In this paper, we extend the numerical solution of linear Volterra integral equation systems of the second kind by incorporating
fuzzy numbers into the model. While the classical approach relies on compactly supported radial basis functions (RBFs) of Wendland type, our contribution reformulates the problem under uncertainty by representing the input data and kernels as fuzzy quantities. Using the α-cut representation, the fuzzy Volterra integral system is transformed into a family of crisp integral equations, which are then approximated by the RBF variational method. The fuzzy solution is reconstructed from the left and right bounds of these crisp solutions across all α-levels. Convergence in the fuzzy framework is analyzed through the Hausdorff metric, ensuring that the proposed method inherits the convergence properties of the classical scheme. Numerical experiments with triangular fuzzy numbers demonstrate the applicability of the method and show that it provides reliable interval-type solutions that capture the inherent uncertainty of the system. This approach provides a promising numerical framework for Volterra integral equation systems involving uncertain or imprecise data. The presented numerical examples demonstrate the construction of fuzzy solution bands and the decay of the endpoint errors as the number of Wendland centers increases. Future work should investigate adaptive center selection, nonlinear fuzzy Volterra systems, Volterra–Fredholm extensions, and applications to engineering and biological models involving uncertain data.

In this paper, we extend the numerical solution of linear Volterra integral equation systems of the second kind by incorporating
fuzzy numbers into the model. While the classical approach relies on compactly supported radial basis functions (RBFs) of Wendland type, our contribution reformulates the problem under uncertainty by representing the input data and kernels as fuzzy quantities. Using the α-cut representation, the fuzzy Volterra integral system is transformed into a family of crisp integral equations, which are then approximated by the RBF variational method. The fuzzy solution is reconstructed from the left and right bounds of these crisp solutions across all α-levels. Convergence in the fuzzy framework is analyzed through the Hausdorff metric, ensuring that the proposed method inherits the convergence properties of the classical scheme. Numerical experiments with triangular fuzzy numbers demonstrate the applicability of the method and show that it provides reliable interval-type solutions that capture the inherent uncertainty of the system. This approach provides a promising numerical framework for Volterra integral equation systems involving uncertain or imprecise data. The presented numerical examples demonstrate the construction of fuzzy solution bands and the decay of the endpoint errors as the number of Wendland centers increases. Future work should investigate adaptive center selection, nonlinear fuzzy Volterra systems, Volterra–Fredholm extensions, and applications to engineering and biological models involving uncertain data.