Improving Quality Control Applications by Increasing Inspection Efficiency and Decreasing Nonconformance Percentage

Abstract

Quality controls are activities to evaluate the level of conformity of product attributes and optimal quality objectives. When 100% inspection is applied in quality control processes, sampling is used because it causes high costs, long control times and product damage. Acceptance sampling, which is a statistical method, determines whether the lot can be accepted or rejected in line with the tests performed on the samples taken from the lot. The acceptance sampling plan depends on multiple factors such as the level used, the degree of control applied, the lot size, and the acceptable quality level. For this reason, the use of standard sampling plans that increase the validity of quality control operations can be expressed. In this paper, acceptance process applications were studied for a the company that demonstrates textile industry studies. In this the company, the control processes entered the products in the batch of different sizes coming from the regulations used for the contract are applied. To determine the acceptance listening, the execution of the lot, the control measurement dimensions entered first, the reasons for the return of the the company’s four product groups and the AQL reports are reviewed, and the statistical evaluations of the quality controls come to an end. Next steps, using ANSI/ASQ Z-1.4, observations suitable for lot sizes and appropriate acceptance-rejection details were determined and compared with the size and decision points of the the company. Cause-effect diagrams do not take into account the reasons that cause the returns to be made so that the possible reasons for the returns can be examined. Finally, according to the results of the sampling, solutions were found to make the dimensions for these reasons. Cause-effect diagrams do not take into account the reasons that cause the returns, so that possible causes of the returns can be examined. Finally, according to the sampling results, solutions were found for sizing for these reasons. It is recommended to taken 50 samples from lot sizes between 281–500, 80 samples from 501–1200 lot sizes, and 125 samples for lot sizes between 1201 and 3200. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.