A combined life prediction method for product based on IOWA operator

Abstract

As the rapid development of modern technology, industrial companies have to manufacture high-reliable and long-lifetime products. How to evaluate these indexes is an urgent problem to be solved. Utilizing the product degradation information may be an effective way to solve this issue. However, most of lifetime prediction models in use are mainly based on single prediction model with the shortage of low robustness and accuracy. In this chapter, a combined prediction method based on the performance degradation data by using the induced ordered weighted averaging operator (IOWA) is proposed. We select two better prediction models, which are time series model and BP neural network, to predict the degradation path of product respectively. The IOWA operator can build a new combination prediction method which can overcome the defect of fixed weight coefficients of the traditional combined method. This method can update the weight coefficients dynamically according to the prediction precision of each model. Then the objective function of the error square sum is established with weight coefficients used to combine these prediction methods and integrate the prediction results.