Enhanced fuzzy subjective evaluation method / Shezrin Hawani Amir Hamzah

It has become a rising concern nowadays where a reliable subjective evaluation performance is necessary in order to measure the level of performance in an organization. Thus, this research begins with analyzing the fuzzy evaluation, including fuzzy rule extraction methods, weight generation method and normalization method available in the literature. Fuzzy approach has been applied in some of the previous studies used in assessment and evaluation methods. However the drawback of the fuzzy subjective evaluation method in using rules and weight has been proposed based on the expert’s view. They failed to integrate or ensure they meet the organization objective in the evaluation process. Hence, the aims of this thesis are to make an enhancement on the existing subjective evaluation method by Othman et al. (2008) and to introduce three main elements which are weight generation, rule extraction method and normalization method in subjective evaluation method. In the second stage of the thesis, five data sets from previous studies were used to validate the subjective evaluation method. The research experiment analyzed the properties of fuzzy rules generated in terms of the total number of rules, size and length. The normalization method was used for all types of data and weight generation method from data collection. The results of the analysis have shown that the consistency of ranking was validated by comparing the performance of the proposed method to several subjective evaluation methods. Hence, the proposed method was able to produce comparable results in fuzzy environments, tn conclusion, the proposed enhanced fuzzy subjective evaluation method offered another alternative framework with efficient approach towards evaluation process. This will contribute to a better evaluating process which reduces the need of human thinking, thus saves costs, reduces time, and helps to increase the high consistency.