This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 1 Shewhart I II (ARL) Saniga(1989) … x I II 01 . = u α 10 . , = u β and 4 = u ATS 21.86% 50.0% Abstract Economic design of a control chart is really lower the cost compared with a conventional Shewhart charts. But economic design does not consider the statistics properties, such as type I or type II error and average run length (ARL). In order to improve these issues, an economic-statistical design of control charts was proposed and most of them are considered under one assignable cause. However, there are multiple assignable causes in the real practice such as machine problem, material deviation, human errors,…, etc. This research proposes an economic-statistical model of x control chart under the consideration of multiple-assignable causes. A numerical example was employed to demonstrate the model’s working. It also shows that the cost is up about 21.86% but the test power increases 50.0% of the original one when the limits of type I error, test power and ARL are set at 0.01, 0.90 and 4. 1. Shewhart k h n Duncan (1956) x Shewhart Duncan (1956) (Single Assignable Cause) Duncan (1971) (Multiple Assignable Causes) Saniga (1977) x- R Lorenzen and Vance (1986) Banerjee and Rahim (1988) Weibull Collani and Sheil (1989) S Duncan (1956) 2 I II (ARL) Saniga(1989) I II (ARL) Saniga 1989 I II McWilliams 1994 X 0.2 ARL 1 100 1.0 ARL 1 150 Grid Lorenzen Vance 1986 Saniga 1995 Lorenzen Vance 1986 Torng et al. (1995) EWMA Rahim Oraini 2002 Gamma λ ,2 X … (Multiple Assignable Causes) 2....
View Full Document
- Spring '09
- Control Chart, W. Edwards Deming, Duncan, American Statistical Association, 14 2.05 2.59 14 2.59 2.59