A VARIANT PERSPECTIVE TO PERFORMANCE APPRAISAL SYSTEM: FUZZY C - MEANS ALGORITHM

Abstract

Performance appraisal and evaluating the employees for awarding is an important issue in human resource management. In performance appraisal systems, ranking scales and 360 degree are the most commonly used types of evaluating methods in which the evaluator gives a score for each criterion to assess all employees. Ranking scales are relatively simple assessment methods. Despite using ranking scales allows the management to complete the evaluation process in a short time, they have some disadvantages. In addition, although, all the performance appraisal methods evaluated the employees in different ways, the employees get scores for each evaluation criteria and then their performances are evaluated according to total scores. In this paper, the fuzzy c - means (FCM) clustering algorithm is applied as a new method to overcome the common disadvantages of the classical appraisal methods and help managers to make better decisions in a fuzzy environment. FCM algorithm not only selects the most appropriate employee(s), but also clusters them with respect to the evaluation criteria. To explain the FCM method clearly, a performance appraisal problem is discussed and employees are clustered both by the proposed method and the conventional method. Finally, the results obtained by the current system and FCM have been presented comparatively. This comparison concludes that, in performance appraisal systems, FCM is more flexible and satisfactory compared to conventional method.