Truck pooling and scheduling in post-distribution cross-docking with JIT demands and synchronized interchangeability

TL;DR

This paper introduces a novel cross-docking optimization problem combining truck scheduling with load and destination assignment, incorporating JIT demands, and proposes a hybrid matheuristic approach validated by extensive computational experiments.

Contribution

It presents a new integrated mathematical model and a hybrid solution method for a complex cross-docking problem with JIT demands and product-truck-destination assignment.

Findings

Hybrid matheuristic effectively solves large instances.

JIT demand integration improves scheduling efficiency.

Symmetry breaking constraints enhance model performance.

Abstract

Various operational optimization problems arise in cross-dock synchronization. The combination of the truck scheduling with other decision problems in cross-docking has been targeted by researchers in the recent years. In most of the truck scheduling researches, the load and the destination of the outbound trucks are predetermined. This paper presents a novel cross-docking optimization problem in which the truck scheduling is combined by two assignment problems: the assignment of received loads and the assignment of destinations to outbound trucks (product-truck-destination allocation). Moreover, a Just-In-Time (JIT) strategy is imposed on the destination demands, whereas in most of the previous researches the time windows are imposed to the trucks rather than the demands. An integrated mathematical model is presented for this cross-docking problem. The mathematical model is reinforced by two categories of symmetry breaking constraints. A matheuristic approach, including two mixed-integer mathematical models, is proposed. An adaptive heuristic is presented to solve the first step of the matheuristic algorithm instead of the mathematical model. The solution approaches are analyzed by the extensive computational experiments. The numerical results show the efficiency of the hybrid matheuristic to solve the real size instances of the studied problem.