Repeatability and reproducibility (R&R) analysis by attributes which values quantitatively the performance of a visual inspection system

Abstract

The present article proposes a system of a quantitative assessment of the ability of inspectors to make correct decisions, applying an R&R study by attributes, with the method of risk analysis, supported by contingency tables and the Cohen Kappa index. Additionally, it establishes a tool to simplify the required calculations and a proposed inspection method that seeks to reduce the variability associated with the inspection process. In this paper, a simulation study is made. We simulate several quality inspectors and three sceneries. Results show a quality level between 95% and 99% was obtained. It is possible to show that reducing the variability of inspection processes results through standardized methods, criteria equalization, and staff training improves the quality level process. The procedure presented and simulated in this article can be applied in the qualitative inspection, such as manufacturing processes as services processes.