Collaborative Research: Leveraging Noncontact Dimensional Metrology to Understand Complex Part-to-Part Variation

该奖项的目标是开发一种范例，用于识别和可视化高维空间密集光学坐标测量机（OCMM）数据中复杂的零件到零件变化模式。使用激光和/或视觉系统进行非接触式尺寸测量的ocmm是离散零件制造中最有前途的质量控制新兴测量技术之一。它们产生大量的轮廓、点云和高分辨率图像数据，这些数据表示参数和非参数表面几何特征。该奖项对零件间变化的强调与当前OCMM数据分析软件的强调根本不同，OCMM数据分析软件将几何特征单独适用于单个零件。为了创建理解变化的拟议范例，将开发用于识别和可视化变化模式的流形学习框架，解决挑战，包括同时识别非参数和参数变化模式，处理与流形学习中通常假设的不同的测量噪声结构，转换OCMM数据以减少模式中的非线性程度，并处理在不同过程阶段获得的异构数据类型。如果成功，这项研究的结果将提供一个强大的工具，以促进发现和消除生产差异的主要根本原因。许多离散零件制造行业在OCMM设备上投入了大量资金，但缺乏知识发现工具来充分利用该技术来了解零件之间的变化。这项研究将通过创建一种更有效地分析高维、空间密集OCMM数据的方法来填补一个关键的空白。它将为采用现代OCMM技术的六西格玛制造项目提供更复杂、更迫切需要的减差工具，这将提高美国制造商的竞争力。它还将允许OCMM技术的更大投资回报，预计将增加对硬件/软件系统的需求，并激发技术的进一步发展。

The objective of this award is to develop a paradigm for identifying and visualizing complex part-to-part variation patterns in high-dimensional, spatially dense optical coordinate measuring machine (OCMM) data. OCMMs for noncontact dimensional metrology using laser and/or vision systems are one of the most promising emerging measurement technologies for quality control in discrete parts manufacturing. They produce large volumes of profile, point cloud, and high resolution image data that represent parametric and nonparametric surface geometry characteristics. The emphasis of this award on part-to-part variation is fundamentally different than the emphasis of current OCMM data analysis software, which fits geometric features separately to individual parts. To create the proposed paradigm for understanding variation, a manifold learning framework for identifying and visualizing the variation patterns will be developed, addressing challenges that include simultaneously identifying nonparametric and parametric variation patterns, handling measurement noise structure that differs from what is typically assumed in manifold learning, transforming OCMM data to reduce the extent of nonlinearity in the patterns, and handling heterogeneous data types obtained at different process stages.If successful, the results of this research will provide a powerful tool to facilitate the discovery and elimination of major root causes of manufacturing variation. Many discrete parts manufacturing industries invest heavily in OCMM equipment but lack knowledge discovery tools for fully utilizing the technology to understand part-to-part variation. This research will fill a critical void by creating a methodology for more effectively analyzing high-dimensional, spatially dense OCMM data. It will provide more sophisticated, badly needed variation reduction tools suitable for manufacturing six-sigma programs that employ modern OCMM technology, which will increase the competitiveness of US manufacturers. It will also allow for a greater return on investment in OCMM technology, which is expected to increase demand for the hardware/software systems and spark further advances in the technology.