A Knowledge Based FMEA to Support Identification and Management of Vehicle Flexible Component Issues

Abstract

The objective of this research is to develop a knowledge-based approach of performing Failure Modes and Effects Analysis (FMEA) on flexible vehicle components. A three-step knowledge discovery and data mining process is used to develop a set of rules. This set provides decision support to engineers while they perform FMEA on flexible vehicle components. The first step is where traditional FMEAs are performed on flexible components. In the second step, a decision tree algorithm (J48) is used to extract rules from the collected data. The third step involves manual post-processing of the rules extracted by the decision tree algorithm. Finally, the set of rules is implemented in a knowledge-based FMEA tool that can be used to provide decision support to engineers while they perform FMEAs on flexible components. The tool is used to identify and review quality issues within a complete vehicle. The top ten issues identified from the FMEA are submitted to design and manufacturing engineers for detailed evaluations.