Learning-Based Algorithm for Fault Prediction Combining Different Data Mining Techniques: A Real Case Study

Abstract

In recent years, new Data Mining (DM) algorithms and methodologies are increasingly used as an industrial solution for manufacturing improvements. In this context, new techniques are widely required by companies in the field of maintenance due to the need to reduce breakdowns intervention and take advantage of the increasing availability of data. This paper aims to propose a new learning-based algorithm to improve knowledge extraction by combining different DM techniques from a predictive maintenance perspective. First, the J48 algorithm and Random Forest (RF) are used as a predictive model to classify a set of failure modes according to their influence on the Overall Equipment Effectiveness (OEE). Then, the Apriori algorithm is used to identify the relationship among failure events belonging to the lowest OEE range for which, therefore, a predictive maintenance strategy should be defined. In order to describe the learning-based algorithm proposed in this paper, a real case study is presented and detailed. The experimental results showed a valuable tool for knowledge extraction and the definition of a set of predictive maintenance strategies for those failures most affecting the process. In this way, the complexity of decision-making on maintenance strategies can be reduced mainly when dealing with a large amount of information or a challenging dataset.

Keywords

• Algorithms
• Knowledge extraction
• Failure prediction
• Data mining
• Case study