A Data-Driven Approach for Spatial-Temporal Analysis of Travel-Related Activity Changes during Disasters

Disasters affect the connectivity of transportation infrastructure systems and the patterns of travel-related activities. Because humans and transportation systems interact, when they get affected, their interaction could exacerbate disaster impacts. There is, thus, a need to analyze and predict how travel-related activities change in disasters to pave the way toward human-infrastructure interaction modeling. As such, this paper proposes a data-driven approach for spatial-temporal analysis of changes in travel-related activities to support activity prediction in disasters. The proposed method includes two steps: (1) extracting information about travel-related activities from the American Time Use Survey (ATUS), and (2) utilizing a difference-in-differences method to analyze the activity duration changes. The proposed method was implemented in analyzing travel-related activities during the COVID-19 pandemic. The implementation results revealed how travel-related activities changed across categories, time, and space. As a preliminary study, this paper focuses on presenting the method and discussing the results from its implementation.