Abstract: To deal with the low operation rate of the firing mechanism of ammunition fuze, robustness analysis and probabilistic design methods were introduced into the fuze design to carry out system simulation research on performance prediction and optimization of the firing mechanism of ammunition fuze. Firstly, the dynamic model of the instant firing mechanism was established, and the finite element simulation method was used to verify that the mechanism could meet the design requirements under the nominal value, that is, the fuze could fire reliably. Secondly, Monte Carlo simulation based on dimensional tolerance dispersion was carried out to evaluate fire performance, and not all the obtained target value dispersion data could meet the design requirements. Then, the contribution rate of each dimension variable to the target value was obtained by using the DOE (design of experiment) of optimal Latin hypercube, and the Pointer algorithm was used to optimize the variable tolerance with high contribution rate. Finally, the optimized results were verified, and the results showed that all the optimized target value distribution data met the design requirements. By analyzing and optimizing the robustness of fuze firing mechanism, it is of great significance to find a balance between performance reliability and manufacturing cost to improve product performance and reduce manufacturing cost.