Sensitivity to Four-Step Versus Activity Based Models to Transportation System Changes

The new travel demand forecasting models deployed in the Mid Ohio Regional Planning Commission area (MORPC) and for the statewide model employ new activity based modeling techniques. These techniques are a radical paradigm shift which seeks to micro-simulate behavior of individuals rather than employing statistical averages, as in the traditional four step travel demand forecasting models. While the national research community is almost unanimously behind this paradigm shift as a way of providing better forecasts by getting at the root causes of how and why people travel, to date there has been little, if any, research proving this assertion. In addition, while conceptually simple, in practice the new approaches are more complicated and difficult for practitioners to understand and because they are new, have so far been more expensive to develop. The models also require significant computer resources and run times thereby degrading turn around time. This research seeks to quantify the potential benefits of the new models in terms of forecasting sensitivities by conducting before and after studies of a variety of projects employing both the new and old models. A unique opportunity exists in that MORPC currently maintains both its new and traditional models. This situation will not last much longer which makes the research urgent. It is hoped that through this research, the Department can determine in what situations, if any, this new modeling paradigm should be pursued in the future versus using traditional techniques. If the research is not conducted, we will lose the opportunity to evaluate these new techniques and be left with the same questions in the future regarding what modeling techniques to employ when updating travel demand forecasting models. Research task for this project are as follows. Task 1: Preliminary Analysis; Task 2: Determine Detailed Analysis Methodologies; Task 3: Selection of Study Projects; Task 4: Determine Data Collection Program; Task 5: Prepare Models and Model Data; and Task 6: Analyze Data and Prepare Final Report and Recommendations. The deliverable will be a final report containing statistical comparisons of the two modeling techniques versus actual conditions. This should also contain recommendations on when, where and why each method might be preferred over the other. Tasks 1-4 will also contain brief technical memoranda to aid the Department in making critical decisions about subsequent tasks.

Language

  • English

Project

  • Status: Completed
  • Funding: $183150.00
  • Contract Numbers:

    134368

  • Sponsor Organizations:

    Ohio Department of Transportation

    Research Program
    1980 West Broad Street
    Columbus, OH  United States  43223
  • Project Managers:

    Giaimo, Greg

  • Performing Organizations:

    University of Texas at Austin

    Austin, TX  United States  78712
  • Principal Investigators:

    Bhat, Chandra

  • Start Date: 20080616
  • Expected Completion Date: 0
  • Actual Completion Date: 20101130
  • Source Data: RiP Project 18086

Subject/Index Terms

Filing Info

  • Accession Number: 01464449
  • Record Type: Research project
  • Source Agency: Ohio Department of Transportation
  • Contract Numbers: 134368
  • Files: RiP, STATEDOT
  • Created Date: Jan 3 2013 2:42PM