Similar outcomes, different paths: Tracing the relationship between neighborhood-scale built environment and travel behavior using activity-based modelling

The relationship between the built environment and travel behavior has long been a hot research topic. However, most existing research focus on aggregate measures of the patterns of daily travel (e.g. vehicle miles travelled, total distance travelled), while the mechanisms and processes that lead to these outcomes are much less researched. We therefore propose an activity-based model that simulates people's detailed decision making in daily travel and the influences of the neighborhood-scale built environment throughout the process, using Beijing as the study case. More specifically, it deals with eight activity types, eleven most common activity plans, 652 traffic analysis zones, six time slots and four travel modes. The model is then applied to simulate how people's travel behavior would change in response to changes in the built environment. The simulation is implemented by changing the built environment conditions. The major advantage of the model lies in its ability to 'trace back' the influence of the built environment on aggregate travel outcomes to detailed travel aspects, such as activity frequency, travel distance for various purposes, mode choice, etc. The simulation results show that although compact design generally reduce VMT, the mechanisms of influence by different built environment variables vary, e.g. through a major impact on the mode choice or through distributed impacts on several travel aspects.