EFFECTIVE METHODS AND ALGORITHMS FOR CREATING A PERSONNEL ASSESSMENT MODEL AND RATING SYSTEM

Abstract

This paper explores effective approaches for building an employee evaluation and ranking system aimed at improving work quality in small and medium-sized enterprises (SMEs). The main objective of the study is to compare the performance of three widely used methods (AHP, TOPSIS, and XGBoost) and to assess their applicability in the SME context. Using an open HR analytics dataset, a set of evaluation criteria was defined, and criterion weights were computed through the Analytic Hierarchy Process (AHP). A multi-criteria employee ranking was then generated using the TOPSIS method, producing an integrated performance score for each employee. In addition, a data-driven predictive model based on the XGBoost algorithm was developed to generate rankings from patterns learned in the dataset. The results indicate that expert-based and multi-criteria decision-making methods provide strong interpretability and ease of implementation, while machine learning approaches offer higher predictive potential when sufficient and high-quality data are available. The study concludes that a hybrid framework combining expert weighting and data-driven modeling represents a promising solution for SMEs seeking practical and reliable employee ranking systems.