Activity-Based Modeling System for Travel Demand Forecasting
Submitted to:
Metropolitan Washington Council of Governments
777 North Capitol Street, N.E.
Washington, D.C. 20002-4226
Sponsored by:
U.S. Department of Transportation
400 7th Street, SW
Washington, D.C. 20590
and
U.S. Environmental Protection Agency
401 M Street, SW
Washington, D.C. 20460
Prepared by:
RDC, Inc.
311 California Street, Suite 620
San Francisco, CA 94109
September 1, 1995
NOTE: This project was jointly sponsored as part of the Travel Model Improvement Program (TMIP), by the U.S. Department of Transportation (US DOT) and U.S. Environmental Protection Agency (US EPA). The sponsors' intent behind the project was to investigate and develop the idea of activity-based forecasting in an applied setting. RDC, Inc. developed the Activity-Mobility Simulator, or AMOS, as a tool in this regard.
The information contained in this report represents the views of RDC, Inc., and is not necessarily endorsed or recommended by the US DOT or US EPA.
The project sponsors are interested, however, in articulating their support for continued investigation and assessment of activity-based forecasting techniques, and studying their potential for: (1) applications in transportation, and (2) any improvements to modeling practice. To date, several metropolitan areas have demonstrated an interest in activity modeling, and have started to undertake activity surveys for their regions; among these are Boston, Oahu, Detroit, Dallas/Ft. Worth, Raleigh/Durham, and Portland, OR. As part of TMIP, US DOT and US EPA anticipate continued exploration of the role an activity-based approach can play in travel forecasting.
Abstract: Activity Based Modeling System for Travel Demand Forecasting
This study is probably the first attempt to develop and implement a full-fledged activity-based policy analysis tool for a metropolitan region and thereby examine whether activity-based approaches can be put to practical use. In particular, the study attempts to determine whether an operational activity-based tool can be developed while utilizing available data, supplemented by a medium-scale survey that can be conducted with modest mounts of monetary and time resources.
Author: RDC, Inc.
Publication Year:
September 1995
Sponsor(s): Metropolitan
Washington Council of Governments; U.S. Department of Transportation; U.S.
Environmental Protection Agency
Table of Contents
1 Introduction
1.1 Background
1.2 Study Objectives
1.3 AMOS Features
1.4 Study Conclusions
1.5 Outline of Report
2 A Critical Review of the Trip-Based, Four-Step Procedure of Urban Passenger Demand Forecasting
2.1 Advantages
2.2 Internal Inconsistencies
2.3 Data Inefficiency
2.4 Lack of Behavioral Foundation
2.5 Resulting Problems as a Policy Tool
2.6 Summary
3. Why the Activity-Based Approach?
4. Overview of AMOS
4.1
Paradigm Shifts
4.2 Structure of
the Model System
4.3 Data Needs
4.4
Areas of Application
5. AMOS System Components
5.1
Baseline Activity - Travel Analyzer
5.2
TDM Response Option Generator
5.3
Activity-Travel Pattern Modifier
5.4
Evaluation Module and Acceptance Routines
5.5
Statistics Accumulator
6. Application to Washington, D.C. Area
6.1
AMOS Survey
6.2 AMOS
Survey Sample Profile
6.3 Analysis
of Stated Responses to TDM Strategies
6.4
Implementation of TDM Response Option Generator
6.5
Implementation of AMOS with MWCOG Databases
6.6
Examples of AMOS Application to Commuters in MWCOG Sample
7. Policy Analysis
7.1
Evaluation Measures
7.2
Micro-Simulation Procedure
7.3
Results of AMOS Prototype Simulation Runs
8. Conclusions and Recommendations
Appendix B. AMOS Survey Instrument - Appendix B is not available at this time.
Appendix C. AMOS Survey Data Bases
Appendix D. Evaluation Module and Acceptance Routine
List of Figures
Figure 4.1 AMOS Model
Figure
5.1 Activity-Mobility Simulator (AMOS)
Figure
5.2 Baseline Activity-Travel Analyzer
Figure
5.3 TDM Response Option Generator
Figure
5.4 Activity-Travel Pattern Modifier
Figure
5.5 Statistics Accumulator
Figure
6.1 Sensitivity of Neural-Network-Based TDM Response Probabilities to
Parking Charge Level
Figure 6.2
Relative Change in TDM Response Probabilities
Figure
6.3 MWCOG Data Flow
Figure 6.4 Age
Distribution in Sample
Figure 6.5
Distribution of Vehicle Ownership
Figure
6.6 Income Distribution in Sample
Figure
6.7 Distribution by Number of Commuters and Household Size
Figure
6.8 Distribution by Type of Residence
Figure
6.9 Distribution of Commute Travel Times
Figure
6.10 Distribution by Mode to Work
Figure
6.11 Distribution by Stops in Work Journey
Figure D.1 Evaluation Module and Acceptance
(Search Termination) Routine
List of Tables
Table 2.1 Sample of Recognized Issues Involved
in the Application of the Four-Step Procedure
Table
4.1 Output Variables for AMOS Modules
Table
6.1 AMOS Survey Completion Rates
Table
6.2 Average Household Characteristics
Table
6.3 Respondent Characteristics
Table
6.4 TDM Strategy Response Distributions
Table
6.5 Statistical Tests of Similarity of Response Distribution
Table
6.6 Congestion Pricing Response Distribution by Work Mode
Table
6.7 Congestion Pricing Response Distribution by Commute Distance
Table 6.8 Congestion Pricing Response
Distribution by Trip Chaining
Table 6.9
Congestion Pricing Response Distribution by Gender Role
Table
6.10 Bike/Pedestrian Facility Improvement Response Distribution by
Commute Distance
Table 6.11
Bike/Pedestrian Facility Improvement Response Distribution by Age
Table 6.12 Bike/Pedestrian Facility
Improvement Response Distribution by Sex
Table
6.13 Bike/Pedestrian Facility Improvement Response Distribution by
Number of Serve-Child Stops During Commute Trips
Table
6.14 Bike/Pedestrian Facility Improvement Response Distribution by
Number of Other Stops During Commute Trips
Table
6.15 Multinomial Logit Model of Response to TDM #2
Table
6.16 List of Initial Assumptions in the AMOS Prototype
Table
6.17 Baseline Travel Pattern for Case 1
Table
6.18 Baseline Travel Pattern for Case 2
Table
6.19 Baseline Travel Pattern for Case 3
Table
6.20 Baseline Travel Pattern for Case 4
Table
6.21 Baseline Travel Pattern for Case 5
Table
6.22 Predicted TDM Response Option for Five Cases
Table
6.23 Modified Travel Pattern for Case 1
Table
6.24 Modified Travel Pattern for Case 2
Table
6.25 Modified Travel Pattern for Case 3
Table
6.26 Modified Travel Pattern for Case 4
Table
6.27 Modified Travel Pattern for Case 5
Table
6.28 Changes in Travel Characteristics for Five Cases
Table
7.1 Baseline Travel Characteristics
Table
7.2 AMOS Simulation Results: Parking Pricing (TDM #1)
Table
7.3 AMOS Simulation Results: Parking Pricing with Employer-Supplied
Commuter Voucher (TDM #4)
Table 7.4
AMOS Simulation Results: Congestion Pricing (TDM #5)
Table
7.5 AMOS Simulation Results: Synergy Combination of Parking Pricing and
Congestion Pricing (TDM #6)
