The purpose of this paper is to develop a statistical control chart based model for earlier defect identification.
The paper used statistical process control methods and an auto‐regression model to model the identification of the initiation point of a random defect. Conventional statistical process control (SPC) methods have been widely used in process industries for process abnormality detections. However, their practicability and achievable performance are limited due to the assumptions that a continuous process is operated in a particular steady state and that all variables are normally distributed. Because the case considered here does not meet the requirement of conventional SPC methods, we proposed adaptive statistical process control charts based on an autoregressive model to distinguish defects from normal changes in operating conditions. The method proposed has been tested on a set of vibration data of rolling element ball bearings
Several control charts have been used and compared in this paper to identify the initial point of a defect. Overall, the adaptive Shewhart average level chart is a good choice since it overcomes the drawback of adaptive moving charts by working out the limits using all the bearings' data, with no such a need for a subjective threshold level. They are also not very sensitive to the small casual changes in the data.
The model developed can be served as part of a prognosis tool for maintenance decision making since once the earlier warning point has been identified, corrective maintenance actions may be taken. It has practical application areas in vibration based monitoring or any monitoring scheme where a trend in the monitored measurements may exist. The method proposed is easy to use and can be implemented in any condition based maintenance software packages.
The approach proposed in this paper is a new application of existing methods and of original contribution from a point of view of applicability. It adds value to the existing literature and is of value to practitioners.
Wang, W. and Zhang, W. (2008), "Early defect identification: application of statistical process control methods", Journal of Quality in Maintenance Engineering, Vol. 14 No. 3, pp. 225-236. https://doi.org/10.1108/13552510810899445Download as .RIS
Emerald Group Publishing Limited
Copyright © 2008, Emerald Group Publishing Limited