|
Activity Number:
|
498
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #309341 |
|
Title:
|
The Use of Bayesian Tolerance Intervals for Batch-Release Testing
|
|
Author(s):
|
Richard Lewis*+ and Steven J. Novick
|
|
Companies:
|
GlaxoSmithKline and GlaxoSmithKline
|
|
Address:
|
PO Box 13398, Research Triangle Park, NC, 27709-3398,
|
|
Keywords:
|
Tolerance Intervals ; Random-Effects Models ; Batch-Release Testing ; Bayesian Tolerance Intervals
|
|
Abstract:
|
When developing quality-control criteria such as batch-release limits, one would like to have a high level of confidence that acceptable batches will meet the criteria. Tolerance limits seem well-suited for this purpose and their use has previously been proposed in a number of settings, ranging from simple random samples from a Gaussian population to random-effects models. We consider using Bayesian tolerance intervals to determine batch-release limits and simulate their performance in some selected settings, including random-effects models with unbalanced data.
|
