The study

A delivery company does not choose a route in a vacuum. It chooses whether to send an electric or fuel vehicle, how to schedule charging around trips, which depot should serve each customer, and what to do when demand changes during the day. This study places those decisions in one multi-depot, mixed-fleet routing problem.

What is connected in the model

• Electric and fuel vehicles can both serve the delivery system.
• Vehicles may complete multiple trips, with charging tied to the actual trip schedule.
• Electricity cost and charging emissions depend on when electricity is used and the grid's time-varying carbon intensity.
• Customer demand can change after the original plan has been made.
• Multiple depots can plan together, with shared costs allocated between them.

How the search works

The paper minimizes total delivery cost, including vehicle activation, travel, charging, fuel, and carbon costs. It develops a hybrid genetic search with vehicle-type exchange and carbon-intensity-guided charging. The model and method are evaluated through benchmark instances, component ablation, and a Beijing simulation case.

Three moments worth following

• A cleaner charging hour does not automatically lower emissions from the entire mixed fleet.
• Two depots can use fewer vehicles and lower total cost even when the combined route distance grows.
• When new orders arrive, reconsidering unfinished work can outperform inserting orders one by one in the tested event sequence.

Paper title: Complex Mixed-Fleet Logistics Model and Solution Algorithm under the Dual-Carbon Background. Authors: Zhou Leixishu, Gan Hongcheng, and Qiu Yingying, School of Management, University of Shanghai for Science and Technology.