Optimal Tolerancing: The Link Between Design and Manufacturing Productivity

这SBIR第二阶段的研究地址的计算环境，最佳尺寸管理的组件和装配。 该方法将遗传优化方法与基于蒙特卡罗的公差测试相结合。 公差范围和标称尺寸均作为设计变量。 目标是生成满足所有设计和制造考虑的最小成本装配，这些考虑可以形成为约束或多个目标。 该技术的成功应用将大大减少产品开发时间和制造成本。 当前的公差分析软件不能以最佳方式分配公差或修改标称尺寸。 传统的非线性规划方法不足以解决高度非线性，多峰问题，从考虑一个“真实的”组件或组件。 遗传算法与容差分析包的组合具有在并行或分布式计算网络上高效操作的优点。 本研究旨在提高遗传优化和公差模型仿真的算法效率和可靠性。

This SBIR Phase II research addresses a computational environment for optimal dimensional management of components and assemblies. The approach couples a genetic optimization method with a Monte Carlo based tolerance test. Both tolerance ranges and nominal dimensions are included as design variables. The goal is to generate the minimum cost assembly which meets all design and manufacturing considerations, which may be formed as constraints or as multiple objectives. Successful application of the technology will dramatically reduce product development times and manufacturing cost. Current tolerance analysis software does not allocate tolerances or modify nominal dimensions in an optimal fashion. Traditional nonlinear programming methods are insufficiently robust to solve the highly nonlinear, multimodal problem resulting from the consideration of a "real" component or assembly. The combination of the genetic algorithm with a tolerance analysis package has the advantage of efficient operation on a parallel or distributed computing network. This research is directed toward increasing algorithmic efficiency and reliability for both the genetic optimized and the tolerance model simulation.