\(\alpha\)-optimal sampling plans for large acceptance numbers (Q1087289)

Lots of size N are inspected without replacement by a single sampling plan (n,c) where n is the sample size and c the acceptance number. A cost model is assumed which is linear in the inspection costs and the terminal decision (acceptance or rejection of the lots). Incorporating the probability (called \(\alpha)\) that the actual lot quality is not worse than the break-even quality, a minimax-regret approach is pursued and mathematically elaborated. An approximate solution is derived for the case \(c>10\).