Consistent inspection processes are the keys to manufacturing high-quality products but also the key to the sustainability and success of a manufacturing company. To make quality assurance more accurate, faster and more efficient, a fully automated process is a possibility. An impact in this regard is the targeted use of artificial intelligence (AI). Fujitsu and VisiConsult, a specialist in industrial X-ray systems, are collaborating in this regard.
Christof Schleidt, project manager at Fujitsu, and Lennart Schulenburg, Managing Director of VisiConsult explain the objectives of this partnership and discuss the benefits for the customer.
Optimal combination of skills
Christof Schleidt: You are experts and market leader in industrial X-ray systems, we are specialists in computer integration, AI and visual inspection. How can we combine our skills optimally so that the customer and the entire market benefit from it?
Lennart Schulenburg: In my opinion, our two companies are like sparring partners, we perfectly complement our skills, our solutions and our technologies, this translates into sustainable added value. With our automated X-ray systems, we have digitized and automated the non-destructive control process (NDT). With high-resolution digital X-ray detectors, data on the images generated is important. With you, we will now take the next step towards Industry 4.0. With your AI expertise, the next step from standard image processing to automated image evaluation will be possible. In this way, we can together help our customers increase their detection probabilities and productivity.
Christof Schleidt: In addition, through the network via cloud connection and AI use, it is much simpler and closer to the process to get the results of detection. You are in contact with industrial sectors where the highest quality standards are implemented. I am thinking of critical components like suspension in the automotive sector or turbine blades in the aerospace sector. In these areas, there is no room for error in terms of accuracy and detection rates of algorithms. We appreciate your years of experience in non-destructive testing and your focus on the industry.
Successful use cases
Lennart Schulenburg: Your offering complements intelligent technologies for visual inspection and fundamental market knowledge with in-depth AI expertise. For example, we have already launched two pilot projects in the welding inspection field, with a welded pipeline producer and a manufacturer of thin-walled titanium welded tubes for the aerospace sector. Both are essential to safety in their areas of use. Water pipes in an aircraft as well as pipelines must be watertight, to ensure safety when used and to avoid any safety problems.
Christof Schleidt: And with these successful use cases, we show that our approach is not just a bland theory. Instead, we bring tangible and practical use cases along the way. Our main goal is the seamless integration of AI into non-destructive testing applications. In order for our customers to be supported and their needs met, have you developed a deployment strategy.
Lennart Schulenburg: We follow a system that contains four steps: select-prove-qualifying-run. First, we offer clients, by actively participating in this program, to be a pilot client. In the "prove" stage, we develop in close communication with the client proof of the concept, we examine the technical probability of the project. Then we move on to the "qualify" stage. Because we operate in a security-critical environment, we need to qualify algorithms carefully. We must demonstrate the accuracy of the results against the high criteria of punishment. Finally, we have the "run" stage where the system switches to live operation.
Benefits for the customer
Christof Schleidt: I think through this process, the client has an idea of what they can expect from this technology and can decide whether they want to invest more or not.
Lennart Schulenburg: Yes precisely. And we show that it is not only technically feasible but also commercially viable. The manufacturer saves money and has a cost-effective and safe technology. The long-term goal is a universal and continuous automated evaluation system.
Christof Schleidt: In addition, customers enjoy many other tangible benefits. Our approach to the subject with our cooperation simplifies the process which would be a huge energy debauchery for a single company. Managers should speak with many different contacts, such as AI specialists and non-destructive material testing experts. In the real world, this would mean a high investment that would not be profitable for most companies. We take this quest on our own and offer an ingenious and easy-to-achieve solution. Another notable advantage is accurate and rapid recognition of defects that was not possible in manual or rules-based evaluation systems. In addition, this will reduce costs, as AI-based quality inspection minimizes significant errors in safety-critical products.
Lennart Schulenburg: Beyond that, we can automate the inspection process and thus significantly increase the productivity of the high-value chain. In addition, parallel communication with regulators simplifies the process for the customer. Because when operationally deploying AI technology, there are several certification guidelines to follow. This allows the customer to benefit from our expertise in both your knowledge of AI and visual inspection and our 26-year experience in the non-destructive testing and market leadership of X-ray systems.
Looking to the future
Christof Schleidt: I think in the future we will also combine our skills focused on the benefits to our customers. Fujitsu and VisiConsult are capable of revolutionizing the market. I am thinking of an innovative opportunity to optimize the process and make it more transparent. One objective could be, for example, communication between operational centres such as a casting unit and the real-time evaluation module via X-ray technology.
Lennart Schulenburg: In addition, it would be convenient, when delivering lower quality materials, to give direct and timely feedback to the supply chain. In addition, in the future, in addition to long-term X-ray technology, there will be various other non-destructive control methods such as ultrasound, Foucault current or magnetoscopy evaluation.
Visit the site: