Application of an Activity-Based Model for Highway Pricing Studies in Chicago
description
Transcript of Application of an Activity-Based Model for Highway Pricing Studies in Chicago
TPAC | Columbus, OH | May 2013
Application of an Activity-Based Model for Highway Pricing Studies in Chicago
CMAP Region Population: 10.5m 21 counties 2K TAZs 17K MAZs
Regional Comprehensive Plan GO TO 2040, adopted in 2010 Notorious traffic congestion Pricing queries lead to ABM development
Why Pricing? Theoretical
Turn external costs inward Practical
Cannot build our way out of congestion
Express or HOT lanes in the US (2012)
Why Develop ABM?6
Advanced Model Work Plan Pricing ABM Transit ABM ABM-DTA Integration
Benefits User segmentation Continuous value of time
Integrated Model System CT-RAMP demand model
Coordinated Travel Regional Activity-based Modeling Platform
Highway Assignments & Skimming Handles route choice for trucks, externals,
& airport traffic
Distributed Model System
Model Flow1. Population Synthesis
2. Long-term
4. Daily
5. Tour level
6. Trip level
2.1. Usual workplace
3.2. Car ownership
4.1. Person pattern type
Mandatory Non-mandatory Home
4.2.1. Frequency
4.2.2. Destination
4.2.3. TOD
4.3.1. Frequency
4.3.2. Party
4.3.3. Participation
4.3.4. Destination
4.3.5. TOD
4.4.1. Frequency
4.4.2. Allocation
4.5.1. Frequency
4.4.3. Destination
4.4.4. TOD
5.1. Tour mode 5.2. Stop frequency 5.3. Stop location
6.1. Trip mode
6.2. Auto parking 6.3. P&R parking
6.4. Trip departure
Individual mandatory tours
Joint Non-mandatory tours Allocated tasks Individual non-
mandatory tours
4.5.2. Destination
4.5.3. TOD
Available time budget
Residual time
Full day
3. Mobility 3.1. Free Parking 3.3. Transit pass 3.3. Toll transponder
2.2. Usual school
4.6.1. Frequency
4.6.2. Destination
4.6.3. TOD
At-work sub-tours
7. Network Simulations 7.1. List of trips 7.2. Trip tables 7.3. Assignment
Model Re-estimated for CMAP Pricing
ABM Auto ownership model Destination choice models Time-of-day choice models Mode choice models
CT-RAMP Person TypesPERSON-TYPE AGE WORK
STATUSSCHOOL STATUS
Full-time worker 18+ Full-time NonePart-time worker 18+ Part-time NoneNon-working adult 18 –
64Unemployed None
Non-working senior
65+ Unemployed None
College student 18+ Any College +Driving age student
16 – 17
Any Pre-college
Non-driving student
6 – 16 None Pre-college
Pre-school 0 – 5 None None
CT-RAMP Activity TypesPURPOSE DESCRIPTION CLASSIFICATIO
NELIGIBILITY
Work Working at regular workplace or work-related activities outside the home.
Mandatory Workers and students
University College + Mandatory Age 18+High School Grades 9-12 Mandatory Age 14-17Grade School Grades K-8 Mandatory Age 5-13Escorting Pick-up/drop-off passengers
(auto trips only).Maintenance Age 16+
Shopping Shopping away from home. Maintenance Age 5+
Other Maintenance
Personal business/services, and medical appointments.
Maintenance Age 5+
Social/Recreational
Recreation, visiting friends/family.
Discretionary Age 5+
Eat Out Eating outside of home. Discretionary Age 5+
Other Discretionary
Volunteer work, religious activities.
Discretionary Age 5+
Road Pricing Essentials Variation in Value of Time
ABM - continuous VOT distribution EMME - discrete VOT classes (high & low)
Vehicle occupancy ABM & EMME - 3 discrete classes
SOV, HOV2, HOV3+ Route type choice
ABM & EMME explicitly treat toll & non-toll users for each segment
Value of Time VOT by travel purpose & person type Car occupancy accounts for cost sharing
VOT(HOV2) = 1.6 * max(individual VOT) VOT(HOV3+) = 2.3 * max(individual VOT)
Static assignment & VOT
Example of VOT Distribution
EMME Macro 0. Initialize matrices mf123-mf166Travel (auto) modes mf123-mf134
Non-travel modes mf135-mf166
1. Base multi-class assignment to get travel times
2. SkimmingPrepare scenario for skimming
Skim for each segment mf175-mf234
3. Utility calculation
4. Route type choice modelTravel (auto) modes mf123-mf134
Non-travel modes mf135-mf166
5. Update demand matricesTravel (auto) modes mf279-mf290
Non-travel modes mf291-mf322
6. Report totals
7. Final assignment with saved class-specific volumes
Spe
cifie
d no
of i
nter
nal i
tera
tions
at e
ach
glob
al it
erat
ion
For base and target year pre-skimming
Integrate travel and non-travel modes
Implement assignment
Bypass auto modes if the core model is used
Bypass auto modes if the core model is used
Bypass auto modes if the core model is used
For the last global iteration only
Travel (auto) modes mf235-mf246
Non-travel modes mf247-mf278
MSA factor
Route Type Choice Currently implemented as binary choice
Toll vs. non-toll Explicit modeling of toll users at OD level Accounts for toll bias Allows for VOT variation/segmentation
beyond 12 assignable classes
Assignable Trip Tables Total of 44
12 for passenger auto (from ABM) 8 for trucks (EMME route type choice
model) 12 for externals (EMME route type choice
model) 12 for airports (EMME route type choice
model)
Initial Demand SegmentationVehicle Type & Value-Of-Time
Non-toll SOV
Non-toll HOV2
Non-toll HOV3+
Toll SOV
Toll HOV2
Toll HOV3+
Auto low VOT
1 3 5 2 4 6
Auto high VOT
7 9 11 8 10 12
Commercial 13 14Light truck 15 16Medium truck
17 18
Heavy truck 19 20External low 21 23 25 22 24 26External high
27 29 31 28 30 32
Airport low 33 35 37 34 36 38Airport high 39 41 43 40 42 44
Desired Multi-Class Assignment
Vehicle Type & Value-Of-Time
Non-toll SOV
Non-toll HOV2
Non-toll HOV3+
Toll SOV
Toll HOV2
Toll HOV3+
Auto + external + airport low VOT
1 3 5 2 4 6
Auto + external + airport high VOT
7 9 11 8 10 12
Commercial 13 14Light truck 15 16Medium truck 17 18Heavy truck 19 20
Multi-Class AssignmentVehicle Type & Value-Of-Time
Non-toll SOV
Non-toll HOV2
Non-toll HOV3+
Toll SOV
Toll HOV2
Toll HOV3+
Auto + external + airport low & high VOT
1 3 5 2 4 6
Commercial + light truck
7 8
Medium truck
9 10
Heavy truck 11 12
Projects New highways
IL 53/120 Elgin O’Hare West
Bypass Add lanes
I-90 I-290 I-55
IL 53/120
EOWB
I-90
I-55
I-290
Setting Toll Rates Current
Set to recover construction & operating costs
Congestion pricing Set to achieve performance objectives
Maintain free flow speed Maximize revenue Maximize throughput
Setting Toll Rates Set toll based on modeled delay Toll Rate = (Hrs of Delay * VOT per Hr) /
Miles
Travel Times (AM Peak)M
inut
es
$0.00
$2.76$1.30
$2.53
$3.41
$0.00
Toll Rate ComparisonsTo
ll ra
te ($
/mile
)
Other Findings Mode Share
Small HOV increase & SOV decrease Traffic Spillover
Arterials & General Purpose Lanes Decreased congestion
Other CMAP Efforts Outreach Further study
29
Questions?
Matt Stratton, [email protected] Wies, [email protected] Vovsha, [email protected] Stabler, [email protected]