Sensor fault detection for manufacturing quality control

Abstract

This paper proposes a W control chart that is able to detect sensor mean shift faults and distinguish them from potential process faults in discrete-part manufacturing processes. The control chart is set up based on a linear fault quality model. The sensitivity of the W chart to the occurrence of sensor faults is studied. An index called the sensitivity ratio is used to investigate the effects of sensor fault locations and the sensor layout on the sensitivity of the W chart to sensor faults. In comparison with traditional control charts, which directly monitor the product quality characteristics, the proposed W chart can effectively separate sensor faults from process faults. An automotive body assembly process is used as an example to demonstrate the performance of the W chart.

[Supplementary materials are available for this article. Go to the publisher's online edition of IIE Transactions for the following free supplemental resource: Appendix]

Keywords:

• Sensor fault detection
• control chart
• manufacturing processes

Acknowledgement

The research is supported in part by the National Science Foundation under grants CMMI-0528735, CMMI-0726939 and CMMI-0758178.