UNIVERSITY PARK, PA – Manufacturing technology, the various devices and systems that enable the production of manufactured goods, is changing rapidly. Due to its digital nature, the accelerated pace at which manufacturing technology advances often makes it difficult for engineers to incorporate the latest technologies and production processes into product designs.
Penn State researchers received a $ 424,743 grant from the National Science Foundation to study how the size and quality of datasets created by digital models relate to machine learning and how this affects the support provided to technical designers.
The three-year project, “Investigating the Effectiveness of Machine Learning Paradigms in Supporting Engineering Designers in Rapidly Changing Digital Manufacturing,” is led by Principal Investigator Christopher McComb, Assistant Professor of Engineering Design, of industrial engineering and mechanical engineering. Nicholas Meisel, Assistant Professor of Engineering Design and Mechanical Engineering, and Timothy Simpson, Professor Paul Morrow of Engineering Design and Manufacturing, are the Co-Principal Investigators. Mechanical engineering doctoral student Glen Williams also recently joined the project.
It is known that today’s manufacturing machines are often based on digital models which produce large amounts of data. The researchers’ approach revolves around a two-step process that uses these existing datasets to develop design knowledge.
“In the first step, we will use deep learning to extract features, or common recurring patterns, from the database,” he said. “In the second step, we use deep learning again to learn how to use these models to predict performance and behavior – things like strength and manufacturability.”
Data sets vary widely in quantity and quality, making the usefulness of machine learning somewhat unknown. The team will use additive manufacturing, or 3D printing, to investigate how the quantity and value of datasets relate to the accuracy and usefulness of machine learning capabilities and how this relates to the support provided. to technical designers.
McComb said additive manufacturing was chosen as the case study because of its status as an emerging manufacturing technology.
“By using additive manufacturing, we will develop methodologies that will help us better support novice engineers and designers when the next breakthrough manufacturing technology arrives,” he said.
Through the combination of additive manufacturing, machine learning, and explainable artificial intelligence, McComb and his team will use data collected from existing 3D printing design challenges to investigate the use of design feedback. automated. Part designs will be gathered from engineering course design challenges and online open sources.
To test the effect of dataset size on comment accuracy and level of detail, the team will create a machine learning pipeline that will pull models from the design datasets.
As the last stage of the project, studies with engineering students will be carried out in order to provide them with qualifying training and to collect data. McComb defines students as key stakeholders and said that by involving them in research, the team helps prepare them for work in the manufacturing industry.
“Ultimately our goal is to serve them better [engineering students] helping them to design new manufacturing technologies faster and more efficiently, ”he said. “By participating in one of our studies, we want them to learn about additive manufacturing and also recognize that they help us better support other learners in the future. “
The research results will lead to a dataset of mechanical part designs stored as voxels, the 3D equivalent of pixels; a better understanding of the impact of the quantity and quality of a dataset on a machine’s learning and feedback capabilities; and first-hand experience of the impact of real-time feedback on additive manufacturing on solutions created by technical designers.
“For companies, this work will help them understand what kinds of information they can expect to extract from the design and engineering data they already have,” said McComb. “For students and novices alike, this work will provide fundamental approaches to help them learn to use new manufacturing technologies. “