Joint optimization of charging infrastructure and fleet mix for CO₂-constrained feeder services

This study addresses the electrification of demand-responsive feeder services, a form of public transport designed to connect rural and low-demand areas to mass transit hubs. Electrifying demand-responsive transport requires planning the charging infrastructure carefully, considering the trade-offs of charging efficiency and charging infrastructure costs. This study addresses the joint planning of fleet size and charging infrastructure for a demand-responsive feeder service under stochastic demand, given a user-defined CO2 emissions reduction policy. We propose a bi-level optimization model where the upper-level determines charging station configuration given stochastic demand, and the lower-level solves a mix fleet feeder (first and last mile) service routing problem under the CO2 emission and capacitated charging station constraints. An efficient deterministic annealing metaheuristic is proposed to solve the CO2-constrained mixed fleet routing problem. The metaheuristic solves up to 500 requests within 3 min, demonstrating the practical applicability of the proposed solution. We applied the model to a real-world case study in Bettembourg, Luxembourg, with two types of electric minibuses and gasoline ones, under different CO₂ reduction targets considering rapid (125 kW) and super-fast (220 kW) chargers, given 200 requests per day. The results show that using 24-seat minibuses leads to significant cost savings (−49 % on average) compared to that of 10-seat minibuses. Due to their larger battery capacity, charger availability has a smaller impact on the operational costs of 24-seat minibuses. The proposed method provides a flexible tool for joint charging infrastructure and fleet size planning.