Machine learning based solution for predicting voluntary employee turnover in organization

Judrups, J.; Cinks, R.; Birzniece, I.; Andersone, I.

The cost associated with employee turnover and the shortage of available workforce in the market creates a situation where employee retention is crucial for the successful operation of an organization. Working remotely, especially in a situation with COVID-19 pandemic, increases the risks of voluntary employee turnover, since it makes the judgment of employee attitudes more difficult for the managers. Voluntary employee turnover (VET) measurements are one of the key indicators for evaluating the effectiveness of personnel management practices in organizations. The paper proposes a solution to decrease the risk of voluntary employee turnover in organizations. The authors propose a machine learning based model to identify the employees prone to voluntary employee turnover based on the employee data gathered and stored by the organization. The model will allow the managers to make a prediction based on data of the risks associated with voluntary employee turnover and to adjust the decision making process based on the information. To create the proposed IT solution for predicting the voluntary employee turnover analysis of models describing it has been performed to identify the most important factors that influence it. 9 factor groups with 67 factors of VET have been identified during the analysis. In the next step, 46 data clusters relevant for the decision making have been identified in specific organization and data from the clusters retrieved for the analysis. Based on the analysis a model for machine learning will be created, developed, and validated for the use in organizations. © 2021 Latvia University of Life Sciences and Technologies. All rights reserved.