Comparison Between Traditional Four-Step Activity-Based Travel Demand Modeling - A Case Study Of Tampa, Florida

摘要

The activity-based travel demand model has been viewed as an advanced approach with higher fidelity and better policy sensitivity. This study aims to compare modeling results from an activity-based model (ABM) developed using the travel diary data collected in the Tampa Bay Region with an existing traditional four-step model - the Tampa Bay Regional Planning Model (TBRPM), based on the four sequential steps: trip generation, trip distribution, model split, and trip assignment. The comparison results show salient differences. Trip production rates calculated from the travel diary data are found to be either double of or a quarter less than the TBRPM. On the other hand, trip attraction rates computed from ABM are found either more than double of or one tenth less than TBRPM. The trip distribution curves from the two models are found similar, but the peaking of travel time is different, with 10 min for the TBRPM, but 15 min for the ABM. Mode split analyses show that the TBRPM may underestimate the driving trips and it cannot capture the usage of other alternative modes, such as taxi and non-motorized. In addition, the ABMs are found to be less capable of reproducing observed traffic counts when compared to the TBRPM, most likely due to not considering the external and through trips. The comparison results presented can help transportation engineers and planners better understand the strengths and weaknesses of the two types of models and this will subsequently assist decision-makers to choose a better modeling tool for their planning initiatives.