Monday, September 8, 2014

Integrating Multimodal Data into Benefit-Cost Analysis for Transportation Planning and Public Policy


A draft manuscript describing my latest transportation research is now available.  The title is, "Integrating Multimodal Data into Benefit-Cost Analysis  for Transportation Planning and Public Policy

In this paper, we explore two types of BCA models.  The first type we refer to as models for public policy analysis.  This is the type of analysis I am familiar with from teaching BCA to graduate and undergraduate students.  We refer to the second type of models as models for planning.  These are used, for example, at state Departments of Transportation.  I was not very familiar with these types of models before beginning this research, but I have learned a good deal about them in the course of writing this report. 

I am very happy with what we were able to produce and I will seeking feedback on it over the next couple of weeks, as this report goes through peer and editorial review.  Please email me with any questions or comments!

The abstract is below:



Abstract

Federal, state and local governments allocate billions of dollars in transportation funds each year.  One useful tool for helping to decide which projects are best investments is Benefit-Cost Analysis (BCA).  Ideally, BCA takes into account all impacts of a decision, and provides a way of selecting investments that maximize social welfare.  However, in practice even the best BCAs only measure select impacts.  This project develops methods of improving BCA by better integrating multimodal transportation data.  It considers both BCA for evaluating past policy decisions, and BCA for planning and programming future transportation investments.  We identify shortcomings of existing models, and propose, implement and evaluate concrete solutions.  Case studies in transportation planning focus on the California Department of Transportation (DOT), but benchmark California’s competencies by exploring methods used by other states and local governments.  In addition, while the focus is on BCA output as a concrete example of the type of performance measure that may suffer from data integration problems, we also consider other important models used by DOTs, especially travel demand models. The conclusion lists all recommendations for improving transportation planning through more integrated models.  These will have immediate use to Caltrans as it considers directions for developing new planning capabilities.  In addition by fitting the planning models we explore in the broader context of transportation planning and policy, this report will also serve as a valuable resource for analysts, managers and others who are interested in better understanding BCA methods and their use.