This paper investigates the problem of optimization method for coupling parametrization design of complex space system. The objective is to develop an integrated optimization framework to design complex space system. An improved collaborative optimization (ICO) strategy for complex design using hybrid optimization algorithm is proposed, which associates genetic algorithm (GA) with tabu search (TS). A multi-objective function with penalty terms for the system-level optimization and a relax factor for the subsystem optimization are adopted, which can avoid unsolvability or converging difficulty. The results demonstrate the applicability and effectiveness of the proposed design methodology.