Sept.9, 2021
When cybersecurity researchers anticipate how cybercriminals might undermine U.S. manufacturing, they come up with scores of doomsday scenarios.
Consider the advanced integrated circuit (IC), known as a semiconductor chip. The IC forms the brains of computing devices in everything from cell phones to space stations. Without the IC, we’d live in an altogether different world – no internet, no Google, no Silicon Valley.
The magic comes from a small wafer, usually made of silicon and imprinted with layers of microscopic masks to produce a pattern of electronic components. The wafer is diced into many pieces, each containing one copy of the circuit, called a die.
Manufacturing an IC may involve more than 1,000 precisely controlled steps. Procuring intellectual property, blueprint design, fabrication, testing, assembly and packaging are largely outsourced to third-party vendors. There are multiple points along this supply chain where things can go wrong.
Imagine a rogue worker sabotaging an otherwise sound chip design by modifying a few masks on it to redirect energy. Once the malicious IC is embedded into a device, energy is redirected to burn out the circuitry, leaving the device (phone, computer, missiles with built-in satellite navigation) incapacitated, leading to potentially disastrous consequences.
If something like this happens, could the culprit be captured, and perhaps more importantly, could the criminal act have been prevented?
Last November, in an endeavor to answer this question and others about vulnerabilities in American manufacturing, the University of Texas at San Antonio launched the national Cybersecurity Manufacturing Innovation Institute (CyManII). The $111 million public-private partnership funded by the U.S. Department of Energy engages in collaborative research and development to help the nation’s manufacturers become more resilient against cyberattacks and ensure U.S. competitiveness in advanced manufacturing.
CALIT2, a CyManII managing partner, will collaborate with 24 other universities to develop tools, technologies and guidance for securing manufacturing supply chains, factory automation and information, and workforce development.
“Our target group is small and medium-size manufacturing companies. Upgrading their efficiency by leveraging data will allow for a surge of demand from internal markets,” says G.P. Li, UC Irvine CALIT2 director. “At the same time, we want to ensure cybersecurity and provide energy efficiencies that support the nation’s goal to combat climate change and its impact.”
According to Li, UCI has a deep commitment to advanced manufacturing as evidenced by the establishment of advanced manufacturing programs at CALIT2. These include:
• EVOKE Lab and Connected Learning Lab
• Data Engineering ThinkTank
• The Smart Connected Worker program
• California Plug Load Research Center
• Integrated Nanosystems Research Facility (INRF), Bio-organic Nanofabrication Facility (BiON) and MEMS fabrication facility
• IoT Cyber Test Range
“These labs and facilities will play key roles as testbeds and development sites for securing clean-energy advanced manufacturing,” Li says.
The Smart Connected Worker program is leading the way by developing a platform that leverages artificial intelligence to transform collected data into metrics, insights and predictions.
This platform creates “digital twins” of human workflows in advanced manufacturing contexts. Digital twins are virtual representations of the real world that incorporate physical objects, processes, relationships and behaviors via augmented and virtual reality. Captured information will help produce solutions to reduce energy consumption and environmental impacts, and produce a more equitable workplace.
UCI’s Cybersecurity Policy & Research Institute (CPRI) also is contributing to the effort. The institute is generating comprehensive technical and policy-driven strategies to address cybersecurity’s technical, legal, policy and human challenges, and building consensus around cybersecurity solutions at the intersection of technology, law and policy.
A key CPRI initiative is the establishment of the IoT Cyber Test Range, which develops protocols for testing IoT systems. With funding from the Taubman Foundation, the IoT Cyber Test Range seeks to detect security vulnerabilities in order to evaluate the robustness of different security approaches implemented in IoT systems and devices.
Then there is V-GER (Virtual/Augmented, Gamified Experiential Reality) Cyber Test Range, a collaboration of the Smart Connected Worker program, CPRI/IoT Cyber Test Range and INRF. V-GER is contributing to CyManII’s goals for work force development and supply chain security tools by producing VR/AR content for gamification of cybersecurity education, exploration and simulation.
Its platform will incorporate real-time operational features through which participants (game designers, game engineers and game players) encounter real-world, real-time security challenges for Internet of Things technologies inherent to advanced smart manufacturing.
“Users can think like an attacker and map out where cyber risk exists,” says Richard Donovan, CALIT2 assistant director of research development/sustainable smart manufacturing. Researchers expect to develop prototype applications that demonstrate workflows associated with semiconductor fabrication. When it comes to preventing intrusions, such as sabotaging IC circuits, a multiscale operations model could be designed that spans tooling design and manufacturing, device making and supply chain operations, Donovan explains.
“These models could connect individual dies with individual masks and mask locations, and further down the road with individual mask suppliers and factories,” he says. Scanners could be equipped to notice when one die looks different from others. Anomalies would be flagged if the mask-writing instructions don’t match the blueprint design. These solutions could detect deviations before their consequences end up on the final device.
While this generation of research is squarely aimed at cyberphysical systems, using data to analyze the implications of machines communicating with each other, Li says future research must encompass human and environmental systems as well. All four elements – cyber-, physical-, human- and environmental systems – including the effects of human interactions on cybersystems and the environment, will be critical to achieving effective, secure and energy-efficient advanced manufacturing. And data will remain every bit as critical to the effort.
“Scientist will need to aggregate data from all four elements and take a holistic approach to find tools and solutions to produce safer, more equitable workplaces and reduce environmental impacts,” Li says. “To do that, data is everything.”
– Sharon Henry
