Cyber-Physical Systems are inherently complex. Reducing the design complexity of such systems, is beneficial for scalability of the design. In this paper, a method for design optimization of these systems is proposed to facilitate the decomposition of the optimization problem for the whole system into smaller sub-problems and coordination of modular solutions to reach the desired optimum. The coordinated solutions are either identical to the optimal solution of the complex optimization problem for the system as a whole or within an acceptable error margin of it. To demonstrate the efficacy of the method, it is applied to a mechatronic case study. The results provide evidence for the potential feasibility of the methodology in terms of meeting the requirements on the solutions, while reducing the computational demand of the design process.