A Tandem Evolutionary Algorithm for Platform Product Customization

Abstract

The increasing demand for reduction in time to market has led to new product development methodologies focusing on reuse. Platform product customization emerges as a basic idea for avoiding designing a new product completely from scratch, but reusing a product platform instead. Key dimensions of some modules of the platform can be stretched or shrunk while modules themselves can be swapped to formulate multiple product variants in a family according to specific customer requirements. This paper formulates such a problem of platform product customization as an optimization problem consistent with the manufacturer’s goal in the design of product variant(s) while satisfying customer requirements and design constraints. Optimal customization takes place at the levels of product structure and module parameters. A tandem evolutionary algorithm is proposed for identifying the optimal structural composition and the optimal parameters of the corresponding structure. The encoding schemes and genetic operators for structure and parameter optimization are designed, respectively. A case study of gantry crane customization is given to illustrate how the proposed evolutionary customization design method is used. The effectiveness of the method is also evaluated through a series of sensitivity analyses.