Cluster Analysis Of Junior K3 Construction Training Participants Based On Competency Using K-Means Clustering Method

Nanda Fitria; Arief Wibowo; Qodariah

doi:10.63922/ijebir.v5i04.4246

Nanda Fitria Master of Management Study Program, Universitas Budi Luhur
Arief Wibowo Lecturer, Master of Management Study Program, Universitas Budi Luhur
Qodariah Lecturer, Master of Management Study Program, Universitas Budi Luhur

DOI: https://doi.org/10.63922/ijebir.v5i04.4246

Keywords: Competency, K3 Construction Training, K-Means Clustering, Data Mining, Human Resource Management.

Abstract

This study aims to analyze the characteristics and grouping of participants in the Junior K3 Construction Training based on competency using the K-Means Clustering method. Participant competency was measured using three variables: pre-test scores, post-test scores, and length of employment. The study employed a quantitative approach with a data mining method. The research population consisted of 54 participants of the Junior K3 Construction Training, all of whom were included as the sample using a saturated sampling technique. Data processing was carried out using RapidMiner Studio software through the stages of data cleaning, data selection, data normalization, clustering process, and centroid analysis.The results show that the K-Means algorithm successfully grouped participants into three clusters. Cluster 0 consists of 12 participants (22.22%) with high competency characteristics and relatively high work experience. Cluster 1 consists of 7 participants (12.96%) with the highest work experience but relatively lower initial competency compared to Cluster 0. Cluster 2 is the largest cluster with 35 participants (64.82%) who have developing competency levels and relatively lower work experience. Centroid analysis shows that the length of employment variable is the main differentiating factor between clusters, while post-test scores are relatively high across all clusters. The findings indicate that the training program was able to improve participant competency evenly despite participants having varying levels of work experience. The research results can be used as a basis for formulating more effective competency development strategies through a participant-characteristic-based training approach.

References

Dessler, G. (2020). Human Resource Management (16th ed.). Pearson.
Ezugwu, A. E., Ikotun, A. M., Oyelade, O. O., Abualigah, L., Agushaka, J. O., Eke, C. I., & Akinyelu, A. A. (2023). A comprehensive survey of clustering algorithms. Artificial Intelligence Review, 56(3), 1921–1984.
Han, J., Kamber, M., & Pei, J. (2022). Data Mining: Concepts and Techniques (4th ed.). Morgan Kaufmann.
Hendrastuty, N. (2024). Penerapan K-Means Clustering untuk pengelompokan hasil belajar peserta pelatihan. Jurnal Informatika dan Sistem Informasi, 12(1), 45–55.
Nababan, R. (2023). Analisis kinerja dosen menggunakan metode K-Means Clustering. Jurnal Teknologi Informasi dan Manajemen, 8(2), 101–110.
Nursyi, A., Pratama, D., & Hidayat, R. (2024). Implementasi data mining untuk segmentasi peserta pelatihan berbasis kompetensi. Jurnal Sistem Informasi Indonesia, 9(1), 33–44.
Sakalessy, M., Rumawas, W., & Tumbel, A. (2024). Human resource competency and organizational performance. International Journal of Management Studies, 11(2), 55–67.
Spencer, L. M., & Spencer, S. M. (1993). Competence at Work: Models for Superior Performance. John Wiley & Sons.