Improving Existing CMMS Software Packages Using Association Rules

Abstract

Since their inception, industries have experienced the negative effects of downtime, lost productivity, lost revenue, as well as layoffs. The prediction of an item's remaining useful life (RUL) enables maintenance techniques to avoid costly and serious damage. As a result, a prognostic is now acknowledged as a crucial activity. Thanks to the Internet of Things (IoT) and IT solutions like Computer Aided Maintenance Management (CMMS) software packages, industries today have a vast amount of data gathered from on-site sensors. This offers real-time data on the equipment's state as well as each piece of equipment's history of interventions from the CMMS software database. By utilizing the vast amount of data that has accumulated over the years, we will be able to extract even more crucial information. The use of artificial intelligence (AI) methods can open up new possibilities for CMMS software packages. In this study, we try to predict RUL (Remaining Useful Time) using an artificial intelligence technique called association rules. This strategy is applied to enhance existing CMMS software programs. Experiment is carried out with a well-known dataset provided by the NASA Ames Research Center and the CoE "Center of Excellence". Experiment results indicate that our suggested approach performs well in forecasting the RUL of turbojet engines and that it also significantly improves the outcomes of predictive maintenance.