Statistical Control: By Monitoring and Feedback AdjustmentA detailed, practical, accessible guide to efficient statistical control. Efficient control is a key element in the improvement and maintenance of quality and productivity. This book shows the advantages of bringing together the more commonly used methods of statistical quality control with appropriate techniques of feedback adjustment. It uses recent research and practical experience to provide feedback methods of immediate use in the workplace. Statistical Control by Monitoring and Feedback Adjustment introduces a new coordinated approach to quality control. The authors' clear and cogent presentation uses extensive graphical explanation supplemented by numerous examples and computational tables. A helpful selection of problems and solutions further facilitates understanding. Topics covered include: * A fresh look at process monitoring * Using feedback adjustment charts * Minimizing the size of adjustments * Feedback techniques that minimize costs of adjustment and sampling * Detection of special causes with Cuscore Statistics * Efficient monitoring of operating feedback systems * The roles of models, optimization, and robustness * A brief review of time series analysis. Statistical Control by Monitoring and Feedback Adjustment is important reading for quality control engineers and statisticians as well as graduate students in quality control, industrial engineering, and applied statistics. |
From inside the book
Results 1-3 of 46
Page 13
... representing the myriad num- ber of " lurking " variables not so far identified . Thus the common causes are represented by , e ( xn + 1 , Xn + 2 , . . . ) , and the discovery of a special cause would correspond to the identification of ...
... representing the myriad num- ber of " lurking " variables not so far identified . Thus the common causes are represented by , e ( xn + 1 , Xn + 2 , . . . ) , and the discovery of a special cause would correspond to the identification of ...
Page 14
... represented by the nonstationary IMA model referred to earlier . Also , that , when there is process inertia , it can be represented by a first - order dynamic system , such as discussed in Chapter 4 , possibly with one or more added ...
... represented by the nonstationary IMA model referred to earlier . Also , that , when there is process inertia , it can be represented by a first - order dynamic system , such as discussed in Chapter 4 , possibly with one or more added ...
Page 160
... represented by a first - order dynamic system with parameter & so that , as in Chap- ter 4 , = X1 + 1 = X1 = ( 1 ― 8 ) ( X + do X1 - 1 + 83 X1 - 2 + · · · ) . 2. The disturbance can be represented by an IMA model ― = - Zt + 1 - Zł at +1 ...
... represented by a first - order dynamic system with parameter & so that , as in Chap- ter 4 , = X1 + 1 = X1 = ( 1 ― 8 ) ( X + do X1 - 1 + 83 X1 - 2 + · · · ) . 2. The disturbance can be represented by an IMA model ― = - Zt + 1 - Zł at +1 ...
Contents
Control Charts for Frequencies and Proportions | 19 |
Control Charts for Measurement Data | 57 |
Modeling Process Dynamics and Forecasting Using | 85 |
11 other sections not shown
Common terms and phrases
a₁ a₁+1 action limits adjustment equation Analysis Appendix applied appropriate approximately assumption binomial distribution Biometrika calculated Chapter control chart control scheme correlation corresponding cost Cuscore statistic dead band deviation from target discrete discussed disturbance estimate EWMA example exponentially weighted factor feedback control first-order forecast errors gX₁ illustration IMA model increase interpolation limit lines linear Luceño mean square error measurements Methods minimum moving average moving range nonstationary normal distribution observations obtained occur optimal output standard deviation overdispersion parameter plotted Poisson distribution probability procedure process adjustment process monitoring produce random reference distribution residuals robustness series model Shewhart chart shown in Figure shows signal smoothing constant special causes standard deviation stationary statistical process control statistically independent Suppose Table target value Technometrics temperature thickness tion unit interval variable variance variation variogram warning limits white noise x₁ Y₁ z₁ zero