A Critique of Deming's Discussion of Acceptance Sampling Procedures.
Abstract
An inspection sampling problem discussed by Deming is analyzed assuming the percent defective, p, is partially unknown with prior pi(*). In this case, the all or none inspection rule discussed by Deming may not be correct. Simple condtions based on the lot size, the ratio of costs, and the posterior mean determine when 100% inspection is best. Efficient algorithms based on the posterior mean are given for determining the optimal inspection plan in general. Keywords include: Acceptance sampling; Bayesian inference; Sampling plan algorithms.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1985
- Accession Number
- ADA154305
Entities
People
- R. E. Barlow
- Xiaoxuan Zhang
Organizations
- University of California, Berkeley