Advanced Activity-Based Models in Context of Planning Decisions

Abstract

Travel demand modeling today is undergoing a transition from the conventional four-step models to a new generation of advanced activity-based models. The new generation of travel models is characterized by such distinctive features as the use of tours instead of trips as the base unit of travel, the generation of travel in the framework of daily activity agendas of individuals, and the use of fully disaggregate microsimulation techniques instead of the aggregate zonal calculations. Although the theoretical advantages of activity-based models—in particular, behavioral realism and consistency across all travel dimensions—are well known, the practical advantages in the context of planning decisions have rarely been discussed and documented. Experiences to date are summarized for application of activity-based models for various planning purposes in metropolitan regions of New York City; Columbus, Ohio; Atlanta, Georgia; San Francisco, California; and Montreal, Canada. The focus is on the practical planning questions and policies that were analyzed with these models and their relative strengths and advanced features compared with the four-step models. The planning questions and policies include congestion pricing schemes, high-occupancy-vehicle facilities, parking policy, testing impacts of demographic scenarios, and so on. It is shown that activity-based models are capable of treating these planning and policy issues at the level at which four-step models become inadequate.