Optimizing timetable and network reopen plans for public transportation networks during a COVID19-like pandemic

Abstract: The recovery of the public transportation system is critical for both social re-engagement and economic rebooting after the shutdown during pandemic like COVID-19. In this study, we focus on the integrated optimization of service line reopening plan and timetable design. We model the transit system as a space-time network. In this network, the number of passengers on each vehicle at the same time can be represented by arc flow. We then apply a simplified spatial compartmental model of epidemic (SCME) to each vehicle and platform to model the spread of pandemic in the system as our objective, and calculate the optimal open plan and timetable. We demonstrate that this optimization problem can be decomposed into a simple integer programming and a linear multi-commodity network flow problem using Lagrangian relaxation techniques. Finally, we test the proposed model using real-world data from the Bay Area Rapid Transit (BART) and give some useful suggestions to system managers.