Macro-, meso-, and micro-level validation of an activity-based travel demand model

摘要

Validation is an essential part of the model development process for practical applications inthe real world context. Spatial transferability has been recognized as a useful validation testfor travel demand models. To date, however, transferability of activity-based models has notbeen frequently assessed. This paper focuses on how and at which level the validation ofactivity-based travel demand models has been performed. More specifically, it examines thespatial transferability of an activity-based model, TASHA (Travel Activity Scheduler forHousehold Agents), which has been developed for the Greater Toronto Area (GTA), Canada.It applies TASHA to the Island of Montreal, Canada, using the 2003 Origin-Destination (O D)travel survey and the 2001 Canadian census, and validates the transfer by comparingmodelled and observed activity attributes from the 2003 O-D survey for five differentactivities (work, school, shopping, other, and return to home) at three different levels ofaggregation, macro-level (aggregation of the entire population), meso-level (aggregation bypopulation segments by age group and gender, and by home location), and micro-level(individuals). In most of the cases, simulated results are significantly different than theobserved activity travel patterns of the individuals of this region. The large variations foundat different levels most likely indicate differences in behaviours between Montreal andToronto. We conclude that re-estimation of model parameters and the use of local activityattribute distributions (frequency, start time and duration) is a desirable step in the transfer ofthe TASHA model from one context to another.