Truck Scheduling: A Case Study in the Automotive Sector

Abstract

Logistic companies face several problems in their services, leading to increased time and costs in their operations. These issues could be mitigated if decision-making were based on models that consider resource optimization and respect process constraints. An example of this problem is the queues generated by trucks at docks for loading and unloading materials in the automotive sector. Therefore, this work studies truck scheduling with the inclusion of forbidden time windows for breaks. The aim is to propose a mathematical model in order to minimize the time spent by drivers within companies and to present exact and approximate solutions with the use of Operational Research techniques and algorithms. Methods and algorithms that adapt to this type of problem were identified. Then, the mathematical model was developed, computationally implemented in the LINGO and Gurobi software, and validated according to the case study conditions. A metaheuristic based on the Simulated Annealing algorithm was implemented in Visual Basic for Applications language, and used as a form of approximate resolution for problems larger than the ones solved optimally. Computational experiments were conducted to evaluate the performance of the model and metaheuristic proposed. The GAP between the metaheuristic solution and the optimal solution is less than 4% for all problems tested. For larger problems, with more than 10 trucks and 5 docks, Gurobi optimization software is not able to find the optimal solution in a feasible time. However, the metaheuristic is able to find good solutions for bigger problems in a reasonable time. Therefore, results proved the performance of the mathematical model, indicating that truck scheduling is an important tool for automotive companies, as it can help to improve efficiency and productivity.