A hybrid invasive weed optimization for an imperfect, two-warehouse, lot-sizing problem. (28th January 2020)

• Record Type: Journal Article
• Authors: Atabaki, Mohammad Saeid; Pasandideh, Seyed Hamid Reza; Mohammadi, Mohammad
• Is Part Of: Journal of modelling in management. Volume 15:Number 4(2020)
• Page Start: 1363
• Page End: 1387
• Publication Date: 2020-01-28
• Subjects: Logistics -- Optimization -- Supply chain management -- Inventory control; Industrial management -- Mathematical models -- Periodicals; Industrial management -- Computer simulation -- Periodicals; Business -- Mathematical models -- Periodicals; Business -- Computer simulation -- Periodicals; 658.4033
• DOI: 10.1108/JM2-03-2019-0059
• Languages: English
• ISSNs: 1746-5664

Abstract

Abstract : Purpose: Lot-sizing is among the most important problems in the production and inventory management field. The purpose of this paper is to move one step forward in the direction of the real environment of the dynamic, multi-period, lot-sizing problem. For this purpose, a two-warehouse inventory system, imperfect quality and supplier capacity are simultaneously taken into consideration, where the aim is minimization of the system costs. Design/methodology/approach: The problem is formulated in a novel continuous nonlinear programming model. Because of the high complexity of the lot-sizing model, invasive weed optimization (IWO), as a population-based metaheuristic algorithm, is proposed to solve the problem. The designed IWO benefits from an innovative encoding–decoding procedure and a heuristic operator for dispersing seeds. Moreover, sequential unconstrained minimization technique (SUMT) is used to improve the efficiency of the IWO. Findings: Taking into consideration a two-warehouse system along with the imperfect quality items leads to model nonlinearity. Using the proposed hybrid IWO and SUMT (SUIWO) for solving small-sized instances shows that SUIWO can provide satisfactory solutions within a reasonable computational time. In comparison between SUIWO and a parameter-tuned genetic algorithm (GA), it is found that when the size of the problem increases, the superiority of SUIWO to GA to find desirable solutions becomes more tangible. Originality/value: Developing a continuous nonlinear model for the concerned lot-sizing problem and designing a hybrid IWO and SUMT based on a heuristic encoding–decoding procedure are two main originalities of the present study.