One of the major challenges of implementing additive manufacturing (AM) processes for the purpose of production is the lack of understanding of its underlying process-structure-property relationship. Parts manufactured using AM technologies may be too inconsistent and unreliable to meet the stringent requirements for many industrial applications. The first objective of the present research is to characterize the underlying thermo-physical dynamics of AM process, captured by melt pool signals, and predict porosity during the build. Herein, we propose a novel porosity prediction method based on the temperature distribution of the top surface of the melt pool as the AM part is being built. Advance data analytic and machine learning methods are then used to further analyze the 2D and 3D melt pool image streams to identify the patterns of melt pool images and its relationship to porosity. Furthermore, the lack of geometric accuracy of AM parts is a major barrier preventing its use in mission-critical applications. Hence, the second objective of this work is to quantify the geometric deviations of additively manufactured parts from a large data set of laser-scanned coordinates using an unsupervised machine learning approach. The outcomes of this research are: 1) quantifying the link between process conditions and geometric accuracy; and 2) significantly reducing the amount of point cloud data required for characterizing of geometric accuracy.
Dr. Mojtaba Khanzadeh received his B.Sc. and M.Sc. in Industrial Engineering from Sharif University of Technology in 2013 and 2015, respectively, and his Ph.D. in Industrial Engineering from Mississippi State University in 2019. He also holds an M.A. in Statistics from Mississippi State University. Dr. Khanzadeh has been a machine learning scientist since 2019 in Entefy, Inc. and he is soon joining Amazon.com Services LLC as a data and applied scientist. During his career in Entefy, Inc. he also served as an assistant professor in Davis School of Business in Colorado Mesa University, and School of Business and Economics at SUNY Plattsburgh. Dr. Khanzadeh’s research focuses on the effective utilization of high-dimensional, image-based data streams to monitor and predict the performance of complex metal printing processes. From a methodological viewpoint, his expertise focuses on the development of new tensor-based, feature extraction methodology for big data (at the scale of terabytes) generated from complex engineering systems; and use of the developed method for the purpose of prediction, diagnosis, and optimization. The methodologies that he has developed are not limited to Additive Manufacturing applications, rather they have direct applicability to many other types of application domains including, healthcare applications, cybersecurity, power generation systems, financial engineering, smart grids, and other data-rich systems. He is a member of the Institute for Operations Research and the Management Sciences (INFORMS), the Institute of Industrial and Systems Engineers (IISE), and American Society of Mechanical Engineers (ASME).