Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
142 Innovative Methods for Pricing Studies Arun R. Kuppam, Cambridge Systematics, Inc. Maren L. Outwater, Cambridge Systematics, Inc. Rob C. Hranac, Cambridge Systematics, Inc. In a recent forum on road pricing (1), attendees dis-cussed limitations with current travel demand fore-casting approaches for pricing studies. In addition, Cambridge Systematics, Inc. (CS) recently completed a paper on the limitations of studies used to advance toll projects (2) and on the opinions of Washington Stateâs community leaders (3). Based on these sources and recent experience in developing forecasting models for toll projects, the authors have identified the following issues as important to improving existing travel models for pricing studies: inaccurate values of time for specific travelers, trip purposes, modes, and time periods; and lack of temporal detail and behavioral choice for time- of-day models. CSâs approach to advance travel models for pricing studies focuses on these issues as the most critical to be addressed in existing models. The authors have been involved in the development and application of these methods for trip-based models in Minnesota and Wash- ington, as well as for activity-based models in San Fran- cisco. The remainder of this paper describes innovative methods to incorporate advances to address these issues. In addition, the authors describe strategies to optimize tolls for pricing studies. Finally, more research is proposed to address additional limitations of existing models. VALUES OF TIME The estimation and application of the value of time in travel demand forecasting models is the most often cited problem for evaluating pricing projects. There are a number of issues that present challenges, including how to distribute values of time: ⢠Across individual travelers (i.e., with different income levels); ⢠Across different trips (i.e., with different purposes and modes); ⢠Across different destinations (i.e., trips to the air- port); ⢠Across different vehicle types (i.e., with different vehicle classes); ⢠Based on the types of goods being carried for truck trips; and ⢠For different types of congestion (i.e., recurring and nonrecurring congestion, such as accidents). In a disaggregate travel demand forecasting system, these values of time could be set based on the traveler, the trip, the vehicle type, and the goods being carried and could remain consistent throughout the forecasting process, eliminating the application-related issues sur- rounding the values of time. At this time, most travel demand forecasting models are aggregate trip-based mod- els, which makes the distribution of values of time for indi- vidual travelers, trips, and vehicles impossible. For these models, the only solution is to identify specific categories of travelers, trips, and vehicles and apply values of time for these categories. This is an effective means of distrib- uting values of time within the forecasting system. However, these trip-based modeling systems do not necessarily contain the same market segmentation throughout the system (i.e., to assess values of time by