Integrating grey relational analysis and support vector machine for performance prediction of modular configured products

Abstract

Evaluating whether a newly configured product can satisfy the customers' individual requirements or not is crucially important for the modular configuration design. Product performance prediction at the end of the configuration process can estimate the performance parameter values through the soft computing method instead of practical test experiments, which enables fast and accurate evaluation of configuration schemes. In this article, we propose a novel prediction approach based on the integration of grey relational analysis and support vector machine through discovering the knowledge from the historical configuration information. The implementation process of the prediction is established, and the procedure in applying the prediction to the configuration design is presented. There are three key steps to achieve performance prediction. First, the module parameters that affect the performance need to be reduced using the grey relational analysis method and then a module parameter reduction is generated. Second, the relationship between the reduced module parameters and the performance parameter is mined from the limited existing product data. A support vector machine model used for regression prediction is constituted. Third, when the values of the module parameter reduction are determined, the performance value of a newly configured product can be predicted by means of the support vector machine model. This methodology can ensure the performance prediction executed in a short period of time with a high degree of precision, even under the small-sample conditions. A design case of the plate electrostatic precipitator is studied to illustrate and demonstrate the feasibility of the proposed method.

Multi-objective optimization of process parameters in an area-forming rapid prototyping system using the Taguchi method and a grey relational analysis

“…In equation (4), the distinguishing coefficient ζ can be set between 0 and 1 for the diversity of grey relational coefficients. 36 In general, the value of the distinguishing coefficient is set to 0.5. 32 In addition, the sensitivity analysis by varying the distinguishing coefficient is shown in the experiment results…”

A scheduling procedure for a flow shop–like knowledgeable manufacturing cell with self-evolutionary features

“…SVMs have gained their wide applications in fields such as pattern recognition 41 and the regression of non-linear functions. The principle of SVMs is to map the input vector from a low-dimensional space into a high-dimensional linear space such that the problem can be treated as linear and solved.…”