Search results

The purpose of this paper is to design a modified chain sampling plan for assuring the product percentile life where the lifetime follows Weibull or generalized exponential distributions (GEDs). In order to reduce the cost of inspection when implementing the proposed modified chain sampling plan, it is also considered the economic aspect of designing of proposed plan in this paper.

The authors have designed the proposed plan on the basis of two points on the operating characteristic (OC) curve approach. The optimization problem is used to determine the plan parameters of the proposed plan so that the specified values of producer’s risk and consumer’s risk are satisfied simultaneously.

The results we have obtained, confirm that the proposed plan will be very effective in reducing the sample size rather than other existing sampling plans. The OC curves of proposed plan, chain sampling plan and zero acceptance number single sampling plan show that the performance of proposed plan in discriminating the good and poor quality lots is better than other two plans. In this paper, it is proved that the value of number of preceding lots required for current lot disposition plays an important role.

The proposed modified chain sampling plan for assuring the percentile lifetime of the products under Weibull or GEDs is not available in the literature. The proposed plan can be used in all the manufacturing industries to assure the product percentile lifetime with minimum sample size as well as minimum cost.

The purpose of this paper is to propose a new mixed repetitive group sampling (RGS) plan based on the process capability index, C_pk, where the quality characteristics of interest follow the normal distribution with unknown mean and unknown variance. Tables are constructed to determine the optimal parameters for practical applications for both symmetric and asymmetric fraction non-conforming cases. The advantages of this proposed mixed sampling plan are also discussed. The proposed sampling plan is also compared with other existing sampling plans.

In order to determine the optimal parameters of the proposed mixed RGS plan based on C_pk, the authors constructed tables for various combinations of acceptable and limiting quality levels (LQLs). For constructing tables, the authors followed the approach of two points on the operating characteristic (OC) curve. The optimal problem is formulated as a non-linear programming where the objective function to be minimized is the average sample number (ASN) and the constraints are related to lot acceptance probabilities at acceptable quality level and LQL under the OC curve.

The proposed mixed RGS plan will be a new addition to the literature of acceptance sampling. It is shown that the proposed mixed plan involves minimum ASN with desired protection to both producers and consumers compared to other existing sampling plans. The practical application of the proposed mixed sampling plan is also explained with an illustrative real-time example.

In this paper, the authors propose a new mixed RGS plan based on the process capability index C_pk, where the quality characteristic of interest follows the normal distribution with unknown mean and unknown variance. Tables are constructed to determine the optimal parameters for practical applications. The proposed mixed sampling plan can be used in all production industries. This kind of mixed RGS plan is not available in the literature.

This study investigates the repetitive mixed sampling (MRS) plan based on the Cpk index that was proposed by Aslam et al. (2013a). They were the first to study the MRS plan, but they did not pay attention to the fact that submitting to the variable inspection a sample that was first submitted to the attribute inspection, truncates the X observations. In addition, they did not work with an accurate expression to calculate the probabilities of the Cpk statistic.

The authors presented the results based on their original sampling plan through Monte Carlo simulation and defined the theoretical results of their plan when the sample submitted to the variable inspection is no longer the same one submitted to the attribute inspection.

The β risks of the optimum sampling plans presented by Aslam et al. (2013a) are pretty high, exceeding 46%, on average – this same problem was also observed in Saminathan and Mahalingam (2018), Balamurali (2020) and Balamurali et al. (2020), where the β risks of their proposed sampling plans are yet higher.

In terms of originality, the authors can declare the following. It is not a big deal to propose new sampling plans, if one does not know how to obtain their properties. The miscalculations of the sampling plans risks are dangerous; imagine the situation where the acceptance of bad lots exceeds 50% just because the sampling plan was incorrectly designed. Yes, it is a big deal to warn that this type of problem is arising in a growing number of papers. The authors of this study are the pioneers to discover that many studies focusing on the sampling plans need to be urgently revised.