Production performance evaluation based on rough set theory and wavelet neural network

Abstract

Aimed at overcoming subjectivity and improving the accuracy of traditional production performance evaluation methods for manufacturing enterprises, a new model of performance evaluation was proposed based on rough sets and a wavelet neural network (RS - WNN). Firstly, an evaluation index system considering innovation performance was constructed. Secondly, a theory of rough sets and fuzzy mathematics was utilized to preprocess and simplify the index system, and then, the input dimensionality of wavelet neural network was reduced. Finally, algorithms of stepwise checkout and iterative descending grads were employed to decide the parameters of WNN and to obtain the synthetic evaluation value of production performance. A case study showed that the proposed model was effective and feasible in measuring production performance.