Fuzzy decision by opinion score method (FDOSM):design and development of new multi criteria decision making method
The purpose of this research was to design and develop a new multi-criteria decision- making (MCDM)method called Fuzzy Decision by Opinion Score Method (FDOSM) to help overcome the problems ofMCDM methods based on the idea of an ideal solution This research used an experimentalresearch design with w...
Saved in:
| Main Author: | |
|---|---|
| Format: | thesis |
| Language: | eng |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://ir.upsi.edu.my/detailsg.php?det=5221 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The purpose of this research was to design and develop a new multi-criteria decision- making (MCDM)method called Fuzzy Decision by Opinion Score Method (FDOSM) to help overcome the problems ofMCDM methods based on the idea of an ideal solution This research used an experimentalresearch design with which FDOSM was applied to individual and group contexts. Essentially,FDOSM contains three main blocks, namely the input data block, data transfer block, and dataprocessing block. For the data processing block, three sets of experiments were carried out tooptimize the parameters of the proposed method. The first experiment dealt with threedifferent configurations, namely Direct aggregation, Compromise Rank, and Distancemeasurement, of a single decision maker. Direct aggregation with arithmetic mean is the main configuration recommended for comparing the results of different experiments.However, if the maximum utility is important to the decision maker, compromise rankingwould be the proper configuration. The second experiment focused on the process ofGroup Fuzzy Decision by Opinion Score Method (G- FDOSM) with two different configurations,namely internal and external aggregation. The main finding of G-FDOSM experiment showed the resultsof internal and external configurations were close, with the ratio of the closeness of theexperimental results of G-FDOSM with 90 alternatives being 71.02%. However, externalaggregation was deemed more appropriate for compromise ranking. The third experimentinvolved several different case studies to examine the suitability of FDOSM in solving differentMCDM problems. The results showed that, compared to the ideal solution, the best player (P16)achieved a ratio of 58.3% from the ideal solution, which was considered to be the best ratio amongother players for the sports science case study. For the GPS case study, experimental resultsshowed the best solution was m8 with a ratio of 67% from the ideal solution. Overall, the resultsof FDOSM and G-FDOSM were close to the humans opinions, suggesting that arithmetic mean is themost suitable aggregation operator for all the experiments and FDOSM can adopt different fuzzymembership. Furthermore, reference comparison used with FDOSM can be implemented moreefficiently compared to the use of the pairwise comparison of the Analytic Hierarchy Process andthe Best-Worst Method. In conclusion, the proposed FDOSM had been successfully modulatedmathematically, tested with different numerical examples, andcompared to other MCDM methods. |
|---|