Akademik Veri Yönetim Sistemi
Journal of Multiple-Valued Logic and Soft Computing, cilt.40, sa.3-4, ss.371-400, 2023 (SCI-Expanded)
Statistical process control (SPC) that generally relies on the use of control charts (CCs) to monitor a manufacturing process for identifying special causes and signals for correct actions in the process is a critical and important approach to evaluate processes for improving quality and reducing reworks and scraps. So that their effects can be found and the necessary preventive action can be suggested before a large number of nonconforming products are manufactured. The determination of variability that can follow by CCS affects the cost and quality of process. To eliminate the high scrap ratio, rework cost, and to ensure customer satisfaction, CCs are effective quality tools to determine whether a process is in-control or out of control in a production environment. Currently, there are many applications in the industry where quality characteristics of a product or a process are analyzed by CCs. Monitoring of some types of quality characteristics are not correct, if the quality-related characteristics cannot be represented in numerical forms, such as characteristics for appearance, softness, color, then CCs for attributes are used. The theory of classical CCs requires all the data to be exactly known. But sometimes, we need human judgments and evaluations to construct CCs or process' uncertainties cannot be defined by crisp numbers. In these cases, we can successfully use the fuzzy set theory (FST) to design CCs based on attributes. The major contribution of FST is its capability of representing vague data or managing uncertainty. FST is a systematic base in dealing with situations, which are ambiguous or not well defined. Fuzzy control charts (FCCs) based on human judgments, uncertainties, or vagueness are utilized when vague data is used as real-valued interpretations of uncertainty and vagueness. Recently, extensions of fuzzy sets are often used to manage unceratinties of process. We also know that one of the extensions of FST named type-2 fuzzy sets has fuzzy membership degrees that ability for more flexible modeling uncertainties than type-1 fuzzy sets or traditional fuzzy sets. This paper presents a design of CCs for nonconformities called control charts for attributes based on type-2 fuzzy sets and a real case application in the automotive sector. The control limits (CLs) and central limit (CL) values have been obtained. Two wellknown attributes control charts named c. and u have been designed and analyzed based on type-2 fuzzy sets for this aim. The obtaid results show that it provides a more sensitive and flexible evaluation opportunity compared to the CCs.