This paper develops a new method for implementing mass-customization, namely, the customization around standard products, or catalog-based customization. The method addresses the customization requirements of a class of products that are complex in configuration, multi-functional and structurally similar. We formulate catalog-based customization as an optimization problem consistent with the manufacturer’s goal of incurring minimal costs in the redesign of existing standard components, while meeting customer specifications and satisfying design constraints. The ‘catalog-based’ nature of the formulation raises concomitant issues of cost function development and problem simplification/solution. We identify the generational structure as best suited to exploit the cost data in existing catalogs and construct a product cost function. The cost-estimation methods used by the generational structure in the construction are identified as weight-based — for modeling the material costs, and methods based on similarity principles and regression analyses — for the production costs. The optimization formulation of catalog-based customization may be simplified by an a priori identification of a standard catalog design as the customization basis. This is accomplished with function costing — a cost-estimation hypothesis that uses product functionality to develop an approximate cost-estimate. The function-costing estimate is also used to abstract features from the standard base design into the optimization formulation. The preferred solution strategy for the optimization formulation is identified as genetic algorithms. We apply the customization method developed to Brushless D.C. Permanent Magnet (BDCPM) motors and obtain optimal minimal cost custom designs (from the standard designs of a BDCPM motor family) for different sets of customer requirements.

This content is only available via PDF.
You do not currently have access to this content.