Multi-factor integrated configuration model and three-layer hybrid optimization algorithm framework: turnkey project-oriented rapid manufacturing system configuration
Shu-Lian Xie,Feng Xue,Wei-Min Zhang,Jia-Wei Zhu,Zi-Wei Jia
DOI: https://doi.org/10.1007/s40436-023-00476-8
IF: 3.837
2024-03-23
Advances in Manufacturing
Abstract:In the context of increasingly prominent product personalization and customization trends, intelligent manufacturing-oriented turnkey projects can provide manufacturers with fast and convenient turnkey services for manufacturing systems. Their key characteristic is the transformation of the traditional design process into a configuration process. However, the scope of configuration resources in existing research is limited; the cost and time required for manufacturing system construction are overlooked; and the integration of the system layout configuration is rarely considered, making it difficult to meet the manufacturing system configuration requirements of turnkey projects. In response, this study establishes a multi-factor integrated rapid configuration model and proposes a solution method for manufacturing systems based on the requirements of turnkey projects. The configuration model considers the system construction cost and duration and the product manufacturing cost and duration, as optimization objectives. The differences in product feature-dividing schemes and configuration of processes, equipment, tools, fixtures, and layouts were considered simultaneously. The proposed model-solving method is a three-layer hybrid optimization algorithm framework with two optimization algorithm modules and an intermediate algorithm module. Four hybrid configuration algorithms are established based on non-dominated sorting genetic algorithm-III (NSGAIII), non-dominated sorting genetic algorithm-II (NSGAII), multi-objective simulated annealing (MOSA), multi-objective neighborhood search (MONS), and tabu search (TS). These algorithms are compared and validated through a hydraulic valve block production case, and the TS and NSGAIII (TS-NSGAIII) hybrid algorithm exhibits the best performance. This case demonstrates the effectiveness of the proposed model and solution method.
engineering, manufacturing,materials science, multidisciplinary