University of Buffalo team use AI to bring efficiency to Additive Manufacturing industry
May 30, 2022
A research team led by the University of Buffalo, New York, USA, is working to improve the quality, production and efficiency of Additive Manufacturing and semiconductor manufacturing. The work is funded by a $2.3 million grant from the National Science Foundation.
“Each step may be optimised, but that doesn’t always mean it’s for the greater good of the overall production process,” stated Hongyue Sun, PhD, the grant’s principal investigator. “What we’re doing is creating an analytical framework that connects and coordinates all these processes. The end result will be a cyber-physical system that uses artificial intelligence and other tools to optimise and ultimately improve manufacturing systems.”
The framework, which the team is calling STREAM, includes the use of Artificial Intelligence, simulation and other technologies. It is intended to create a public online repository where researchers and industry professionals can share information and their experiences regarding data, models, simulators, controllers, analytics and empirical studies.
Additionally, the project includes three interconnected research tasks:
- Creating software that enables efficient communication and computing in cyber-manufacturing systems
- Creating a modelling system to achieve an accurate and efficient process quality control
- Creating a simulation and production control system for continuous improvement of quality, manufacturability, and productivity of future multistage and distributed manufacturing systems
In the semiconductor manufacturing process, for example, Sun says there are numerous dependent steps. “This includes tens of stages such as crystal growth, ingot slicing, wafer lapping and polishing, lithography, etching, chemical mechanical planarisation,” Sun said. “These stages have strong dynamics and dependencies. The operations at downstream stages are affected by operations at upstream stages, quality wise and productivity wise – for instance, multiple lapping machines need to collaboratively process hundreds of wafers from ingot slicing; and the real-time process and production information of machines are interdependent and jointly determine the system’s performance.”
As part of the project, researchers intend to create new outreach and workforce development activities for students ranging from primary to graduate students. They also will work with professionals in the field of manufacturing.
These research results, Sun says, may be included in the university curriculum for advanced manufacturing, architecture design, machine learning, simulation, and system control and optimisation. The work is also hoped to support UB’s efforts to advance the greater Buffalo area’s economy.