An Approach Based on Improved Grey Model for Predicting Maintenance Time of IPS2

Abstract

Maintenance service is the most important part in industrial product-service system (iPSS). To enhance the proactiveness of maintenance service, the failure time of product need to be predicted. This study provides insight into how the maintenance time can be predicted accurately for the service decision-making of IPSS based on the historical failure time data, avoiding the dependence on mechanical condition monitoring. An improved grey prediction model is introduced to obtain a predictive interval of maintenance service time which is used as basis for making a decision of proactive maintenance service. The method is applied in an instance of maintenance service for agricultural machinery. In three examples of different types of agricultural machinery, the predicted intervals of maintenance time are given and the fitted values have high accordance with the real ones. The result shows that this approach can be used effectively to predict production failure time. Moreover, the output results of this approach can provide the necessary support to make a decision for proactive maintenance service without the dependence on mechanical condition monitoring in a reliable way.