Abstract

This research applies unsupervised machine-learning techniques to explore agile workstyle segmentation within organizational contexts. Using the Employee HR Dataset from Kaggle, comprising 14,999 employee records with variables such as satisfaction, performance evaluation, workload, and tenure, the study aims to identify natural behavioral clusters that reflect diverse productivity profiles. After comprehensive preprocessing—including data cleaning, feature selection, and Min–Max normalization—the K-Means clustering algorithm was implemented to group employees into homogeneous segments. The optimal number of clusters (𝑘=4) was determined through the Elbow Method and Silhouette Coefficient (0.57), ensuring statistical validity and interpretability. The resulting clusters revealed four distinct agile workstyle archetypes: Collaborative Core Workers (balanced and satisfied), High-Impact Performers (high evaluation and motivation), Low-Engagement Staff (underutilized and less satisfied), and Overloaded Experts (high performance but low satisfaction). These profiles provide a multidimensional perspective on workforce diversity, connecting quantitative analytics to agile management principles such as sustainable pace, self-organization, and continuous improvement. Visualization through Principal Component Analysis and boxplots confirmed clear separations among clusters, validating the algorithm’s ability to capture meaningful behavioral distinctions. Findings indicate that unsupervised learning can effectively support agile HR decision-making by quantifying intangible dimensions of engagement and workload balance. The approach demonstrates that machine learning extends beyond predictive modeling into strategic diagnostics for workforce optimization. Practically, the results guide agile managers in identifying potential burnout risks, reinforcing engagement programs, and sustaining team motivation. The study concludes that integrating explainable data-driven insights into agile HR practices enhances transparency, adaptability, and organizational resilience—cornerstones of modern agile transformation.

Keywords: Agile Management, Human Resource Analytics, K-Means Clustering, Machine Learning

How to Cite:

Setiawan, I. & Wulandari, A. N., (2026) “Clustering Agile Workstyles and Productivity Profiles Using KMeans Machine Learning on Employee HR Data for Workforce Sustainability ”, Agile Management 1(1): 123, 1-16. doi: https://doi.org/10.63913/am.v1i1.123