【讲座名称】
Scalable data fusion for generating disaggregate activity schedules
【主讲人】
Prateek Bansal 助理教授 新加坡国立大学
【主讲人介绍】
Dr. Prateek Bansal is a Presidential Young Professor at the National University of Singapore. He leads the Behavioural Computational Science Lab and serves as a co-principal investigator of the Adaptive Mobility module at Future Cities Laboratory Global. His research focuses on mobility behavior, emerging transport technologies, and urban-scale behavioral modeling. He has published more than 80 journal articles in leading transportation and interdisciplinary journals, and currently serves as Associate Editor of Transportation Research Part A and Transportation Research Part B.
【讲座内容简介】
Understanding individual-level mobility is essential for transport and urban planning, but existing datasets often face a trade-off between behavioral detail and spatiotemporal coverage. This study proposes a two-stage data fusion framework that combines household travel surveys with passively collected mobility data. First, a cluster-based encoder–decoder model aligns demographic, behavioral, and spatiotemporal features across datasets. Second, a context-aware Markov process generates realistic daily activity schedules. Applied to Singapore and Seoul, the framework improves behavioral realism, corrects survey biases, greatly expands spatiotemporal coverage, and provides high-quality demand-side data for adaptive transport and urban planning.
【时间】
2026年6月11日（周四） 上午10:00
【地点】
犀浦校区交运楼304会议室
主办：研究生院 交通运输与物流学院
承办：学院学工组 交通运输科技协会 交通工程系
邀请人：翟国聪老师
