Subnetwork Analysis for Dynamic Traffic Assignment Models

Abstract

Dynamic traffic assignment (DTA) can be used to model impacts of traffic control plan scenarios on travel behavior. However, using DTA for modeling construction project impacts is limited by the computational time required to simulate entire roadway networks. DTA modeling of a subarea surrounding these work zones extracted from a larger network can decrease the overall run time. A particular issue of interest is estimating dynamic demand along the boundary of the selected subnetwork. Often demand at the subnetwork boundary is based on flows extracted from a full, base network analysis. However, it is likely that impacts within the subnetwork caused by traffic control plans will extend beyond the boundary and affect this demand. The use of a logit model to reassign demand at subnetwork boundary centroids on the basis of differences in internal (subnetwork) travel times between base and impact scenarios is presented. The proposed methodology was implemented by using several software programs linked through automated scripts. The method's implementation and the results from a case study are presented. Implementing the logit formulation was found to provide better estimates of subnetwork demand, specifically along the boundary, when compared with use of a fixed demand table extracted from the vehicle trajectories of the full network model under base conditions. Ultimately, the findings were encouraging and suggest that the strategy could enhance the accuracy of subnetwork demand estimation.