Constraints on membership functions of rules in fuzzy expert systems
Ismail Burhan Turksen, Yuan Tian · 2002
The compositional rule of inference used in fuzzy expert systems is investigated for multiple rules. Constraints on membership functions of rules are analyzed under the Sub-T composition and the R-implication from the perspective of the basic requirement for fuzzy reasoning. A graphic method is suggested to obtain the bounds on overlaps between membership functions of rules. An example is given to illustrate the impact of the overlaps on the inference result.>