Residual Responses to Change Patterns of Autocorrelated Processes |
| |
Authors: | Mohamed El Ghourabi Mohamed Limam |
| |
Affiliation: | 1. Université de Tunis , Tunisia mohamed. elghourabi@laposte.net;2. Université de Tunis , Tunisia |
| |
Abstract: | This article studies the residual behaviour of various stationary processes in the presence of change patterns. Three types of change patterns are considered, Additive Outliers, Innovative Outliers and Level Shift. The knowledge of the residual behaviour is important for monitoring production processes. A new method of residual process control is proposed, the patterns chart. In addition to the advantage of detecting change patterns, it distinguishes their nature. The patterns chart's performance is compared to the performance of the special causes control (SCC) chart based on average run length. The results show that the proposed method performs better than a SCC chart. A real case study illustrates that the patterns chart has all the desirable properties of a SCC chart and it overcomes the negative ones. |
| |
Keywords: | Autocorrelation outliers residual responses control chart ARL |
本文献已被 InformaWorld 等数据库收录! |