Predictive assessment of staff involvement based on machine learning methods (on the issue of neuro-taylorism)

Abstract

The authors of the scientific article consider the problem of transformation of the classical principles of scientific organization of labor (F. Taylor) in the age of digitalization. The concept of «neuroteilorism» is proposed, which is a synthesis of Taylor timing and neural network algorithms for predictive assessment of staff engagement. Based on Kaggle's open data (n=2000 employees), the authors conducted a comparative study of three machine learning models: Random Forest, direct propagation neural network (MLP), and gradient boosting (XGBoost). The target variable was selected as a binary indicator of low engagement (satisfaction 2.5 on a 5-point scale). The analysis focused on employees from the IT, Marketing, and Finance departments. The results showed that XGBoost achieved the highest accuracy (83.2%), and overtime hours (Overtime_Hours) were the key predictor of burnout. The authors of the scientific article provided practical recommendations for HR departments on digital labor rationing.