Hierarchical fuzzy inference system for staff performance evaluation: case study

Evaluating the performance of employees is one of the most critical requirements for senior managers in organizations. One of the challenges in this regard is the assessment of the performance of staff personnel. Due to the nature of their activities, defining quantitative indicators alone cannot provide an acceptable evaluation. Therefore, this research focuses on presenting an applied fuzzy model for the evaluation of staff personnel.

This study involves developing a hierarchical fuzzy performance evaluation system composed of both qualitative and quantitative indicators simultaneously. The indicators consist of 7 quantitative and 5 qualitative ones, identified and fuzzified with the input of 11 senior managers of the organization. Considering all indicators simultaneously in a system would require a very large number of fuzzy rules. Therefore, a step-by-step model with continuous and hierarchical fuzzy systems is used in this article, significantly reducing the number of rules. Finally, the developed model is implemented in Simulink, a Matlab tool, using the Mamdani method.

The results of the `evaluation of the developed model closely matched expectations with good accuracy, providing a suitable basis for employee assessments. The advantages of this model include its relatively good accuracy compared to traditional models, higher employee satisfaction, and reducing subjective assessments resulting from bias, relationships, etc. It also enables faster and simpler evaluations for supervisors and managers.

Originality/Value:  

The presented model is a practical approach that has been implemented in a real organization. It offers the possibility for other organizations to conduct employee evaluations by adjusting some indicators according to their activities, thereby adding value to the scientific field.