Fujitsu’s mission is to achieve human centric innovation – a concept that has co-creation at its very heart as a fundamental element of our R&D strategy. At Fujitsu Laboratories of Europe, our teams are pioneering a host of state of the art Artificial Intelligence advances, adopting a tangible co-creation strategy and working with customers, collaboration partners and society to develop a new generation of user-centric applications and services, under-pinned by creative information analytics.
At Fujitsu Forum last month in Munich, we announced a major co-creation project with Siemens Gamesa, a respected leader in the renewable energy industry. Each year, Siemens Gamesa produces over 5,000 wind turbine blades for use in on/offshore wind farms. Before manufacture, each of these blades undergoes a rigorous quality assurance process, as the structures can contain multiple defect types, including wrinkles in the fibreglass that can emerge during the production process. This has the potential to cause a catastrophic failure once deployed into a wind, so it is crucial that 100% of wrinkles are identified.
Measuring up to 75 metres in length, ultra-sound scans from the surface of each blade have to be meticulously inspected, involving a 6-8 hour process by a highly-skilled Quality Control engineer. Siemens Gamesa turned to Fujitsu to apply innovation to this time-consuming and costly process, and we were able to transform the process through a combination of AI Deep Learning technology and a highly effective co-creation approach.
At Fujitsu Laboratories of Europe, we are focusing on a co-creation strategy to solve real-world manufacturing problems in record time, using the knowledge base of Fujitsu’s extensive manufacturing expertise combined with our state of the art AI innovations. For Siemens Gamesa, our innovative AI technology uses a new framework that combines image and signal processing techniques with deep learning technology. This enables an approach which translates any raw data analysis problem into one involving image pattern recognition. In the case of the Siemens Gamesa quality control application, the solution is applied automatically to analyze and detect relevant patterns in Non-Destructive Testing (NDT) ultrasound data. NDT is a non-invasive technique that is used for determining the integrity of materials or structures and enables the detection of internal defects, a technique that is widely used in industry as part of manufacturing quality control processes.
Close collaboration between our teams and specialists from Siemens Gamesa highlighted the solution’s capabilities and performance through workshops and the completion of a Proof of Concept. The results far exceeded expectations, with Fujitsu Laboratories of Europe’s AI solution reducing the inspection region of the blade to just 20%, and therefore dramatically cutting the inspection time by 80% – resulting in significant cost savings.
Siemens Gamesa awarded Fujitsu the contract in March 2017, and the intention is that within a year from roll-out, the solution will be deployed across all of its factories around the world. We believe that this example highlights the key elements of real co-creation – where R&D, an imperative business requirement, and a strong customer relationship all work seamlessly together and demonstrate the power of AI to solve real-world problems.