<p>The growing complexity of deepwater oil exploration presents considerable challenges for naval and offshore engineering, particularly regarding the reliable station-keeping of floating production units under harsh environmental conditions. Among these, the mooring system plays a critical role in ensuring positional stability. Its design must adhere to stringent performance standards while balancing cost-efficiency and structural integrity. Advances in computational tools and Artificial Intelligence (AI) have enabled more efficient evaluation of complex design scenarios. This paper presents a structured three-stage methodology that combines multi-objective optimization with surrogate modeling to support the design of offshore mooring systems. The main contribution of this work does not lie in the modeling technique itself, but in the way consolidated AI methods are applied and validated in a real offshore mooring system design context. In Stage 1, the problem is modeled in the Synapse Offshore platform and solved using the Non-dominated Sorting Genetic Algorithm (NSGA-II), aiming to minimize the positional offset from the neutral point. A case study is performed on an FPSO with 24 mooring lines, assessed under 56 critical environmental conditions selected from a database of over 1700 scenarios. In Stage 2, a surrogate model based on Artificial Neural Networks (ANNs) is trained using simulation data, incorporating normalization, cross-validation, and hyperparameter tuning. In Stage 3, the trained ANN is integrated into the optimization loop, replacing the original solver and accelerating the design process by approximately threefold. Although the techniques employed are widely established, the validated implementation illustrates how AI-assisted modeling can enhance computational efficiency without compromising accuracy in realistic offshore applications.</p>