Integrated warehouse assignment and carton configuration optimization using deep clustering-based evolutionary algorithms. (February 2023)

• Record Type: Journal Article
• Authors: Das, Jyotirmoy Nirupam; Tiwari, Manoj Kumar; Sinha, Ashesh Kumar; Khanzode, Vivek
• Is Part Of: Expert systems with applications. Volume 212(2023)
• Publication Date: 2023-02
• Subjects: Last-mile -- Warehouse Operations -- Deep Clustering -- Multi-objective; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2022.118680
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Optimization of the packaging and delivery of orders from warehouse to customers. A multiobjective formulation has been proposed to solve the optimization problem. Clustering and evolutionary algorithms have been used to get optimal solutions. Proposed methodology reduces the carbon footprint and carton manufacturing costs. Abstract: A rapid rise in e-commerce has forced logistic companies to invest in efficiency and reduce last-mile delivery costs. A significant part of last-mile delivery operations is the cartonization of orders and economically delivering them. An optimal carton configuration leads to a better cartonization and reduces the carton manufacturing costs and carbon footprint, whereas the optimal warehouse allocation directly reduces the transportation costs. Therefore, a multiobjective formulation has been proposed in this study to address the warehouse assignment and carton configuration optimization problem. A novel Interdependent Pareto Ant Colony Optimization (IPACO) has been integrated with a Deep Embedded Clustering algorithm (DEC) to form a DEC-based IPACO (DECIPACO) model to solve the proposed formulation. The integrated model was tested on 54 different datasets and compared against other clustering-based evolutionary algorithm models. The DECIPACO model provided an optimal or a non-dominated solution in all cases against the k-means clustering-based evolutionary algorithm models. Hence, the proposed DECIPACO model was able to explore an optimal trade-off between fuel costs and total carton volume while fulfilling the customer demand from inventory-constrained warehouses.