SPC and the Smart Factory, Part One
Whether you call it Industry 4.0, the smart factory, or Industrial Internet of Things (IIoT), technology and the need for increased efficiency and reduced waste are pushing manufacturing toward a revolution in how we produce things. Quality sat down with Jason Chester, Director of Global Channel Programs for InfinityQS, to discuss the role of SPC and other important information for those aspiring to a smart factory operation. This is part one of the three-part series of our conversation.
Quality: Where does SPC fit into the key characteristics of the smart factory or Industry 4.0, i.e. integration across entire enterprises/supply chains?
Jason Chester: The smart factory, or Industry 4.0, is the application of modern connected digital technologies across the entire factory, enterprise, or supply chain to optimize those operations across three core dimensions:
- Minimizing cost—such as operating costs, efficiency, productivity, waste, etc.
- Maximizing value—such as agility, flexibility, responsiveness, product quality, and customer satisfaction, etc.
- Mitigating risk—such as supply and demand volatility, production risks (e.g., equipment failures), and, of course, product quality risks.
Industry 4.0 is a concept that can dramatically improve the performance of manufacturing organizations across these three dimensions. However, it is a concept—not a solution. There is no singular off-the-shelf Industry 4.0 solution that can be implemented. Instead, it is the integration and application of a wide-ranging portfolio of technologies—from low-cost smart sensors, ubiquitous wired and wireless communication networks, big data storage and compute, edge computing, analytics (predictive, prescriptive, and visual), and artificial intelligence and machine learning amongst, many others. Each component provides a critical piece of the jigsaw that is the smart factory, or Industry 4.0. One critical piece, in our view, is SPC.
While many who are familiar with SPC often associate it with traditional control charts, these are just the common visual representations of SPC. Behind those are a suite of powerful and proven techniques and algorithms that analyze, monitor, and predict the performance of a particular process or product characteristic. It effectively identifies abnormal trends in real time and provides alarms and notifications before a non-conformance (e.g., specification limit, net content control, etc.) occurs. When used across multiple processes, facilities, and even entire supply chains, SPC provides manufacturers with a standardized, real-time, and (importantly) evidence-based view of performance across their entire operations.
Yet SPC cannot operate in a vacuum. SPC requires data from operators, machines, Internet of Things (IoT) sensors, or programmable logic controllers (PLCs), for example, and the ability to present analysis and notifications to the right people, at the right time, and in the right place in order to make effective, immediate short-, medium-, or long-term decisions.
Increasingly, modern SPC solutions are leveraging the power of cloud-based deployment models, which provide more scalable and flexible platforms, enabling SPC to be cost-effectively deployed across entire enterprises and supply chains.
SPC is therefore one of the many critical components required as part of a smart factory, or Industry 4.0, strategy.
Quality: How important is standardization of interfaces; can effective SPC be achieved without it?
Chester: SPC fundamentally relies on data—whether entered directly by an operator into the SPC application, imported from files, or directly integrated into data sources, such as gauges or PLCs. As long as data is able to reliably enter the SPC platform, then effective SPC can be achieved. However, while integration with non-standardized interfaces will not prevent effective SPC solutions to be deployed, often the cost and complexity of creating and maintaining those interfaces, especially with a large number of non-standardized interfaces, can be much higher. So, standardization of interfaces is important, not only to reduce the cost and complexity, but also to facilitate the rapid deployment of effective SPC solutions across the manufacturing enterprise.
Quality: If Industry 4.0 requires both products and physical systems to be intelligent enough to know what they are, what product they will become and where they need to go, as well as smart enough to communicate among all systems, what are some of the conditions that need to be met in industries where products and equipment are not intelligent enough?
Chester: I do not believe that Industry 4.0 requires products and specific physical systems to be intelligent to any such degree, but rather it is in the systems that are governing those processes or products where the value of intelligence can be truly leveraged. The vast majority of manufactured components are inert physical objects, and the physical systems involved in manufacturing those components most often perform a very specific task. A plastic electronics case, an engine gasket, a tub of ice cream, or the wing spar of an airplane will never be “intelligent,” nor do they need to be. The finished entity of which they become a part may be intelligent, such as smart consumer devices, self-driving automobiles, operator-less cranes, and pilotless airplanes, but that is a different part of the picture. Likewise, a milling machine will require no more intelligence other than being able to perform its specific task to specification. If a deviation from target occurs, it will need the intelligence to readjust the machining process to keep on target. That level of intelligence can be achieved by a local controller, intelligent enough to perform that task successfully, but little else.
But more broadly, the term “intelligence” is an often overused and misapplied concept in Industry 4.0. Just like human intelligence, we rely on our senses (sight, sound, smell, touch, and taste) to gather information from the world around us. When this information is combined with past experience, we form knowledge about the world in which we live. Intelligence is the ability to apply that knowledge to drive a desired outcome. Likewise, in an Industry 4.0 context, we use sensors and other such technologies (like RFID) to “sense” the industrial environment. When that information is collected, integrated, aggregated, and analyzed across the whole supply chain, we can start to turn that information into knowledge about processes and products, and then apply intelligence to respond—either automatically or via human collaboration—whether that is instructing a specific piece of equipment to respond to a certain variance or given condition elsewhere in the supply chain, or knowing where a raw material is in the supply chain or where a consumer product is located in a warehouse or retail store. This intelligent system may be a centralized platform, a number of interconnected systems across the supply chain, or at a specific location (such as through edge computing).
Applying these interconnected intelligent systems across the supply chain utilizing a myriad of feedback and control mechanisms is, in my view, the reality of Industry 4.0 and the smart factory. To use the human analogy one last time—they become the “brain” of the manufacturing supply chain, able to control and orchestrate the manufacturing and supply chain much more effectively than human, manual processes alone could ever achieve.
For more information, visit www.infinityqs.com.