Name: Cristian David Correa
Institution: Universidad Nacional de Colombia, Colombia
E-mail: crdcorreaal@unal.edu.co
Co-authors: Arturo Fernández, Luis Raul Pericchi
Abstract:
In industrial processes it is of interest to judge a batch of products in such a way that the acceptability of this can be decided. For this, it is necessary to select a sample of the lot with a proportion p of defective units and make the decision to reject or not the selected lot. One way to select the sample is setting the producer’s risk and minimizing the consumer’s risk or by considering a loss function which minimizes the sampling plans cost taking into account the risks of both the consumer and the producer. We consider a binomial distribution for the number of defective units in the sample because each item inspected may be or not defective. In this paper, we propose an alternative methodology that includes frequentist and Bayesian aproach for determining the sample size by using a weighted sum of the producer and consumer risks. We also show a decision rule to decide the acceptability of a submitted lot. Keywords: Producer’s and consumer’s risk, Sampling plans, Weighted producer’s and consumer’s risk, Binomial distribution.