The Travel Model Improvement Program (TMIP) is a multi-year, multi-agency program sponsored by USDOT and EPA, with the mission of supporting and empowering planning agencies through leadership, innovation and support of travel analysis improvements, to better meet current and future mobility, environmental, safety and security goals. TMIP began operations in 1992, and has three goals:
One specific challenge that the TMIP is helping DOTs and MPOs address is developing methods for estimating the land use impacts of transportation projects. Traditionally, a variety of methods have been used for this purpose, ranging from planning judgement and other qualitative techniques, to complex urban land use models. Models have the advantage of rigor; they can incorporate and analyze a large amount of information, which is important given the increasingly complex and dynamic relationship between land use and transportation. However, land use models are relatively expensive to use and their results are not always widely accepted by stakeholders, in part because models are sometimes viewed as a "black box", and decision makers and the public do not always trust the results. Planning analysts and development experts can sometimes provide a more nuanced and realistic understanding of a given land use and transportation context, but may be perceived as insufficiently objective or rigorous. Thus, there is a need to identify analytical methods that combine the rigor and objectivity of land use models with the sophisticated realism of analysts and experts.
Expert panels provide these strengths, and have been used with substantial success in varying contexts throughout the United States and the rest of the world. An expert panel is a diverse group of individuals with access to current, high quality information relevant to a given land use and transportation context. In the case studies reported here, expert panels generally utilized some variation of the Delphi Method, whereby the group works within a structured process to respond to two or more rounds of questions in order to develop forecasts of land use impacts, typically employment, population, and housing distributions. The Delphi Method typically involves each panelist developing analysis independently, and then sharing that analysis with the group so that the panel as a whole can report out a single set of findings1. Within these fairly broad parameters, there are important variations in how any one expert panel may be structured and how it may function. Thus, there is a need for better understanding of the ways that expert panels may be used, their characteristics, and the issues associated with their use.
