Electric delivery fleets coordinate charging to avoid long waits
Emergent Charging Coordination in Electric Delivery Fleets
Machine LearningMultiagent SystemsNeural and Evolutionary Computing
Summary
Electric delivery vehicles need to recharge during their shifts without causing long lines at charging stations. The authors created a system where each vehicle makes its own charging decisions based only on local information, like how busy stations are. This local decision-making leads to smart, group-wide coordination without a central controller. Their approach works almost as well as a perfect planner and much better than simple rules, cutting wait times from 45 minutes to under 2.
What this means in practice
- •For electric fleet operators: Improve electric delivery fleet efficiency by enabling vehicles to self-coordinate charging without centralized scheduling or infrastructure support.
- •For logistics software developers: Embed learned local charging policies into fleet management software to reduce wait times and increase on-time shift completions in EV fleets.
Authors
Javier Vales-Alonso, Juan J. Alcaraz
Abstract
In electric delivery fleets, mid-shift charging is non-trivial: each vehicle must decide when, where and how much to charge to finish on time with battery above a safety floor. The choices are coupled: queues build where too many vehicles pick the same station. Prior work resolves this coupling with central dispatching, precomputed schedules or reservations, machinery that charging infrastructure rarely supports. Instead, we use a family of learning agents under purely local control: every vehicle runs the same policy, deciding alone from its time budgets and broadcast station occupancies, leading to emergent coordination without central control or messaging. We validate this paradigm in simulation on real OpenStreetMap networks of twenty cities, each with a frozen scenario calibrated by an omniscient Oracle (99.5% of shifts completed on time), whereas a naive greedy rule (nearest station on low battery) completes just 73%. Agents trained with neuroevolution (NEAT) and policy gradients (PPO) on four cities and deployed zero-shot across all twenty, sixteen never seen in training, complete 96.8% and 98.6% of shifts, with the policy-gradient controllers proving more robust when demand or vehicle characteristics drift beyond the trained regime. In contrast, tuned threshold heuristics that read vehicle urgency alone fall short in contended cities (~80%). Through training, these learning agents rediscover partial charging and short opportunistic sessions, and route around busy stations, cutting per-session queue waits from about 45 minutes to under 2. In summary, this coordination paradigm balances local urgency against public occupancy, reaching near-Oracle performance at minimal implementation cost.