Real-world Truck Scheduling for Cross-dock Operations : an alternative solution approach

Owing to the growing market in transportation networks and the mostly truck’s arrival in peak hours, trucks have to wait for available docks in a crossdocking platform (called also a cross-dock); this leads to increased congestion at the gate. Every time point where trucks have to wait is a waste for cross-dock yield. The congestion causes in turn planning problems in two ways. First, trucker has to do more than one shift per day. Waiting for his first task means that it will be too late against his workload per day. Second, trucking companies have to lose more of time to waiting in each node of transportation network so the arrival time of trucks is uncertain at each destination. One option to spread the workload is the use of truck appointment system. In this paper, we focus firstly to expose a state-of-the-art map of scientific research based on the problem type. We propose also a model to derive a robust plan to attenuate the impact of uncertain arrivals in a cross-dock. The objective is to dimension a planning horizon and to link adjacent planning runs using an ATN network and multi-agent system. Lastly, we implement a first prototype to test and to validate the proposed approach using a JADE platform. keywords: Logistics, Cross-dock, Real-world scheduling, Classification, Multi-agent system, Truck appointment system.