Graphical fuzzy algebra applied to schedulability analysis of real time systems

This paper proposes a graphical representation for the computation of complex functions with two fuzzy arguments. It is based on the theory of evidence, and has been compared with the theory of falling shadows with similar results for the AND operation. It is graphically shown that no general correspondence between grades of membership correlation and t-norms can be established when the extension principle is applied. Some new interpretations of membership correlation, based on imprecise observations, are provided. The proposed method has been used to solve real-time schedulabilty analysis with fuzzy task execution times. The example shows that interval computations can be performed to reduce the number of time consuming simulations.