Large-Scale Analysis and Modeling of Mobility Changes under Remote Work in the United States

The widespread adoption of remote work has profoundly reshaped mobility patterns following the COVID-19 pandemic, yet a comprehensive, data-driven model capturing these dynamics remains limited. In this study, we analyze mobility data from approximately seven million individuals across the United States. Comparing mobility in 2019 and 2022, we find a marked reduction in both travel distance and trip frequency after the rise of remote work, accompanied by greater temporal irregularity and less predictable departure times from home. While office-bound mobility has declined, other trip types have increased, potentially enhancing urban vibrancy. Building on these empirical observations, we develop an early-stage model that jointly captures the temporal and spatial dynamics of individual mobility, drawing inspiration from activity-based frameworks and Markov chain. In this framework, agents seek boundedly optimal solutions that reduce total travel time while fulfilling recurring obligations such as work, sleep, and other daily needs. The model highlights how the interplay between spatial choices and temporal dynamics drives mobility changes in the post-remote-work era.