Modified divergence measures based on fuzzy MEREC and TOPSIS for staff performance appraisal

The aim of this study is to establish a divergence measure integrated with the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) approach for crisp evaluation that can overcome limitation of previous divergence measures, as well as to describe its properties. The proposed...

Full description

Bibliographic Details
Main Author: Saidin, Mohamad Shahiir
Format: Thesis
Language:English
Published: 2023
Subjects:
Online Access:http://psasir.upm.edu.my/id/eprint/119040/
http://psasir.upm.edu.my/id/eprint/119040/1/119040.pdf
Description
Summary:The aim of this study is to establish a divergence measure integrated with the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) approach for crisp evaluation that can overcome limitation of previous divergence measures, as well as to describe its properties. The proposed divergence measure has been enhanced by utilising fuzzy α-cut, in which experts can identify a wide range of rankings when their levels of confidence vary since uncertainty or ambiguity is an essential feature of multi-criteria decision-making (MCDM) cases. This study also provides a modified technique, the fuzzy MEthod based on the Removal Effects of Criteria (MEREC), by modifying the normalisation technique and enhancing the logarithm function used to assess the entire performance of alternatives in the weighting process. The comparative analyses are conducted through the case studies of staff performance appraisal at Universiti Putra Malaysia (UPM) and Universiti Malaysia Perlis (UniMAP) that consist of 6 and 13 sub-criteria, respectively. The simulation-based study is used to validate the effectiveness and stability of the proposed method. Regarding correlation coefficients and central processing unit (CPU) time, the findings of this study were compared to those of other MCDM methodologies. Based on the results, the proposed technique performed in a manner consistent with the current distance measure approaches since all of the values of the correlation coefficient were greater than 0.8. Besides, the proposed technique provides the advantage of being able to assess all potential score values of alternatives, including 0 and 1. Furthermore, the simulationbased study demonstrates that even in the presence of outliers in the collection of alternatives, fuzzy MEREC is able to offer consistent weights for the criterion. Since the criteria weights significantly affect the results of rankings, the sensitivity analysis is used to reveal how the rankings change due to the variation of criteria weights, which mainly explores the influence of single criterion weight changes. The correlation coefficient values between the original rankings and the rankings with decreasing and increasing criteria weights are presented. Based on the analysis, the most affecting criterion to the ranking of staff performance in each category has been identified. In addition, it has been identified that the proposed technique has the shortest CPU time when compared to the other divergence measurement methodologies. As a result, the proposed technique provides more sensible and practicable results than the others in its category.