A Comparative Study of the Defuzzification Methods in an Application
Nidhi Kataria · SSRN Electronic Journal · 2010
This paper briefly reviews some of the different defuzzification methods reported in literature, using the base problem of feeding according to the protein and energy level. The defuzzification methods studied are: centroid, bisector, Largest of Maximum (LOM), Mean of Maximum (MOM) and Smallest of Maximum (SOM). In the problem under consideration, two input parameters - energy and protein, and one output parameter (feeding) - have been taken. In this work, four rules have been made, which decide the level of feed required. Ten observations for energy and protein levels are taken and the fuzzy rules are applied on them to find a solution. The fuzzy results obtained are then defuzzified using the five methods stated above. From the results obtained for the different observations and different defuzzification methods, it can be concluded that centroid, bisector and MOM give almost the same values. Therefore, there is more consistency in these results. On the other hand, there is a wide variation in the results obtained from LOM and SOM.