Sie sind auf Seite 1von 14

7/13/2010

Atlanta Travel Forecasting Methods:


Traditional Trip-Based
&
Activity-Based Model
TMIP Webinar,, Julyy 14,, 2010
Guy Rousseau, Modeling Manager, Atlanta Regional Commission

Travel Demand Forecasting @ ARC


Where Do We Start?

Travel
Behavior

L d Use
Land
U &
Socio-Economic
Characteristics

Existing/Planned
E
i ti /Pl
d
Transportation
System

The Travel
Demand
Model Set
(The Black Box)

7/13/2010

ARCs Trip Generation


6 Trip Purposes:
Home-Based Work
Home-Based Shopping
Home-Based Grade School
Home-Based
Home Based University
Home-Based Other
Non-Home Based

ARCs Trip Production


IS NOT Cross-Classification, its Logit
Why? Allows more independent variables:
Household size (1, 2, 3, 4+)
Household income (<$20K, $2050K, $50
100K, $100K+)
# of workers per household (0, 1, 2, 3+)
3 )
# of children per household (0, 1, 2, 3+)
# of autos per household (0, 1, 2, 3+)

7/13/2010

ARCs Trip Attraction


Developed from Household Travel Survey
&W
Workplace
k l
E
Establishment
t bli h
tS
Survey
Features an Attraction Share Model based
on 4 market segments:
Households without cars
Households with cars < workers
Low income hshlds with # cars >= # workers
High income hshlds with # cars >= # workers

ARCs Trip Distribution


Still using Gravity Model, but looking into
D ti ti Ch
Destination
Choice
i
Makes use of Topographic Penalty to
compensate for area bias created by a
river crossing
The topo
topo penalty is a lump sum of time in
minutes (2 to 3) added to the composite
time of interzonal times for all zone pairs
on opposite sides of the river

7/13/2010

ARCs Mode Choice Fully Nested Logit Structure

ARCs Air Passenger Model Mode Choice Structure


Non-residents

Non Auto

Rental Car

Dropped Off

Taxi

Transit

Residents

Private Auto

Non Auto

Transit
Drive Self

Taxi

Dropped Off

7/13/2010

ARC Commercial Vehicle & Truck Model


based on
FHWA 13-Bin
Vehicle Classification

1: 2/3 Wheels

2: Automobiles

5: 2 Axle/6 Tire 6: 3 Axle Single


Unit Trucks
Trucks

9: 5 Axles Single
Trailer Trucks

3: 2 Axle/4 Tire
Trucks

7: 4+ Axle Single 8: 4 or Less Axles


Unit Trucks
Single Trailer Trucks

10: 6 or More Axles


Single Trailer Trucks

12: 6 Axles Multiple


Trailer Trucks

4: Buses

11: 5 or Less Axles


Multiple Trailer Trucks

13: 7 or More Axles


Multiple Trailer Trucks

91 External Stations (2024 + 91 = 2115 total taz)

7/13/2010

Atlanta Traffic Distribution By Hour


9.00%

8.00%

7.82%
7.33%

7.00%
6.71%

6 53%
6.53%

Pe rc e nta ge of Traffic

6.00%

5.83%

6.01%
5.60%
5.49%

5.09%
4.89%
4.75%

5.00%
4.61%

6.69%

5.12%

2000 Percent
1999 Percent
2 Year Average

4.04%

4.00%

3.40%

3.00%
60%
2.60%

2.00%

1.95%

1.89%

1.14%

1.00%

0.71%
0.54%

0.76%
0.51%

0.00%
1

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hours in a Day

7/13/2010

The 4-Step Process

ARC Travel Demand Forecasting


Land Use,
Population &
Employment

Traffic
Analysis
Zones (TAZ's)

Land Use
Models

Travel Pattern
Surveys
Trip
Generation

Trip
Distribution

Transportation
Facilities &
Performance

Highway and
Transit
Networks

Transit
Assignments

Mode and
Path Choice

Modal Trip
Tables

Auto
Occupancy
P/A to O/D

Highway
Assignments

Skim Trees

No

Speeds in =
Speeds Out
?

Yes

Vehicle
Emission
Estimates

ARC Model Applications: Using Radar


Graphs to Visualize Performance Measures
Scenario 1

Scenario 1a

Demand

Accessibility

Demand

Mode Share

Accessibility

Congestion

Mode Share

Congestion

Scenario 2

Scenario 2b

Demand

Demand

Accessibility

Mode Share

Congestion

Accessibility

Mode Share

Congestion

7/13/2010

ARCs Activity-Based Model


Provides results similar to 4-step trip
based model
Ok,
Ok so th
then why
h b
bother
th with
ith an ABM?
Because ARCs ABM provides additional
details, more info about travel patterns &
market segments
ABM allows to answer questions the 44
step model is not capable to provide
For internal use only, not for official
purposes, hence dual/parallel track of
models

ARC Activity-Based Modeling System


Based on the CT-RAMP1 family of ABMs developed in New York,
NY, Columbus OH (MORPC) and others
- Explicit intra-household interactions

- Continuous temporal dimension (Hourly time periods)


- Integration of location, time-of-day, and mode choice models
- Java-based package for AB model implementation

Implemented with the existing Cube-based networks, GUI and


ancillary models (external model, truck model, assignments, etc)
Households: 1.7 million in 2005, 2.7 million in 2030
Model development parallel effort with MTC

1Coordinated

Travel-Regional Activity-Based Modeling Platform

7/13/2010

Synthesized Population: Person Age Share


Age 80+

2030

Age 65-79

2005

Age 50-64
Age 35-49
Age 25-34
Age 18-24
Age 16-17
Age 12-15
Age 6-11
Age 0-5
0%

5%

10%

15%

20%

25%

30%

Share

Trip-Based Model Mode Share


Compared to ABM Mode Share
90%

80%

70%

60%

50%
HBW
40%

HBO
NHB

30%

20%

10%

0%
WLKL

WLKP

DRVL

DRVP

SOV

Trip-Based Model

HOV

WLKL

WLKP

DRVL

DRVP

SOV

HOV

Activity-Based Model

7/13/2010

Line Boardings: Trip Based Model Versus ABM

AM SOV Free:
Trip Length Frequency Distributions
450000

400000

350000

TBM
300000
Frequency

ABM
250000

200000

150000

100000

50000

0
1

11

13

15

17

19

21

23

25 27 29
Distance

31

33

35

37

39

41

43

45

47

49

51

10

7/13/2010

VMT by Time Period


80,000,000

70,000,000

60,000,000

VMT

50,000,000

40,000,000

Trip Concept3
ABM Concept 3

30,000,000

20,000,000

10,000,000

0
AM

MD

PM

NT

ARCs ABM Year 2005 Volume/Count Scatterplot

11

7/13/2010

What Sort of Performance Measures & Visuals are


Possible with an Activity-Based Model?
ABM results in a complete
p
activity
y diary
y for all
ARC residents
A wealth of activity/travel results
Just about any custom report/query/visual is
now possible
ibl
Performance Measures also available by
Age, Gender & Household Types

Mean Delay, Peak Period Travel

12

7/13/2010

Travelers By Age

Persons Not At Home By TAZ and Hour

13

7/13/2010

Persons By TAZ and Hour

Questions / Comments

Guy Rousseau (404 463-3274)


g
grousseau@atlantaregional.com
@
g
Atlanta Regional Commission
40 Courtland Street, NE
Atlanta, Georgia 30303
www.atlantaregional.com
g
Acknowledgements: PBS&J, AECOM, Parsons Brinckerhoff,
John Bowman, Mark Bradley, Bill Allen

14

Das könnte Ihnen auch gefallen