An algorithm to compute time-balanced clusters for the delivery logistics problem. (May 2022)

• Record Type: Journal Article
• Authors: Menchaca-Méndez, Adriana; Montero, Elizabeth; Flores-Garrido, Marisol; Miguel-Antonio, Luis
• Is Part Of: Engineering applications of artificial intelligence. Volume 111(2022)
• Publication Date: 2022-05
• Subjects: Balanced clustering -- Biobjective optimization -- Simulated annealing; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2022.104795
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: An effective supply chain organization is fundamental for any manufacturing, distribution, retail or wholesale business. New technologies have made considerable improvements in the whole process of inventory management; Artificial Intelligence (AI) represents one of the best options for the industry and their search for more intelligent and robust logistics solutions. Based on a real-world scenario, we approach the challenge of defining delivery routes within a city such that the time they require to be traveled is approximately the same. Moreover, while the routes must ensure that drivers' workload is time balanced and contract regulations can be met, they also must correspond to a customers' partition (sectorization) according to well-defined, non-overlapping delivery areas. We introduce an approach to solve the problem through the algorithm HSAC (Hierarchical Simulated Annealing Clustering). The proposed algorithm first applies a divisive approach to the data, using simulated annealing at each step to create time-balanced partitions, and then solves the TSP problem to create optimal routes within the defined groups. Based on real data concerning two Mexican cities, our experimental results show that HSAC can solve the sectorization problem efficiently. Highlights: A novel approach to create time-balanced delivery areas within a city. The approach combines divisive clustering and simulated annealing. Real-world data is used in the evaluation. Comparison against a classic clustering algorithm is presented.