Substitution-boxes (S-boxes) are essential nonlinear elements in modern symmetric cryptographic systems, offering confusion and resilience against cryptanalysis. The rising need for secure communication highlights the imperative to develop resilient S-boxes with formidable cryptographic properties. This work aims to develop novel 8×8 S-boxes utilizing a hybrid approach that integrates the flower pollination algorithm (FPA) with chaotic maps. The suggested method uses a one-dimensional discrete chaotic map to generate high-quality initial S-boxes, addressing the constraints of traditional FPA techniques that may become trapped in local optima. These preliminary solutions facilitate an effective optimization procedure. The FPA metaheuristic optimizes the S-box configurations by refining a fitness function that embodies critical cryptographic requirements. The produced S-boxes are assessed using conventional metrics, such as bijectivity, rigorous avalanche criterion, nonlinearity, input/output XOR distribution, bit independence criterion, and maximal predicted linear probability. The results are juxtaposed with alternative approaches for S-box generation utilizing optimization techniques. The results demonstrate that the suggested technique generates S-boxes with robust cryptographic attributes and enhanced resilience to various forms of cryptanalytic assaults. This paper offers both theoretical and practical advances by presenting an effective hybrid optimization framework for S-box designs, enhancing the security of contemporary encryption systems.