Adaptive large neighborhood search for drayage routing problems involving longer combination vehicles
Daniel Bustos-Coral,Alysson M. Costa
DOI: https://doi.org/10.1016/j.cor.2024.106826
IF: 5.159
2024-09-03
Computers & Operations Research
Abstract:Longer combination vehicles (LCVs)—tractor-trailer combinations with two or more trailers—can move more cargo per trip than standard trucks, potentially reducing costs and emissions. Acknowledging this, the drayage industry has recently launched initiatives to increase the adoption of LCVs, aiming to mitigate supply chain issues. This paper aims to support this kind of initiative by tackling drayage routing problems involving LCVs and further aspects, such as heterogeneous truck fleets, containers of any size and cargo category, and compatibility constraints based on the containers' sizes and loads. As solution method, we propose an adaptive large neighborhood search (ALNS) heuristic, whose search operators account for aspects such as truck/container compatibility and load configurations specific to each truck. The proposed ALNS heuristic also incorporates a novel acceptance criterion that seeks to escape local optima while allowing for revisiting the current best solution and avoiding search stagnation. Through a series of benchmark tests against a state-of-the-art exact approach, we show that the proposed ALNS heuristic can consistently find high-quality solutions for a wide range of instances while, in some cases, cutting runtimes from hours or even days to a few minutes. We also show that the proposed acceptance criterion enables improved performance. Finally, we use the proposed ALNS heuristic to derive managerial insights into the savings delivered by different types of LCVs in drayage operations.
computer science, interdisciplinary applications,engineering, industrial,operations research & management science