Spatial Microassignment of Travel Demand with Activity Trip Chains

Abstract

As activity-based approaches gain acceptance for travel demand analysis and forecasting in practice, the unit of analysis in transportation planning models needs to shift from a one-way trip to an activity schedule with associated trip chains. A spatial network assignment process and its application to an actual network are described; the process can be used in conjunction with activity-based travel demand forecasting procedures as well as for traffic operational studies. Two simulation-based dynamic spatial microassignment procedures for travel demand with activity trip chains are presented. The first represents a one-step simulation-assignment procedure in which assignment of trips is based on prevailing travel time. The second is an iterative simulation-assignment procedure in which a user equilibrium solution is obtained. The models are illustrated through various experiments conducted using an actual network (in Fort Worth, Texas) to examine network performance under different activity and trip-chaining scenarios. Different demand levels with different trip chain patterns are considered and compared for an actual transportation network. In addition, a case is made for dynamic microassignment of activity trip chains by contrasting it with current practice and illustrating the pitfalls of inappropriately recognizing trip chains using current-practice assignment. In particular, two different procedures are considered and compared for selected test scenarios with the dynamic assignment models presented. The first procedure corresponds to the case in which the activity and trip-chaining behavior is completely ignored, and only the final destination in the chain is considered. The second procedure treats the links of the chains separately as independent trips, without appropriate linkage between them.