Industry 4.0 comes out of its Shell
29 July 2021
Industry 4.0 is revolutionising manufacturing. Remarkable changes are taking place – often behind the scenes – to turn expectations into reality. Steve Sands, Head of Product Management at Festo, explains why the unassumingly named Asset Administration Shell is so crucial to delivering the Industry 4.0 vision.
Digitalisation and data will have an incredible impact on today's engineers. Industry 4.0 provides a roadmap by which all parties involved in business processes within the manufacturing industry will be seamlessly connected. Information from suppliers and customers is transparent and readily available. Components, machines and systems are intelligent and able to communicate autonomously. This opens up a whole world of new opportunities for improved efficiency, cost-saving and collaboration. The big question is, how do we get there?
The name "Asset Administration Shell" (AAS) may not slide easily off the tongue, but it describes an essential tool in realising the Industry 4.0 vision. The AAS is the 'digital shell' that wraps around each smart element of a machine, enabling it to explain to the complete machine architecture its function, identity and a whole host of additional information.
In other words, the AAS enables assets to be integrated into the world of information. These assets can be machines and their components, supply materials, technical documents (such as wiring diagrams), business documents (such as contracts and purchasing information) and any other information that will help machines and humans to collaborate effectively using common standards, protocols and languages.
While the cloud, networking and automation devices with internet access can deliver a form of connectivity, the AAS is truly universal, which opens up exciting new possibilities. It enables information to be exchanged between value chain partners, and uses neutral and common standards for communication, services and semantics across companies and across sectors.
Evolving picture
The digitalisation process encompassed by the AAS is evolving extremely quickly. Instead of creating brand new standards that can take many years and endless wrangling between vested interests, the Industry 4.0 vision has galvanised a unifying approach. Where existing open standards do not meet the need, they are modified or extended. The vision and the benefits are so compelling that many potentially rival organisations have worked together to create a working, albeit still developing, solution.
Festo is using its experience in the field of mechanical engineering and automation to support the development of the AAS. Increasingly we are developing smart digital solutions combining extensive know-how and experience with the latest information technology developments. The outcome is that Festo is increasingly utilising hardware plus firmware and software in industrial automation applications.
Much has already been achieved through this multi-organisational collaboration. Examples of incorporated standards include OPC-UA for the communication of data. Automation ML is evolving to be a global standard for transferring the description of a physical item containing the information usually found in 2D or 3D CAD drawings and complete electrical circuit diagrams, programming information, kinematics and behavioural description.
Practical application
So, what is so important about the AAS? This is best described by an example journey where a designer is looking for a particular component: let’s say an electro-mechanical axis that can move a part of their machine with a defined mass, at the required velocity and acceleration.
Picking the component from an easy to use, online selection and configuration package, the designer needs to match the axis to a motor and gearbox combination to deliver the required torque and performance characteristics. Next, the matching servo drive is selected. At each stage of the selection process, the digital description of the system is built up, one brick at a time: for example, describing the parts' properties and capabilities. As each selected part is configured in the virtual world, its digital descriptor is created in parallel, providing its Identification (what it is), Technical data (how it works) and Documentation.
As the design process progresses from a single asset to a sub-system, the digital descriptor grows and integrates the sum of all the parts. The virtual model is accessed throughout the machine lifecycle e.g. for CAD design, CAE simulation, virtual programming, documentation etc., spanning from the machine inception through to build, operation, maintenance and even re-purposing. It upscales in a modular way as the solution grows. The AAS includes a further digital model that captures the Operation data. Under each of these headline modules are submodels that semantically enable the transfer of data – independent of the supplier or manufacturer. Common standards allow the seamless transfer of machine-readable data.
Digital Twins
The AAS contains a complete description of the properties and capabilities of the system. In other words, the AAS is the means by which Digital Twins become possible. Having a Digital Twin means a machine can be built, programmed, commissioned and operated from virtually anywhere in the world. Eliminating costly and time-wasting mistakes is only one of the benefits that engineers can look forward to. Currently, once a machine is built and shipped to the customer, it is extremely expensive (if not impossible under today's restrictions) to conduct maintenance or investigate a problem. A Digital Twin can replicate its physical counterpart, enabling corrective actions to be tested without the costs or delays of long-distance support visits.
Predictive maintenance
The AAS is also critical to optimising predictive maintenance – the ability to anticipate a component's failure and plan a maintenance intervention, thereby preventing unplanned stoppages. The AAS is an excellent boost to artificial intelligence (machine learning) algorithms' efficacy. Artificial Intelligence is one of the Industry 4.0 technologies recognised to have the highest impact on engineers in the near future. An Edge or Cloud-based processor running a combination of algorithms is tuned to the application (such as the Festo AX Scraitec package), monitoring vast data tranches. The system observes and identifies patterns within the data lake. During the set-up period, a human-in-the-loop trains the system by categorising the anomalies it detects. The human operator teaches the system; rewarding and labelling predictions of failures and 'punishing' unhelpful observations. In this way, the system is trained and improves its useful predictions whilst tuning out unhelpful noise.
Predictive maintenance isn't just a vision, it is already being applied. Festo has several use cases: one of which comes from the automotive industry, where a body-in-white assembly line is equipped with hundreds of pneumatic clamps. Whilst these are relatively low-cost products, if one fails, the losses can cost £100,000s in lost production: or, even worse, undetected, poor-quality welds. The ability to foresee pneumatic clamp failure, identify the individual clamp and categorise the failure (solenoid valve, pneumatic seals or clamp pivot bearings) enables planned maintenance interventions up to 14 days before a breakdown situation.
Too much information?
The Industrie 4.0 workgroup developing the Asset Administration Shell has done a fantastic job within a short space of time. The big question is: can an engineer with this extra information improve or accelerate their work? Potentially there is a balance to be had between the enormous benefits to be gained through using smarter products coupled to immense processing power and investing in more complex and expensive systems at the outset. End-users rightfully need to see the evidence of improved lifetime efficiency and payback before specifying these technologies to their line and machine builders.
Quick wins, though, are already at hand. For example, Festo offers a simple to use, free of charge software tool called Quick Search plus, a powerful extension to our digital catalogue. Here you can find very quickly all the information that you require for the whole product range. Eliminating the manual transfer and re-entering of data eliminates wasted time and the possibility of errors. All product data within the bill of materials is captured and seamlessly transferable. A designer can then pick a part from the BoM, such as a pneumatic cylinder, and immediately use it within the Festo 3D CAD design tool. The correctly matched accessories such as piston rod and rear mounting clevis are added to the drawing in seconds. Identifying perhaps six additional components in the past could have taken even a familiar designer many minutes, but here, the task is completed, right-first-time in seconds. The complete assembly can be offered up in the machine general assembly drawing and simply dropped into place.
Conclusion
The importance of the Asset Administration Shell in the implementation of Industry 4.0 cannot be underestimated. Hidden behind this technically descriptive (but possibly rather dull?) term are genuinely exciting engineering changes. The AAS is a crucial enabler in speeding up the design of complex machines and enhancing their whole life operating effectiveness.
Understanding AAS
Festo recently hosted an insightful webinar on the Asset Administration Shell within the concept of Industry 4.0 presented by Dr Michael Hoffmeister of Festo, a visionary and expert in this field. As convener of the IEC workgroup on the AAS he is deeply involved in developing the vision for Industry 4.0 and how critical information about each part of the overall picture can be connected and seamlessly pass on information. He is on the board of the recently formed International Digital Twin Association.
To watch the recording visit: www.festo.co.uk/webinars
Contact Details and Archive...