Innovation in smart manufacturing can come in a flash from a lightbulb moment, but those instances are few. More often than not, breakthroughs in technology, such as bioprinting, blockchain, cloud-based manufacturing and real-time production control, happen after years of careful study accompanied by painstaking, methodical work done sometimes in academic settings. That careful, plodding work is led by the people profiled in the following pages. All have earned doctorate degrees. Many of them count among their significant achievements not only their personal work but also the legions of students they have trained to be the next generations of innovators. In the end, all are people who ask not only “Why?” but also “Why not?”
Dr. Laine Mears
Mears wants to be a smart manufacturing matchmaker. “There is so much to do in the area of how smart manufacturing solutions can make their way onto the manufacturing floor and a lot of technology is emerging,” he said. “I would like to design a process where promising technologies can find the right customers in a controlled and scalable manner, and not get lost in the sea of potential solutions.” Mears founded the THINKER (Technology-Human INtegrated Knowledge, Education and Research) program at Clemson with a five-year National Science Foundation grant to educate students in how humans can best be integrated into the digital manufacturing enterprise. “This goes beyond traditional human-machine interface designs, seeking instead to understand how humans generate and use information and the best ways to convert combined data from people and machines into the most effective information,” he said. Want to follow in his footsteps? Mears advised building a wide-ranging network. “I have found manufacturing researchers (both industrial and academic) to be quite a collaborative group, so the larger one’s network, the greater the opportunities.”
Dr. Satish Bukkapatnam
Bukkapatnam and his team of “renaissance engineer” students used Python to create an open-source CAD/CAM interface to generate G codes for hybrid 3D printing/milling with metals. “Metal-based additive manufacturing processes are still lacking in … open-source software and a supportive community (like the one for FDM printing),” they wrote on the university’s website. They successfully integrated their open-source software with the university’s Optomec hybrid machine and also demonstrated various off-the-shelf software and hardware modules to collect, manage and analyze large data streams from the process quickly enough to allow faults to be detected early for quality assurance. “My students develop an excellent understanding of manufacturing technologies, as well as of the latest measurement and data analysis approaches. They gain hands-on experience in advanced manufacturing platforms, undergo training in advanced mathematical and data science approaches to deal with the data and the complex challenges in smart manufacturing systems.” His own research is on the harnessing of high-resolution, nonlinear dynamic information, particularly from wireless MEMS sensors, to improve the monitoring and prognostics of manufacturing processes and systems of interest to the industry.
Dr. Jian Cao
Cao’s lab has developed a completely die-less forming system called double-sided incremental forming. The system can form a 3D sheet-metal part without using a geometry-specific die set as is currently practiced. “Therefore, we can reduce the design-to-part cycle time from up to 12 weeks to under one week, and eliminate the need to make dies,” she said. “The challenges in DSIF include geometric accuracy and formability prediction, for which we have developed an in-situ compensation method using in-machine sensors and off-line, mechanics-based computational models.” Currently, she is working on the concept of a manufacturing process compiler that integrates knowledge from multiple domains on a single platform so that one can identify what manufacturing processes are most suitable for a given design. “Ultimately, one can use this compiler as the basis for new process innovation,” she said. “This is more easily said than done and will require some long-term work.” She noted that manufacturing education has implications beyond the STEM subjects. “Therefore, my advice is to have a system view, be broad and then find your own specialty and collaborate.”
Dr. Adam W. Feinberg
In the next decade, Feinberg wants to help translate 3D bioprinted scaffolds and tissues from the bench to the bedside. In five years, he wants to demonstrate small-scale functional organs like a multi-chambered heart that pumps blood and is viable for more than 90 days. He may get help from people who attend his lab’s workshops on how to build the open-sourced 3D bioprinters he uses. In the meantime, Feinberg’s lab developed new 3D printing technology for liquid and soft materials. Known as FreeForm Reversible Embedding of Suspended Hydrogels, or FRESH, they print materials inside a support gel. “This is like having support material everywhere, and it allows us to 3D print pure liquids, or a liquid polymer that may take time to solidify or gel,” he said. “We first published this method in Science Advances in 2015 and just last year published work in Science on 3D bioprinting collagen to rebuild the human heart.” For anyone wanting to follow in his footsteps, he advised, “seek out like-minded people and identify the universities where manufacturing-related education and research are thriving.”
Dr. Ajay P. Malshe
In the industrialized and industrializing world, every human encounters at least 10 machines on a daily basis, Malshe has observed. Manufacturing, operational and maintenance challenges, including friction, wear, machining and corrosion—what he calls “cancers of mechanical machines”—seriously challenge their performance, resulting in billions of dollars in losses. These cancers occur at the nanoscale and, therefore, nanomanufacturing is the only smart solution for curing them, Malshe said. “Nanomanufacturing innovations from me and my team have contributed to solving those manufacturing, operational and maintenance challenges with potential impact worldwide,” he said. Malshe envisions Industry 5.0, which will be human- and earth-centric and will be true smart manufacturing by and for humans in harmony with the planet. “We are experiencing unprecedented growth in the world population, in the size of the middle class and overall life expectancy,” he said. “As a civilization, we need more and more good jobs for democratic survival on the planet.” The definition of “smart manufacturing” needs to be revisited because, he said, we’ll soon reach the theoretical limit of available natural resources on earth.
Dr. Denis Cormier
Much of Cormier’s 25 years in additive manufacturing has been focused on the design and fabrication of engineered lattice structures that are now widely used in things like light-weight aerospace structures, bone implant surfaces, filters and heat exchangers. “It all started in the late 1990s when I saw a piece of copper foam on a colleague’s desk,” Cormier said. “He explained how properties such as surface area, porosity and tortuosity of the cell structure were critical to its functional performance. That was in the early days of 3D printing, and I instantly started thinking about 3D arrays of Bucky balls or other geometric building blocks that would allow designers to optimize cell structure relative to the performance requirements of a given application.” Cormier eventually started calling these things engineered cellular materials. That was the beginning of a career-long research thrust. “If you go to an additive manufacturing trade show today, virtually every booth will display examples of engineered cellular materials,” he said. “It’s gratifying to see that and to know that I was among the pioneers in that field.”
Dr. Glenn Daehn
Daehn said he’s been able to work with some really smart people to develop two visions: one is impulse manufacturing, using explosive-like force in factory or lab environments. The other is in metamorphic manufacturing, aka robotic blacksmithing, or using numerically controlled deformation to make parts. “We are hoping to see impulse and metamorphic manufacturing developed as well-used commercial processes,” he said. “Both hold great promise for solving real problems in forming and joining advanced materials and structures.” Daehn thinks advanced control and artificial intelligence hold great promise for making many niche manufacturing technologies mainstream, reproducible and agile. “Imagine a robot system that can do what a skilled craftsman can do, but more reproducibly and with the steps clearly recorded,” he mused. In terms of academia working with industry, he said there is too much divergence between the two sectors. “We in academia have to focus more on integration, engineering and real problem solving and developing talent who have more of a bias for doing, rather than analysis,” he said. “This may lead to more labs that are shared between industry and academics.”
Dr. Jun Ni
Ni is like the mouth of a smart manufacturing river, with close to 100 doctoral and 70 master’s graduates, and hundreds of engineering undergraduates forming the tributaries that fan out to faculty positions at other universities and executive positions at global companies. “I am also very proud of my accomplishment as a founding dean to establish the University of Michigan-Shanghai Jiao Tong University Joint Institute,” he said. “Thousands of American and Chinese students have benefited from this global innovation in engineering education.” His international efforts don’t stop there. From 2017-2019, Ni was co-chairman for the Global Future Council on Advanced Manufacturing and Production at the World Economic Forum. After 40 years in academia, Ni wants to help smart manufacturing entrepreneurs, so three years ago he helped launch a startup. His vision for future smart manufacturing systems includes: 4 “R”s (responsiveness, resilience, reconfigurability and reusability) in addition to traditional qualities; self-aware, self-adaptive machines that can evaluate their condition and make necessary compensations; zero defects with predictive automatic root cause identification; near-zero downtime, and every part right, from first to last.
Dr. Xun W. Xu
To facilitate his research on Industry 4.0, Xu established New Zealand’s first and only laboratory for smart manufacturing, the Laboratory for Industry 4.0 Smart Manufacturing Systems. The lab is a training ground for students, demonstrates how businesses can benefit from smart manufacturing and facilitates collaboration between researchers and industry. Earlier in his career, Xu made original contributions toward developing next-generation computer numerical control machining systems based on a new CNC standard, STEP-NC, that enabled smart machining processes. In 2012, he published a seminal article on cloud-based manufacturing, a new manufacturing paradigm at the time. “Cloud manufacturing extends the concept of cloud computing to manufacturing so that manufacturing capabilities and resources are componentized, optimized and provisioned as services,” he said. His vision for smart manufacturing’s future includes humans who are empowered by smart and autonomous tools. “Manufacturing systems will also continue along the path of being ‘flattened’ in that any boundaries between manufacturers, what is to be manufactured and whom it is to be manufactured for become blurrier and blurrier,” Xu said.
Dr. Binil Starly
Starly’s most significant career achievement so far was making an LED light glow. That light meant his group made a physical manufacturing machine communicate with a global blockchain via a digital twin. The machine was able to autonomously initiate transactions on a smart contract stored on the blockchain and, as a result, triggered the LED on another networked physical machine. “This moment demonstrated the enormous potential for the application of blockchain technologies toward reducing the gap between manufacturing service companies and their potential clients, improving transparency and trust,” he said. “It would also mean that entire machines can now be connected on a global decentralized network of manufacturing resource nodes enabling the realization of cyber manufacturing.” He envisions smart manufacturing making progress in three interlinked areas that span the product lifecycle. First are intelligent interfaces that jointly collaborate with humans. Second are decentralized manufacturing service marketplaces. Third, he envisions manufacturers responding to user preferences by digitally connecting their machines from the shop-floor level to the business/IT systems, integrating humans, processes and technology.
Dr. Lihui Wang
In 1998, Wang was already working on web/internet-based, model-driven, real-time monitoring and control of machines and robots. In 2008, his work included human-robot collaboration. Combined, the monitoring/control and human-robot collaboration formed the basis of the digital twin and cyber-physical systems. His research team is actively working on big data analytics and AI applied to predictive maintenance, machining process planning and scheduling, and human-robot collaborative assembly. “The combined use of big data and AI algorithms can foster the full potential of varying decision processes with real-time manufacturing intelligence,” he said. “This moves manufacturing toward better productivity, efficiency, profitability and long-lasting sustainability.” His vision for the future is one driven by data, AI models, knowledge and human skills, empowered by cloud/fog computing in cyberspace but with humans at the center. “On one hand, AI and AR will provide on-demand decision support to human operators,” he said. “On the other hand, human perceptive and adaptive ability will be used to drive manufacturing equipment in the form of brainwaves to replace rigid control codes.”
Dr. Thorsten Wuest
Wuest and the University of South Carolina’s Dr. Ramy Harik last year wrote “Introduction to Advanced Manufacturing,” a textbook targeted at filling the gap in manufacturing education for engineering students. It has a chapter dedicated to smart manufacturing, which is a “first of its kind to my knowledge,” Wuest said. In 2018, he started representing the U.S. and smart manufacturing on the editorial board of the World Manufacturing Forum Report. He’s passionate about: emphasizing the human element in smart manufacturing systems; bridging the gap among experts’ knowledge; physics-based modeling and data-driven methods in a hybrid approach, and supporting a collaborative approach to help smaller enterprises develop a smart manufacturing roadmap. Wuest would like to see academia modernize the manufacturing content it teaches. Academia would also do well to embrace inter-disciplinary, cross-program classes and project-based learning for engineering students, he said. And colleges would be wise to work with high schools to change incoming students’ perception of manufacturing from “dark, dirty, and dangerous” to reflect today’s reality of providing well-paying and fulfilling high-tech careers that help society at large.
Dr. Qing “Cindy” Chang
By introducing novel concepts and methods like opportunity window, direct virtual data modeling and permanent production loss, Chang is a pioneer in data-driven modeling and real-time production control and decision-making to improve the efficiency of manufacturing systems. “Several aspects of my research have been realized and validated in physical form, which I am particularly proud of,” she said. Chang has developed and implemented a real-time, data-driven decision support system to optimize production operations in dynamic and stochastic operational conditions. Her work has been implemented in many General Motors plants in North America and demonstrated significant improvement in operational efficiency and economic benefit. If adopted, it will potentially enable even greater economic benefits for many other industries. She sees recent developments in artificial intelligence and machine learning showing great potential to transform manufacturing through advanced analytics tools for processing vast amounts of manufacturing data. The focus on data-driven manufacturing requires future engineers to acquire training in data science, which is an enabling skill in the smart manufacturing field, Chang said.
Dr. Tony L. Schmitz
Schmitz sees the symbiotic workings between academia and industry as key to the future of smart manufacturing. “In academia, we are at the front lines of training the next generation manufacturing engineer,” he said. “It is important for academia to understand industry needs so that education addresses those needs. Similarly, it is important for industry to be engaged with academia so new ideas and technology can be implemented successfully.” In turn, Schmitz sees smart manufacturing as a catalyst for the growth of American industry. “I see smart manufacturing as the basis for increasing the U.S. manufacturing base, including machining. As we are able to make better decisions about manufacturing processes, we will increase our competitiveness in the global market.” Schmitz’s career focus has been on developing predictive models for manufacturing processes, including an approach for predicting tool point dynamics in milling and milling process simulations. He believes there are great opportunities for combining physics-based manufacturing process models with machine learning algorithms to enable autonomous operation. This will guide his future smart manufacturing research.
Dr. Ramy Harik
Harik established the Future Factories laboratory at the University of South Carolina and would like to see academia contribute to smart manufacturing’s future by creating similar labs to form a network. “The network would integrate fundamental concepts from cyber manufacturing, automation and advanced manufacturing to form an ecosystem where future students will be exploring and working with these concepts prior to graduation,” he said. Harik’s lab is a unique experimental platform that includes a concert of industrial equipment: robots, drones, real-time cameras, conveyors, smart glasses and augmented-reality devices. The platform is digitized with an active digital twin, and a digital engine processing all the incoming data and running potential conflict/failure scenarios. He wants to continue innovating on the platform while basing an online curriculum for smart manufacturing on it. “The Future Factories platform will be an active test bed for the online course,” he said. “I want to make this curriculum as widely available as possible to entice the future workforce on the topic of smart manufacturing and manufacturing jobs that are passionately intriguing.”
Dr. Jay Lee
Lee began his academic career in 2000 after many years of diversified experience in government and manufacturing. He was the founding director of a National Science Foundation Industry/University Cooperative Research Center on Intelligent Maintenance Systems (IMS), which has been a catalyst for the transformation of industrial big data, machinery prognostics and industrial AI. The IMS Center has worked with more than 100 global companies to develop and deploy smart manufacturing to achieve zero downtime (ZDT) and worry-free performance. Since last year, a number of IMS company members, including Foxconn, have received the Lighthouse Factory recognition by the World Economic Forum with the involvement of IMS. “My current work is to develop a systematic, industrial AI system to realize ZDT and worry-free manufacturing in industry,” he said. Lee’s vision for the factory of the future includes not only smart machines and operations but also the transformation of data to predictive decision and new knowledge. “With the advent of the industrial Internet, 5G and industrial AI, we can discover and develop many new opportunities for, and exciting solutions for, people who are interested in smart manufacturing,” Lee said.
Dr. Satyandra K. Gupta
Gupta’s group develops smart robotic assistants to increase human productivity in manufacturing applications. These smart assistants are capable of programming themselves from task descriptions, learning from the observed performance, safely operating in the presence of uncertainty, appropriately calling for help during the execution of challenging tasks, and interacting with humans in a user-friendly manner. The group last year posted a YouTube video showing a group of next-generation robots automating a composite sheet layup process. “The robotic cell is smart, and it can adapt to uncertainties during the layup process,” Gupta said. “The cell uses artificial intelligence-based algorithms combined with force and vision sensors to automate the fabrication process. The system uses advanced computer vision to detect defects and, if needed, calls a human for assistance.” He is also interested in developing smart manufacturing tech that can reduce human error by monitoring human performance and assisting humans appropriately, as well as exploiting smart manufacturing tech to accelerate the training process. “This will require developing technology solutions that also pay attention to privacy and security considerations,” he said.
Dr. Jim Davis
Davis was there when smart manufacturing (SM) was born. He’s been at it since the 1970s when he worked on digital data and control systems for industry. Since then, he has researched AI in manufacturing, helped start Internet2 and helped found not only the Smart Manufacturing Leadership Coalition but also the Clean Energy Smart Manufacturing Innovation Institute. He now sees AI as the way to align OT with IT to further smart manufacturing. “When I look across my history with SM, industry is only just now ready to rethink manufacturing foundationally,” he said. “I would like to see SM turn this corner toward its full potential.” For those who want to participate, he advised, “SM is a high tech, data-driven profession that is dealing with how to make things the right way for the world. If you are looking to be part of an exceptional diversity in needed perspectives to address a global grand challenge, SM offers technical, practice, education, policy, communication and people-oriented career pathways. We are beyond saying what manufacturing isn’t, i.e. dirty, dumb and dangerous. Smart manufacturing is about ‘making,’ not just manufacturing.”
Dr. Robert X. Gao
Gao developed a systematic approach to the design, modeling, characterization and experimental evaluation of structure-embedded, multi-physics wireless sensors and related machine learning (ML) methods for the condition monitoring, fault diagnosis and remaining service life prognosis of manufacturing machines and product quality control—with applications in plastic injection molding, sheet-metal stamping and electrically assisted micro-rolling. He wants to integrate physics-informed AI algorithms with process-embedded sensing methods to further improve production control and material and energy efficiencies. For smart manufacturing, he envisions more digitization that spans operations and supply chain and enables safe and seamless human-machine collaboration, automated performance optimization, prescriptive maintenance and environmentally responsible production. Academia could contribute to this vision by introducing students to some of the building blocks that make manufacturing “smart,” such as ML, in addition to the fundamental physical sciences. His advice to those who aspire to an academic career in manufacturing is to establish a solid foundation in physical science first while becoming competent with the fundamentals of data science.