Dimensional Variation Analysis for Rigid Part Assembly With an Improvement of Monte Carlo Simulation

Dimensional variation has significant effect on the quality of product. Recently, Monte Carlo (MC) simulation is widely used in dimensional variation analysis, with high accuracy and adaptability, but there is the problem of low computational efficiency. Aiming to address this problem, an improvement of MC simulation is proposed through a two-phases solution. In the first phase, surrogate model is used to approximate the locating constraint equations for a 3D part, which reduces the nonlinear coupling between dimensional deviations and avoids large-scale solution of nonlinear equations in dimensional variation analysis on the condition of ensuring the accuracy. In the second phase, random samples used in MC simulation are replaced by low discrepancy sequences, which enable the samples to have better homogeneity and representativeness and reduce the number of samples required in the dimensional variation analysis. Finally, two examples are used to demonstrate the effectiveness of the method, and the results show that the two-phases solution has advantages both in the accuracy and efficiency.