The article considers the problem of labor productivity growth of the enterprise, taking into account economic, demographic, social factors and subjective information about the quality of staff health. A critical analysis of existing approaches, methods and models in this area has been carried out and a number of significant shortcomings of the presented approaches has been identified that limit the scope of their application. There are no. methods for quantifying the impact of health on labor productivity and developing further recommendations for managing employees health in order to increase their labor productivity. The technology for labor productivity management has been developed based on the phased processing and modeling of quantitative and qualitative data taking into account objective information about economic, demographic, social factors and subjective information about the factors of the health quality. The technology is based on statistical analysis and machine learning, and support managerial decision-making in planning health-saving strategies to increase labor productivity It is proved that to solve the problem of employee clustering and forming their homogeneous groups, it is most advisable to use the k-means method, which is more stable compared to the clustering method based on Kohonen neural networks. In the problem which contains many qualitative variables, such as gender, education, high-quality self-esteem of health for identifying the profile of new employees, the classification method based on the support vector method has demonstrated great efficiency.