Abstract: This research presents a new sampling plan called the group-adaptive multiple dependent state sampling plan (GAMDSSP) that aims to assess product quality in ongoing production by using a truncated life test. The new method takes parts from both the group acceptance sampling plan (GASP) and the adaptive multiple dependent state sampling plan (AMDSSP), enabling decisions to be made based on the total number of defective items found in different groups. The core statistical foundation of the plan relies on the Weibull distribution, which is commonly used to model product lifetimes. In this study, the shape parameter of the distribution is assumed to be known, while the scale parameter is defined in terms of the mean lifetime of the product. To enhance the efficiency of the proposed sampling plan, a genetic algorithm is employed for estimating the optimal plan parameters. The objective of optimization is to minimize the average number of samples required while simultaneously safeguarding the interests of both consumers and producers by maintaining acceptable risk levels. Extensive numerical experiments are conducted to validate the plan’s performance. These experiments demonstrate that GAMDSSP offers greater flexibility and efficiency compared to existing sampling plans, particularly in terms of the average sample number. Two real-world examples further demonstrate the usefulness of the GAMDSSP in various industrial situations. The findings suggest that the GAMDSSP is a robust and adaptable tool for improving quality control in continuous production environments.

Keywords: Average sample number, Group-adaptive multiple dependent state sampling plan, Mean lifetime, Truncated life test, Weibull distribution

Pages: 130-140