Air cargo network planning and scheduling problem with minimum stay time: A matrix-based ALNS heuristic

Abstract

In air cargo delivery, cargo usually needs to stay in hubs after unloading operations until the designated outbound flight is ready to leave. The stay time may account for a significant part of the overall delivery time and is closely related to the service level of the cargo delivery. However, the prior studies on the air cargo network planning and scheduling problem mainly focused on the cost and revenue but paid little attention to the stay time. To bridge this gap, we introduce a new facet to the air cargo network planning and scheduling literature. The air cargo network planning and scheduling problem with minimum stay time involves the optimization of the hub location, flight deployment, flight timetabling, and flight-sequence-based cargo distribution to minimize the total stay time of freighter cargoes while guaranteeing a rational cost level. We develop a hierarchical modeling framework to handle this problem, where a time–space network is built to model the cargo stay time and the flight departure time. Next, to efficiently address the framework, the concept of the capacity matrix is introduced for model reformulation, and a matrix-based ALNS heuristic is proposed for model solving. The heuristic designs eight destroy operators from the column, row, and element perspectives, and develops a repair model for the repair operation. Finally, the proposed framework and algorithm are tested with a real-life instance from a leading express company in China.