Özet:

With the development of technology, large databases are becoming more accessible. Today, it is possible to use large databases in many fields. By using data kept in databases and data mining approaches, meaningful information and rules are discovered. Thus, information discoveries are made that will be useful for businesses. In this study, the production data of a factory producing surgical (medical) masks was used. Clustering analyzes of the variables that are effective in the production of faulty or error-free masks in the quality control stage were made with the Two Step method and the properties of the resulting clusters were compared. The most effective variables in cluster partitioning are; It has been observed that the amount of speed used in the creation of the mask body, the type of fabric of the middle layer of the mask body consisting of 3 layers, and the amount of ultrasonic heat used to create the pleats on the body. The information obtained to improve the performance of production activities will be used to improve the process.

Anahtar Kelimeler: