ICCP: A heuristic process planning method for personalized product configuration design

• Published in: Applied intelligence (Dordrecht, Netherlands) Netherlands), 2023-12, Vol.53 (24), p.30887-30910
• Main Authors: Hu, Kerui, Qiu, Lemiao, Zhang, Shuyou, Wang, Zili, Fang, Naiyu
• Format: Article
• Language: English
• Subjects: Artificial Intelligence; Computer Science; Configuration management; Coupling; Crystallization; Customization; Decoupling; Heuristic; Heuristic methods; Industry 4.0; Knowledge based engineering; Machines; Manufacturing; Mechanical Engineering; Modules; Process planning; Processes
• ISSN: 0924-669X; 1573-7497
• DOI: 10.1007/s10489-023-05186-z

Product personalization, a popular topic of Industry 4.0, requires enterprises to develop customer-centric products in a cost-efficient way. To achieve this goal, a personalized product configuration design that divides a product into multiple modules and instantiates them in parallel has been proposed and proven to be an effective method. However, the complex coupling relationships between component instances and their adaptability to individual requirements directly affect configuration design efficiency and accuracy, which requires reasonable instantiation process planning. Thus, this paper proposes a heuristic process planning method for personalized product configuration design. It first defines the personalized product configuration unit (PPCU) that encapsulates module information and design knowledge, avoiding repeated knowledge retrieval in the instantiation. Using the relevant data contained in the PPCU, a characteristic quantification method (CQM) is proposed to measure the coupling characteristics and adaptability characteristics of the PPCU. Taking characteristic values as input, the proposed heuristic approach, which simulates the crystallization growth process at imbalanced temperatures (ICCP), is applied to configuration design process planning to quickly identify appropriate instantiation sequences. In addition, the novel decoupling strategy we proposed considers the time-coupling ratio and the deformation workload expectation to split the strongly coupled PPCU group. Comprehensive performance experiments on a real dataset and an application experiment on personalized elevator design are conducted. The results verify that our method can achieve the configuration design with the highest planning efficiency and best design configuration result compared to five state-of-the-art methods, especially in the configuration design of complexly structured personalized products.