Retirement marks the end of a Civil Servant’s (PNS) service period and requires proper preparation to ensure a smooth replacement process that meets organizational needs. Dinas X of XYZ Regency still faces challenges in managing employee data, particularly in identifying non-teaching employees approaching retirement. This study aims to implement the K-Means algorithm to classify employees based on job level and remaining years of service and to develop a web-based system that provides replacement recommendations using a rule-based matching approach based on Job Position, Competency, Educational Background, and Regional Apparatus Organization (OPD). The clustering process produced three groups: Near Retirement (0–11 years) with 503 employees, Medium Retirement (12–21 years) with 746 employees, and Far from Retirement (22–34 years) with 632 employees. The clustering quality achieved a Silhouette Score of 0.55, indicating reasonably good cluster separation. The clustering results were then used to support employee replacement recommendations. The evaluation results showed that among 62 employees approaching retirement, only 2 had replacement candidates who met all specified criteria. These findings indicate limited availability of suitable replacements. This study can support human resource planning, including job transfers, competency development, and recruitment strategies according to organizational need.
Copyrights © 2026