Comparing Microscopic Activity-Based and Traditional Models of Travel Demand: An Austin Area Case Study

摘要: Two competing approaches to travel demand modeling exist today. The more traditional “4-step” models rely on aggregate demographic data at a traffic analysis zone (TAZ) level. Activity-based microsimulation methods employ robust behavioral theory while focusing individuals and households. While currently not widely used in practice, many modelers believe that activity-based promise greater predictive capability, accurate forecasts, realistic sensitivity policy changes. Little work has examined detail the benefits of models, relative approaches. In order better understand tradeoffs between these two methodologies, this paper examines model results produced by both, an Austin, Texas application. Results reveal several differences performance accuracy. general, are sensitive changes inputs, supporting notion ignore important distinctions across population. However, they generally involve much calibration application effort, ensure synthetic populations match key criteria activity schedules surveyed behaviors, being consistent household members.