Performance prediction of a modular product variant with RS-SVM

摘要

A hybrid performance prediction approach for newly modular configured products is put forward, which can rapidly and accurately estimate the performance parameter values of a product variant. This method integrates the rough set (RS) theory and support vector machine (SVM) Dimension reduction is achieved by the RS theory that could reduce the configuration parameters. Then these extracted attributes are used as input variables in SVM model. The performance values of a newly configured product can be predicted by means of the trained SVM model. This RS-SVM method helps to evaluate whether the product variant can satisfy the customers' individual requirements.