February 18, 2022

About the speaker

Dr. Zhaohui Geng is an assistant professor in the Department of Manufacturing and Industrial Engineering at The University of Texas Rio Grande Valley. He received his Ph.D. in Industrial Engineering from the University of Pittsburgh in 2021, his Master of Arts in Statistics and Master of Science in Industrial Engineering from the University of Pittsburgh in 2018 and 2016, respectively, and his Bachelor of Engineering in Electronic Science and Technology from Nankai University, China. Dr. Geng is broadly interested in developing statistical inferential methodologies, statistical/probabilistic machine learning algorithms, and large-scale optimization methodologies to solve the problems at the intersection of production systems, methodology, and advanced manufacturing.

Volumetric Data Analysis for Reverse Engineering and Solid Freeform Additive Manufacturing: A Framework for Geometric Metrological Analysis

The poor geometric quality of parts made by additive manufacturing (AM) and other advanced manufacturing processes is a major constraint in their wide industry adoption. Conventional metrological methodologies overlook the three-dimensional (3D) feature-independent processing of these techniques. A 3D point cloud collected by 3D scanning or reverse engineering (RE) has the potential to provide a detailed description of the part, but it is difficult for analysis due to the complex data structure and large data volume. In order to add “intelligence” to the process, a novel statistical data analysis framework, volumetric data analysis (VDA), is developed to extract process knowledge of RE and AM from the 3D point cloud samples, and improve model and product accuracy using analytical, especially statistical, methodologies.

In this talk, we first provide the essential technical foundation and major steps of the VDA framework, where the goal is to bridge the gap between complex closed-shape 3D data structures and advanced multivariate analytical methodologies. This framework is then applied to solve process planning, variation modeling, and tolerancing problems in both traditionally machined parts and general freeform RE, especially laser scanning, and remanufacturing applications.

Several applications of VDA are suggested to transform state-of-the-art analytical methodologies in the modeling/metrology of additive manufactured parts. By utilizing the VDA framework, 3D point clouds can be analyzed using the design of experiments for offline AM quality control. This framework can also be applied for configuration prediction and process parameter optimization for product quality improvement.