This Engineering Emerging areas of Advanced Manufacturing (ENG-EAM) award supports research that will focus on establishing systemic and robust resilience to cyberphysical attacks on connected digital manufacturing systems. Digitization and connectivity are the cornerstones of modern manufacturing, but these very qualities allow cyberphysical attacks to negatively impact part performance by stealthily altering the digital representations of geometry, process plans, and/or in-situ sensing signals. This has the potential to pose a significant threat to societal well-being, economic stability, and national security by introducing defective parts into electronics, spacecraft, planes, automobiles, biomedical devices, and energy components. The state-of-the-art practice of dealing with such attacks by sacrificing productivity, yield, cost, and connectivity to ensure part performance critically limits pervasive and trustworthy adoption of Industry 4.0 and digital manufacturing. This research project will create and validate a novel computational paradigm called Smart-Recover that actively assures every part’s performance despite cyberphysical attacks and with minimal loss in productivity, yield, connectivity, or cost-effectiveness. The research will be complemented by developing a multi-institutional manufacturing cybersecurity education program for workforce development across high school, undergraduate, and graduate educational levels.<br/><br/>The specific goal of the research is to establish the mathematical basis for the Smart-Recover paradigm, which combines pre-fabrication correction of attack-altered geometric models with stoppage-free in-process mitigation of defects created by attack-modified process plans and attack-distorted in-situ sensing signals. To this end, the research objectives include the creation of techniques for: (1) pre-fabrication computational reconstruction of only the attack-altered features of the digital geometric model; (2) in-process remodification of process plans to disrupt formation of local defects induced by atypical attack-driven alteration of exogenous process parameters; and (3) in-process restoration of defect prediction accuracy for attack-altered sensor signals at speeds necessary for local defect mitigation. The research team will further explore the generalizability and collective interaction of these elements of Smart-Recover with stealthy and system-spanning cyberphysical attacks via two manufacturing testbeds. These advances will be achieved via innovations at the convergence of geometric design, machine learning, in-situ sensing, and physics-based modeling.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.