Mixed-Integer Programming Strategies for Reducing Rehandling in Multi-Customer Distribution Operations

Abstract

In the current context of global logistics, the optimization of operational costs is essential to maintain the competitiveness of supply chains. In particular, decisions related to load planning and unloading operations in multi-drop distribution systems play a crucial role in overall operational performance. In this regard, this paper addresses the Multi-Drop Container Loading Problem, integrating critical variables such as customer delivery sequences and the dimensional characteristics of the items. A mixed-integer programming model is proposed to efficiently organize the load, minimizing unnecessary rehandling during the unloading process through the incorporation of a last-in, first-out (LIFO) constraint. In addition, an auxiliary mathematical model is introduced to adjust the initial solution and ensure a physically coherent and stable load configuration. The proposed methodology is validated through case studies implemented in AMPL and Python Colab, demonstrating that the proposed approach not only optimizes the use of available space but also facilitates faster and more orderly unloading operations.