EURO 2025 Leeds
Abstract Submission

1769. An actor-critic model for real-time heavy-duty electric vehicle charge scheduling with flexible pricing at Community Charging Hubs

Invited abstract in session TD-59: Innovations for a Greener City and Sustainable Urban Logistics II, stream Transportation.

Tuesday, 14:30-16:00
Room: Liberty 1.14

Authors (first author is the speaker)

1. Alexander Rose
Chair of Logistics Management, WHU Otto Beisheim School of Management
2. Stefan Spinler
Kuehne Foundation Endowed Chair in Logistics Management, WHU - Otto Beisheim School of Management

Abstract

Achieving climate targets for road transport depends on the rapid adoption of heavy-duty battery electric trucks and buses. Ensuring the economic viability of this transition will require cost-competitive Community Charging Hubs, coordinated and shared between commercial neighbors. At these hubs, vehicle arrival times, charging demands, electricity rates and renewable energy potentials are subject to stochastic uncertainties, requiring real-time dynamic algorithms and predictability through reservation options.
We extend existing research by applying an actor-critic model to jointly optimize the charging schedule and pricing for each period under uncertainty, considering both reservations and spontaneous charging demands. Our actor decides on charging power and pricing actions through reinforcement learning with proximal policy optimization, thus handling the high-dimensional, continuous action space of trucks, buses and municipal vehicles charging at different charger types. Our critic learns the charging state value function to evaluate and update the actor’s policy parameters, also considering dynamic pricing schemes to incentivize user flexibility in charging schedules.
We model uncertainties using real-world telemetry data and electricity rates for a charging hub in Switzerland. Our results show a 50 percent reduction in grid load and charging rates compared to first-come, first-served and priority-based charging algorithms.

Keywords

Status: accepted

Back to the list of papers