Integrating estimated time of arrival in truck scheduling: a real time-reactive model and application

Logistics 4.0 exploits cyber-physical systems to improve logistics processes through real-time informed decision making enabled by data analytics. Despite the great interest around the topic, the application of Logistics 4.0 solutions still remains blank space for many processes. We present a new approach that integrates one of these solutions – the prediction of trucks estimated time of arrival (ETA) – into truck scheduling, to solve truck arrival time uncertainty issues. The approach exploits ETA information to re-schedule trucks in real-time, according to the actual delays, and aims to minimize the sum of trucks waiting times along the day. The approach relies on an optimization model that is formulated as a mixed-integer programming problem, and solved with a genetic algorithm heuristic. We demonstrate the use of our approach in a numerical case study, were we evaluate it against the first come first serve (FCFS) truck scheduling rule. Compared to FCFS, the new approach has most value when ETA accuracy increases.