A multi-objective evolutionary algorithm based on decomposition (MOEA/D) for the cellular manufacturing systems design problem

• Main Author: So, Jia Yi
• Other Authors: Wu Kan
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/74998
• Institution: Nanyang Technological University

Cellular Manufacturing Systems (CMS) are preferred layout for manufacturing factories due to its flexibility and adaptability to handle the dynamic and continuous changes of product demands. This project focused on proposing a method to solve multi-objective CMS facility layout problems (FLPs) that is capable to produce a Pareto Frontier of optimal solutions. The aim of this project is motivated by the research gap that past studies only targeted on combining the two conflicting objective functions of CMS FLPs and generate only one end result. However, it is noted that the decision maker preferred to choose from a set of solutions due to qualitative considerations. The author has proposed a multi-objective approach and successfully developed the corresponding algorithm to solve CMS FLPs. In comparison to a unified objective function methodology by past studies and two other metaheuristics, the author’s proposed approach outperforms with regards to solutions quality and computation time. Companies would be able to apply the author’s proposed approach for their own facility layout problems and obtained an efficient arrangement of machines, resulting in significant cost savings. There is room for exploration for this project such as modifying it to be able to generate solutions real time to suit frequent changes of part demands or allow the decision maker to play a part during solution generation process.