Accessibility and location choice: Innovations in measurement and modeling.

摘要

Transportation planning has traditionally focused on travel demand forecast models that assume land use as exogenous inputs and ignore the feedback effects between changes in the transportation system and urban development. Research advancements in this arena have been shifting away from these narrow views towards large-scale integrated transportation and land use micro-simulation systems. The increasing recognition of the transportation-land use feedback relationship provides numerous timely research opportunities. This dissertation explores two such active areas of research, accessibility and residential location choice, and presents innovations in measurements and modeling in a series of three related papers. The first paper presents a parcel-level measure of public transit accessibility to destinations that explicitly considers the spatial, transportation, and temporal dimensions of accessibility, and uses disaggregate land use and transit schedule information. The second paper shows an operationalization of a time-space prism accessibility measure in a building-level framework and explicitly examines the influence of non-work accessibility at both the local- and person-level. The final paper presents a novel nested logit model of residential mobility and location choice with sampling alternatives and a proposed procedure for sampling bias correction.