With the increased autonomy and connectivity that will arrive with NDE 4.0, new skills such as programming and problem-solving will likely replace traditional inspection and monitoring tasks. However, although the change is significant, so are the potential benefits, with major opportunities for improvements in the efficiency and effectiveness of inspection activities across the value chain.
NDE stands for non-destructive evaluation, and NDE 4.0 is a subset of Industry 4.0. Similarly, therefore, NDE 4.0 will see inspection tasks take advantage of emerging technologies such as artificial intelligence, machine learning, digital twins, augmented reality, cloud computing and big data. In short, NDE 4.0 is the Industry 4.0-driven transformation of the NDE domain.
GROUP WORKING
Established in summer 2020, the NDE 4.0 Group of the British Institute of Non-Destructive Testing (BINDT) seeks to promote and enable the introduction of new NDE 4.0 technologies by identifying and tackling barriers and, in particular, educating the manufacturing and maintenance supply chains.
“The reason for establishing the NDE 4.0 Group was because, within Industry 4.0, there could be people expecting NDE to facilitate continuous sensing and data analysis – we didn’t want NDE to be the bottleneck,” says Robert Smith, pictured left, chair of BINDT’s NDT technical committee and director of the UK Research Centre for NDE.
“Cost is not a big barrier to adoption in my opinion, as many of the things that require implementing as part of NDE 4.0 are cost-savers,” says Smith. “There are two larger barriers than cost. The first is how to qualify autonomous NDT systems. We can make sure processes, techniques and equipment are fit for purpose and reliable, but it’s a person who decides they are reliable and makes decisions about whether to report a defect and what to do about it. It’s really hard to get systems accepted that automatically sentence components based on results.”
He adds: “The other barrier is skills. The whole NDE industry hinges on people making decisions. So what happens to those people? Some have concerns that automation will result in the loss of skills to industry. After all, how can you program a system if no one knows what a defect looks like? Research is currently in progress on this topic, mainly in Germany.”
Traditionally, when the design of a part is complete, someone takes a decision about what size of defect is acceptable. NDE 4.0 presents a new opportunity.
“What NDE 4.0 potentially allows is ongoing adaptive decision-making about the structural integrity of the component, plant or machine in its current state, thus facilitating better asset management,” explains Smith. “Sensors report the current state to a digital twin, which in turn pumps information into a model that simulates the situation and calculates whether it’s still fit for purpose, and possibly how long until the next maintenance is required.”
Service life extension effectively becomes a daily activity with NDE 4.0, reflecting actual loads incurred rather than hypothetical loads and design limits. This capability also benefits manufacturing. Instead of having a hard limit on maximum acceptable defect size, it is possible to rate the nominally-defective component for a shorter lifespan or fewer cycles.
“In many circles, the NDE community is seen as just another cost factor and a ‘bringer of bad news’, but actually the NDE 4.0 mindset and transition provides a route for the community to add value for other stakeholders,” explains Dr Nick Brierley, chair of BINDT’s NDE 4.0 Group and head of R&D at Diondo (pictured above), a specialist in industrial CT scanners. “Providing detailed information on the quality of parts currently in production means you can improve. It becomes a valuable source of useful information.”
Sectors with strong NDE 4.0 uptake include oil and gas, where pipeline monitoring is an established practice. The industry is enthusiastic about the NDE 4.0 framework, readily putting systems online and adopting AI to process signals. In contrast, a sector such as aerospace has been slower-moving due to its heavy regulation.
“In general, moving to a 4.0 working paradigm is tougher in a manufacturing setting than an asset operations setting, but both present significant opportunities, as well as some costs,” states Brierley. “NDE 4.0 is not a single well-defined destination that you reach by using technology X in combination with process Y, making it difficult to set out the exact costs involved. However, there are some generic costs involving data management, which also means standardising how data is stored and exchanged. So, even if you have a highly digital process [as opposed to digitalised], then getting sensor data from a range of different instruments into a database ready for consistent, systematic processing is an obstacle and likely cost point.”
Of course, there are such things as ‘little steps’ into the world of NDE 4.0, such as equipping field operators with AR goggles so they can automatically see where they have actually scanned a component, while consulting standards and colleagues in real time. It is an exciting time for NDE, with many new and evolving roles, such as a ‘user experience designer’, which Brierley describes as a first for the sector.
“If algorithms become a core component of the NDE process, there will always need to be a human involved at some level, even if it’s just to look at the system occasionally and make sure it’s ticking along,” he says. “This is a human interface that needs designing in a way that ensures a reliable process.”
SAFETY-CRITICAL
“NDE is closely related to safety-critical functions, so it’s only the industries that deal with high-integrity products that worry about it,” states Professor Pete Loftus, pictured below, deputy director at RCNDE with responsibility for NDE 4.0 strategy. “The latitude you can give to a human performing an inspection task is orders of magnitude greater than you can give to a computer. Look at automated cars: one crash makes headlines worldwide, whereas crashes happen thousands of times a day with humans at the wheel. So we have a journey to automation beset with legislative hurdles that need resolving.”
NDE 4.0 is the future, which RCNDE is trying to explore in terms of research, and BINDT is trying to explore in broader ways, including skills and standards.
“I’ve spent the last year trying to figure out why more NDE 4.0 isn’t happening already,” says Loftus. “One reason could be that NDE personnel in large companies aren’t necessarily that well-connected to those looking at data streaming back from continuous monitoring. Some of the battle is therefore helping companies achieve more joined-up working; helping them appreciate the switch from a traditional NDT regime to a system health/condition monitoring regime.”
The NDE 4.0 process sets out many objectives which, if achieved, will double up as potential benefits. These include: improved through-life asset performance; more efficient production; better, quicker and more cost-effective NDE; efficient quality control for bespoke products; reduced safety risk for personnel; and efficient quality control for decentralised manufacturing. Let the transition begin.