2019/08/28 by Ítalo Santana, Santana, Ítalo, Plastino, Alexandre +1
Engineering · #Advanced Manufacturing and Logistics Optimization #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Optimization and Packing Problems #Vehicle Routing Optimization Methods
paper · pdf · doi:10.48550/arxiv.1908.10705
openalex publication_date 2019/08/28 · openalex created_date 2022/07/13 · openalex updated_date 2026/07/28
Recently, hybrid metaheuristics have become a trend in operations research. A\nsuccessful example combines the Greedy Randomized Adaptive Search Procedures\n(GRASP) and data mining techniques, where frequent patterns found in\nhigh-quality solutions can lead to an efficient exploration of the search\nspace, along with a significant reduction of computational time. In this work,\na GRASP-based state-of-the-art heuristic for the Minimum Latency Problem (MLP)\nis improved by means of data mining techniques for two MLP variants.\nComputational experiments showed that the approaches with data mining were able\nto match or improve the solution quality for a large number of instances,\ntogether with a substantial reduction of running time. In addition, 88 new cost\nvalues of solutions are introduced into the literature. To support our results,\ntests of statistical significance, impact of using mined patterns, equal time\ncomparisons and time-to-target plots are provided.\n