The main purpose of the optimization design of Dump truck cab‘s ROPS system is to reduce the weight on the premise that cab’s strength, stiffness and dynamic performance are not significant changed; The cab is made up of many components and each component has different influence on the cab’s performance, so the design belongs to the multi-parameter and multi-objective optimization problems. Significant degree analysis method based on Orthogonal Experimental Design was raised. This method evaluated the significant degree of each parameter through calculating the design parameter’s variance to response. On the basis of these data, the design parameters which had more important influence were chosen as optimization variables, an approximate model between design variables and response was built by using TPS-HDMR, then get the optimized solutions by using NSGA-II genetic algorithm, The optimized results was verified. The experimental results agreed well with the simulation results, which can meet the requirements of the enterprise.