Optimizing reconfigurable manufacturing system configuration selection with multi-objective grey wolf optimization

Reconfigurable Manufacturing Systems (RMSs) represent a pivotal paradigm in modern manufacturing, offering the flexibility to adapt to varying production demands. The configuration selection of an RMS significantly influences its performance and responsiveness to dynamic manufacturing environments. In the present work, multiple objective grey wolf optimization (MOGWO) is implemented for the optimal configuration design of an RMS. The real encoded solution assisted in maintain the feasibility of solutions and minimization of search space. The discrete set of feasible machine configurations are handled efficiently to be utilized for the RMS configuration design. The non-dominated solutions obtained by MOGWO are an asset to the manufacturing system design. The decision manager may select a suitable candidate from among the non-dominated solutions in light of the current market situation. Evolutionary algorithms generate initial populations by randomly selecting variables. At stage S-1, operation 15, with a real value of 0.6529, is assigned one of three configurations: {\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{mc}_{2}<^>{2}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{mc}_{3}<^>{2}$$\end{document},and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{mc}_{5}<^>{3}$$\end{document}}. The selected configuration is determined by multiplying the encoded solution's real value by the number of alternatives and rounding up to the nearest integer. This approach confines the search space to feasible regions, ensuring equal probability for all alternatives. The production sequence is 15 -> 1 -> 17 -> 3 -> 5, with a demand rate of 50 units per hour. Performance metrics use power indices Z and Y set to 2, with parameters w1, w2, and w3 valued at 0.5, 0.4, and 0.1, respectively.