Route E
Skims
Non-Toll
Route
Skims
Toll
Route E
Demand
Non-Toll
Route
Demand
Demand Segmention
(LOGIT)
Toll
Route A
Demand
Toll
Route B
Demand
Toll
Route C
Demand
Toll
Route D
Demand
Cars
LCV
HCV
Market Trip Assignment
Toll Route A
Assignment
Toll Route B
Assignment
Toll Route C
Assignment
Toll Route D
Assignment
Total Flows
Convergence
Results
Toll Route E
Assignment
Non-Toll
Assignment
No
Logit based demand segmentation
Benefits of logit based demand segmentation assignment:

Most common method in Australian context

Strong financial market acceptability

Can address toll capping or user budget limits
Disbenefits of the logit based demand segmentation assignment:

Skimming and assignment to toll routes can be complex and
error prone

Limitations on toll route and vehicle class markets
combinations

Often used project specific application

Difficult to adopt for general city wide use or use for many toll
facilities
Distributed VOT
MC Assignment
Vehicle
Demand
Value of
Time
Segmention
16
14
Proportion of DEmand MArket
12
10
8
6
VOT
Group 1
4
VOT
Group 2
VOT
Group 3
VOT
Group 4
VOT
Group 5
VOT
Group n
2
0
1
2
3
4
5
6
Increasing Value ot Time (Higher willingness to pay toll)
Density
Density
Probability
Probability
5
4.5
5
4
4.5
3.5
4
3
3.5
2.5
3
2
2.5
1.5
2
1
1.5
0.5
1
0
0.5
0
Business
Commute
Business
Other
Commute
LCV
Other
HCV
LCV
HCV
Multi-class Equilbrium
Assignment
VTTS Parameters by Class
Redo
Assignments
Total Flows
Cars (3 Trip Purposes)
LCV
HCV
0
12
24
0
12
24
36
48
60
72
84
Value
36 of Time
48 ($/hr) 60
72
84
Convergence
Value of Time ($/hr)
Results
No
Distributed VoT multi-class assignment
Benefits of Distributed VOT multi‐class equilibrium assignment approach:

All possible toll route combinations are assessed at similar level of detail.

Use of transport software built‐in equilibrium assignment algorithms

Potential for reduced model run times and more stable outputs

Can address many toll roads together

Model could be applicable for general planning use

Less potential for user specification error
Disbenefits of Distributed VOT multi‐class equilibrium assignment approach:

Less commonly used and possibly not as well accepted

Cannot handle toll capping easily

Number of classes may limit VOT segmentation

Requires more innovative SP/RP survey analysis
Conclusions



Toll roads are an increasing feature of large
Australian city road networks
Methods for patronage forecasting (i.e. for bid
teams) have been complex and unwieldy
Distributed VOT MC assignment techniques may
be adaptable for general planning agency use