Optimal Robust Fixture Configuration Design Using Computer Experiment
A new FEA based design approach of optimal robust fixture configuration is proposed in this paper, which employs a surrogate model through computer experiment to significantly reduce the intensive computing effort involving numerous FEA system response evaluations. The effects of the fixture variability to the workpiece performance variability are assessed through an efficient robustness evaluation method, First Order Reliability Method (FORM), based on the surrogate computer model. Not restricted to primary datum surface, this new approach enables simultaneous determination of robust locator/clamp locations and clamping forces for a deformable workpiece and thus captures interaction between locating and clamping. The effectiveness of this approach is illustrated though an application example. The results of robustness analysis reveal new information and suggest that the optimal solution resulted from deterministic optimization may not be the best solution when the design is subjected to variability.