PREDICTIVE ANALYTICS FOR EMPLOYEE TURNOVER: A COMPARATIVE STUDY BETWEEN INDUSTRIES

Intan Susilawati (Universitas Riau Kepulauan)
Oktavianti (Universitas Riau Kepulauan)
Rizki Eka Putra (Universitas Riau Kepulauan)

Multidiciplinary Output Research for Actual and International Issue (Morfai Journal)
Vol. 5 No. 6 (2025): Multidiciplinary Output Research For Actual and International Issue

Employee turnover poses a significant challenge across industries, yet its drivers are often assumed to be universal. This study challenges that assumption through a comparative analysis of predictive analytics in the technology, healthcare, and manufacturing sectors. Utilizing human resources data and machine learning models, we identified profoundly industry-specific predictors and model performances. Results revealed distinct turnover dynamics: career-centric in technology, well-being-driven in healthcare, and structurally transactional in manufacturing. Consequently, no single predictive algorithm was universally superior. The discussion concludes that effective turnover prediction and mitigation require tailored, context-aware models aligned with the unique operational and psychological realities of each industry, rendering one-size-fits-all HR strategies obsolete and advocating for a decentralized analytical approach.