Individualized knowledge transfer in the field of production

Anyone who has tried to assemble anything knows that there is often a significant difference between what needs to be done to get the job done and what actually works. On factory production lines, this means workers often face problems. When they try to follow Standard Operating Instructions (SOIs), they rarely represent the actual situation.

To address this problem, EIT Digital, a leading European digital innovation and entrepreneurial education organization, launched the AI-Aid innovation activity “Just-in-Time AI-Based Problem-Solving and Training Support for the Digital Factory Floor” in 2018. This project addresses the problem of the growing knowledge gap between highly skilled employees and assembly line workers, caused by the Internet of Things (IoT).

Using the example of project partner Airbus, one of the largest players in the aerospace industry, this means: Final assembly of aircraft remains a labor-intensive process. It requires highly skilled assemblers who often have to work in confined spaces with components whose installation does not allow for much tolerance. Because of the manual nature, level of complexity and number of components, there is a risk of human error. Mistakes can be costly and time-consuming, as a single component misassembled or broken during the process can result in large parts of the aircraft having to be disassembled and reassembled.

The AI-Aid digital innovation activity assumes that knowledge assistance systems based on artificial intelligence can bridge the lack of knowledge on highly complex manufacturing floors by providing individualized, context-sensitive support to workers based on their previous human-machine interaction. The goal is to provide on-site problem-solving assistance through step-by-step instructions that respond to employee actions and long-term development by providing relevant information.

DFKI is leading the activity. Other partners in the initiative include London-based Digital Catapult, which is contributing its expertise on Internet-of-Things connectivity solutions, and business champion NEOCOSMO, a German company working on advanced learning environments. Airbus Group is participating as a pilot customer.

The project involved recreating a portion of an aircraft interior that contained sensors from Digital Catapult for various critical components. These sensors record, for example, whether a connection was made correctly, the order in which various components were assembled, and whether each component was properly aligned.

DFKI and NECOSMO are using the data obtained to provide a digital, adaptive knowledge and learning environment. The basis for this is the artificial intelligence-based knowledge and support service for intelligent production (“adwisar 2.0”). This architecture interprets production line data to provide possible support actions for workers.

NEOCOSMO integrated this service with the PIIPE microlearning and communication platform, which serves as the knowledge service in the project. To do this, intelligence support developed by DFKI was added to the personalization technology in PIIPE to provide the ability to proactively identify the individual knowledge needs of workers working in the factory and provide the knowledge needed for individual gaps.

On PIIPE, users can find articles, dossiers, step-by-step guides, videos and other learning materials in the sense of “learning nuggets” regardless of location and time. A company can offer its employees content channels to which they can subscribe. When new content such as articles, videos or knowledge dossiers are published or updated, PIIPE notifies users and helps keep them up to date.

The year-long EIT project has AI-Aid was able to show that IoT assets can collect valuable data to make manufacturing processes and training more effective. A multi-network solution has been implemented in Airbus’ future factory, enabling the company to communicate in ways not previously possible.