Abstract: A reliability-based design optimization method is proposed in this paper for the deep-water composite catenary risers. Since the load carrying capacity of anisotropic laminated structures is very susceptible to changes in the sequence, orientation, and thickness of plies, it is necessary to introduce a reliability index as a boundary condition during the optimization process. Compared to traditional steel risers, the composites have a broader application prospects due to several advantages, such as high specific strength and stiffness as well as high fatigue and corrosion resistance. The classic lamination theory is used to compute the equivalent mechanical properties of composite risers, and the global model is then established to obtain the response of the key section that serves to be the boundary for the local model. In addition, the Design of Experiment （DOE） method is used to construct an approximate model, and the Monte Carlo method is utilized to analyze the result of the Deterministic Optimization（DO） and the Reliability-Based Design Optimization （RBDO）. This study shows that the optimized results not only satisfy the strength requirements but also possess a lower weight and a better reliability， which are feasible for composite catenary risers.