Resumen de Trade-off optimization of a location-routing problem involving open routes and flexible fleet: a case study in Guadalajara, Mexico

Joel Novi Rodríguez Escoto, Samuel Nucamendi Guillén, Elias Olivares Benítez

• This study investigates a location-routing problem with self-supply, open routes, and a fixed fleet, focusing on the trade-off between total traveling costs and vehicle contracting costs. A bi-objective approach is developed, comparing the impact of minimizing traveling costs versus vehicle contracting costs, which is the cost associated with vehicle selection. The model is solved using three multi-objective methods: improved augmented epsilon constraint, weighted revised multi-choice goal programming, and chebyshev-based method, to determine the most suitable for the proposed approach. The methods are tested on three instances, two with homogeneous fixed fleets and one with a heterogeneous fixed fleet. Four performance metrics are used to compare the methods, and a real-world case study in Guadalajara, Mexico, is also solved. The results show that the augmented epsilon constraint method outperforms the other methods, improving on average up to 40% in NPS, 50% in CPU time, and 30% in performance metrics. The study also finds that homogeneous instances have, on average, up to 87% unused utilization capacity, while heterogeneous instances have 26% unused capacity. The economic analysis shows that reducing transport costs, on average, by 40% requires a 200% increase in hiring costs for homogeneous instances and a 60% increase for heterogeneous instances to obtain a reducing transport cost of 30%. A bi-objective approach achieves substantial cost savings of over 18% for the company.