Shift Scheduling for Electric Service Vehicles: A Public Transportation Case Study

As global efforts to mitigate climate change intensify, the electrification of road transportation has emerged as a key solution among strategies for reducing greenhouse gas emissions (GHG). For public transportation agencies such as the Société de transport de Montréal (STM), this transition extends beyond passenger buses to include the fleet of vehicles dedicated to maintaining bus service continuity (referred to as service vehicles). A significant challenge in this transition is the substantial capital investment required for electric vehicles, necessitating optimal fleet sizing to minimize costs while maintaining the same quality of service. This strategic problem is strongly related to the tactical issue of shift scheduling where shifts of vehicles are to be determined with variable start times and possible overlaps. This paper focuses on the integration of shift scheduling with the deployment of electric service vehicles in the context of a large metropolitan transit network. We propose a Mixed-Integer Linear Programming (MILP) framework to minimize the fleet size when deploying and scheduling electric vehicles shifts while ensuring charging feasibility based on forecasted demand. Numerical experiments demonstrate that small and medium-sized instances can be solved to optimality with a commercial solver. For large-scale instances, the model provides optimal or near-optimal solutions within a 30-minute computational limit. These preliminary results confirm the framework's effectiveness as a potential robust decision-support tool for assessing trade-offs between investment costs and operational efficiency, thus supporting the achievement of strategic sustainability goals in large urban transit environments.