Xuesong Zhou Research Group

Arizona State University (Tempe)

How can multimodal transportation planning algorithms be made scalable, accurate, and deployable for large metropolitan systems? The ASU Transportation+AI Lab develops methods for dynamic traffic assignment, traffic estimation and prediction, large-scale routing, and rail scheduling, and implements open-source packages like DTALite, NEXTA, and OSM2GMNS. Projects couple differentiable programming, layered computational graphs, and cross-resolution performance models to integrate traffic state estimation with queue and routing models. Researchers use optimization, nonlinear programming, integer programming, and distributed computing to solve time-dependent routing and timetabling problems for transit and rail. The lab documents deployments and downloads of its open-source tools used by metropolitan planning agencies and state DOTs.