Trucks Allocation in Logistics Service Providers' Depots within a Synchromodal Transport Network: A Bayesian Optimization Approach

Abstract

Effective management of resources is a significant challenge in the logistics and freight transport sector. This gets more challenging within synchromodal transport systems, which are characterized by high levels of uncertainty due to real-time decision-making. A primary goal in these inherently flexible systems is to optimize the allocation of truck fleets across depots to enhance operational efficiency and maximize profitability. In this work, we propose a Bayesian optimization approach to optimize the truck-depot allocation strategy for logistics service providers in a synchromodal network in order to maximize their profit. This approach minimizes the need for extensive simulation runs to identify the optimal solution. Instead, it efficiently explores and exploits the decision space, converging to the best outcome within a limited computational budget. The findings underscore the effectiveness of this approach in improving decision-making within a synchromodal transport network, offering a robust and computationally efficient solution for truck allocation problems. This study advances the literature by tackling the vehicle allocation problem in synchromodal networks and by integrating advanced optimization techniques with the concept of synchromodality. In doing so, it supports the development of more resilient and efficient freight transport systems.