Generalized Bayesian Double Group Sampling Plan for Manufacturing Industry

Abstract

Quality is not simply a goal or a choice for organizations, it is also a need for success in the global market. Acceptance sampling is one of two key strategies for quality assurance in manufacturing industry, along with statistical process control. After inspection the lot is either accepted or rejected based on the acceptance criteria. If historical information about the product is available, then the most effective approach for making the appropriate judgement is the Bayesian approach. To estimate quality regions, this work presents a Bayesian double group sampling plan (BDGSP). Based on acceptance criteria, the binomial distribution is used to build a likelihood function for defective and non-defective items. The beta distribution is utilized as the prior distribution to determine the average probability of acceptance. For some stated values of producer’s and consumer’s risks, four different quality regions are estimated. The suggested plan estimates variation point values based on various design parameter combinations. Producer's and consumer's risks correlate with acceptable quality levels and limiting quality levels of regions, respectively. Operating characteristic curves are used to monitor the effects of change in the values of specified parameters and for comparison with existing sampling plan. Application based on real data set proves that the proposed plan is applicable for existing manufacturing industry policies.