摘要:

Nondestructive evaluation (NDE) techniques are used widely in many manufacturing and operational areas of industry. NDE is a highly statistical science that has developed rapidly over the past decade with relatively little input from individuals with forma1 training in statistics. This review article describes some of the many methods and applications of NDE, including eddy-current methods for detecting fatigue in jet-engine turbine disks, ultrasonic methods for detecting flaws in forged titanium, and radiographic methods for detecting out-of-control conditions in a casting process. We will also outline the important potential for increased use of modern methods of statistics to improve the practice of nondestructive evaluation. Areas of statistics important in NDE include data analysis and signal processing, decision theory, screening and calibration, the assessment of measurement capability, and the use of designed experiments for improvement of measurement/detection processes.

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484451

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484452

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484453

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484454

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484455

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484456

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484457

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

We discuss models for recurrent events that incorporate both time trends and effects of past events, such as renewal-type behavior. Inference procedures, including tests for trend, are developed and illustrated on repairable-systems failure data. Simulations are used to examine the accuracy of large-sample approximations used for tests or interval estimation.

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484458

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

Often the least appropriate assumption in traditional control-charting technology is that process data constitute a random sample. In reality most process data are correlated—either temporally, spatially, or due to nested sources of variation. One approach to monitoring temporally correlated data uses a control chart on the forecast errors from a time series model of the process with, possibly, a transfer-function term to model compensatory adjustments. If the time series term is an integrated moving average, then a sudden level shift in the process results in a patterned shift in the mean of forecast errors. Initially the mean shifts by the same amount as the process level, but then it decays geometrically back to 0 corresponding to the ability of the forecast to “recover” from the upset. We study four monitoring schemes—umulative sums (CUSUM's), exponentially weighted moving averages, Shewhart individuals charts, and a likelihood ratio scheme. Comparisons of signaling probabilities and average run lengths show that CUSUM's can be designed to perform at least as well as, and often better than, any of the other schemes. Shewhart individuals charts often perform much worse than the others. Graphical aids are provided for designing CUSUM's in this context.

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484459

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

Control charts are widely used in manufacturing to distinguish between variation that is inherent to the process and variation that signals a special event or problem. Several control charts have recently been proposed for the detection of a periodic component, but their analysis is limited to a moving span of recent observations. This article develops tests for periodicity that make use of independent information about the variability of the time series. These tests are then used to construct control charts that search for periodicity in a span of recent observations but estimate the variability of the process from a longer span of earlier observations. By employing more of the available information, the new charts are able to detect cyclic disturbances more quickly than the existing charts.

ISSN:0040-1706    

DOI:10.1080/00401706.1996.10484460

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor