Activity Number:
|
284
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 14, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Bayesian Stat. Sciences*
|
Abstract - #301944 |
Title:
|
Bayesian Determination of Calibration Interval Length
|
Author(s):
|
Robert Noble*+
|
Affiliation(s):
|
Miami University
|
Address:
|
294 Bachelor Hall, Oxford, OH, 45056, USA
|
Keywords:
|
Bayesian ; calibration interval length
|
Abstract:
|
The length of time between calibrations of various types of equipment or processes is a function of the quality of the equipment or process itself and the risk associated with operating outside prescribed specifications. It is not enough to control the accuracy at the time of calibration but control needs to be extended to cover the periods of usage between calibrations as well. The principle method of extending this control has traditionally been to calibrate at optimal time intervals, but deciding what the optimal length actually is has been an evasive problem. Determining the length of the optimum interval between calibrations is the focus of this research. The method proposed is a Bayesian approach to the problem that controls the risk of operating outside specified limits and is shown to be much more powerful than the currently recommended method for small inventory sizes that are typically found in practice.
|