Runge-Kutta method for solving fuzzy differential equations under generalized differentiability (Q725760)

A Runge-Kutta method of order four is developed for solving first-order fuzzy ordinary differential equations based on a type of characterization theorem proposed by \textit{Y. Chalco-Cano} and \textit{H. Román-Flores} [Chaos Solitons Fractals 38, No. 1, 112--119 (2008; Zbl 1142.34309)]. The authors develop the method based on RL fuzzy numbers and discretize it to a system of two ODEs. Thereafter, the system is solved using the aforesaid method to get an approximate fuzzy solution in the sense of triangular fuzzy numbers. The authors consider the method for two cases. The first one is based on Hukuhara differentiability, the second one on generalized Hukuhara differentiability. The fuzzy ordinary differential equations are classified and solved accordingly to these two types of fuzzy differentiability. A type of an exponential decay model is employed to validate the proposed approach for both cases. The approximate solutions are compared with the exact ones.