Employee performance improvement using clustering method: A case study in PT. XYZ

The garment industry is part of the manufacturing industry which is currently growing rapidly. Therefore, the level of competition will be increasing among competitors. PT XYZ is a garment company located in the city of Semarang. During the production process, defective product that occurred when orders increased. The product defect impacts the company getting a penalty and several other losses. Product defects occur due to a decrease in the performance of employees. Then through this research, researchers want to know the characteristics performance of employees in the sewing department and will be grouped using the clustering method. Sewing department employees were chosen as research subjects because the focus of garment production is on this department who turn fabrics into finished products. The grouping method aims to determine the factors that can influence the quality of performance and can find out the right recommendations to deal with problems that cause a decrease in performance quality. The clustering process is carried out using the K-Means algorithm because it is one of the algorithms in clustering that is able to group quickly and easily. (Davies Bouldin Index) DBI is also used in this study to help determine the optimum number of clusters formed. This study resulted in 4 clusters with the most frequent mistakes being when combining panels on a garment and when doing construction. The error occurred due to employee negligence and did not follow the applicable SOP (Standard Operating Procedure). Increasing motivation for all clusters can be done by giving bonuses to employees. The recommendation to minimize errors when combining panels is to provide a detailed explanation of the product design sketch. Then for mistakes during construction is to pay attention to the arrangement of threads and increase the brightness of the lights for employees in the sewing department.