Prediction of manufacturing resource requirements from customer demands in mass-customisation production

• Published in: International journal of production research Vol. 47; no. 5; pp. 1245 - 1268
• Main Authors: Dean, P.R., Xue, D., Tu, Y.L.
• Format: Journal Article
• Language: English
• Published: Abingdon Taylor & Francis Group 01.01.2009; Taylor & Francis; Taylor & Francis LLC
• Subjects: Applied sciences; Customization; Demand; Exact sciences and technology; fuzzy pattern clustering; Inventory control, production control. Distribution; linear regression; Manufacturing resource planning; mass-customisation production; Neural networks; Operational research and scientific management; Operational research. Management science; Regression analysis; Studies; Manufacturing resource planning; Branching; Mass production; Linear regression; Customization; neural networks; Production line; Regression analysis; Windows; Neural network; Modeling; Linear model; fuzzy pattern clustering; Fuzzy logic; Multiagent system; Sales; Classification; Data models; mass-customisation production
• ISSN: 0020-7543; 1366-588X
• DOI: 10.1080/00207540701557197

Mass-customisation production is a new manufacturing approach to produce customised products based on requirements of individual customers while maintaining the quality and efficiency of mass production. Due to the large variations of customised products, the traditional methods for planning manufacturing resources based on volumes of mass produced products are not effective for mass-customization production. In this research, a new manufacturing resource planning method is developed by studying the relations between customer demands and manufacturing resource requirements based on the true data from a mass-customisation production company-Gienow Windows and Doors. In this research, first the relations between the customer demands, modeled by sales data at levels of whole company, sales branches, and markets in sales branches, and the manufacturing resource requirements, modeled by labour requirements of different production lines are studied. Fuzzy pattern clustering method is employed for classifying the resource requirements into patterns to further understand the relations. Based on this study, linear regression and neural network are used to model the linear and non-linear relations between customer demands and manufacturing resource requirements, and to predict the manufacturing resource requirements from available customer demands. A manufacturing resource planning system was developed to demonstrate the effectiveness of this introduced approach.