📝 SummarXiv

Abstract

The rapid electrification of city bus fleets offers significant environmental benefits, including reduced greenhouse gas emissions and air pollution. However, it also introduces complex challenges in energy management and infrastructure planning for public transport operators (PTOs). This study develops a novel mixed-integer linear programming (MILP) approach to minimize daily operational costs for electric bus (EB) networks. The model integrates on-site photovoltaic (PV) generation, energy storage systems (ESS), and Vehicle-to-Grid (V2G) capabilities, while explicitly accounting for dynamic electricity tariffs, peak demand charges, and battery degradation costs. A discrete-event optimization (DEO) scheme is employed to balance computational efficiency with operational accuracy. The framework is applied to a real-world case in Brussels involving 28 articulated electric buses (EBs) and 232 trips over a 24-hour horizon. A scenario-based analysis is conducted to evaluate the impacts of the extended components. Key findings show that incorporating demand charges into the optimization reduces daily costs by 5% and decreases the share of peak power costs by 9%, underlining the importance of load management. Integrating PV and ESS leads to a total net cost reduction of up to 56%, with ESS primarily used for energy arbitrage rather than direct bus charging. V2G participation is highly sensitive to battery degradation costs and policy incentives: it can become economically viable under high tariff margins and decreased replacement costs. When all extensions are combined, the model achieves a 58% reduction in total operational expenses compared to the baseline, highlighting the substantial value of smart (dis)charging optimization tools for PTOs.