Employee Mutation Map for PLN Employee Placement Using Machine Learning

Abstract

 This research aims to develop an employee transfer mapping system for PLN employee placement using a machine learning approach, specifically K-means clustering. The dataset used includes internal PLN data such as employee job history, asset distribution, operational performance, as well as external data from BPS (Badan Pusat Statistik) that includes regional characteristics, demographics, and infrastructure [1][2]. The data preprocessing process involves handling missing values, normalizing numerical features, and encoding categorical features to ensure data quality and consistency [3][4]. The K-means algorithm is applied to cluster units into categories such as Technical (TEK), Marketing (SAR), and Energy Transaction (TEL) based on selected features [5][6]. After clustering, each unit is given a unique identification that describes the characteristics of the cluster and region [7]. The clustering and identification results are
visualized using Convex Hull and PCA in two-dimensional (2D) and three-dimensional (3D) formats [8][9], demonstrating a clear separation of categories and subcategories within PLN units. Therefore, the results of this research show that the use of clustering can effectively support decision-making processes related to employee placement [10].