A hybrid model for designing attributes sampling plans

Abstract

In single sampling plans by attributes, statistical and economical considerations have traditionally been discussed separately. An approach taking into account both considerations simultaneously would be more useful in terms o f quality assurance. The suggested model involves minimization o f the expected total cost comprising the inspection cost, the annoyance cost o f rejecting a lot and the cost o f outgoing defective items, while the producer's risk and consumer's risk are not allowed to be greater than specified values. To find the optimal sample size and the optimal acceptance number, a two stage solution method is proposed. The accuracy and the efficiency o f the solution procedure are tested on randomly generated problems, by comparing the solutions obtained by the proposed method to those obtained by enumeration. Sensitivity o f the model is discussed by analyzing the effects o f parameters on the optimal sampling plans.