As anyone who ever tried to assemble anything knows, there's often a significant difference between what you are supposed to do to get the job done, and what actually works.
In factories' assembly lines, this means that blue collar workers often hit a wall, when trying to follow the Standard Operating Instructions (SOIs) they are given to complete a task, as they are seldom representative of the real situation.
To fix this, EIT Digital is launching AI-Aid, an innovation activity belonging to the Digital Industry Action Line, which will make SOIs smarter and more interactive, therefore reducing production and assembly errors and increasing production speed in the manufacturing industry.
This takes the shape of a mobile application which, by connecting with sensors positioned on the assembly line, is able to detect whether the worker is performing a specific sub-task well or not.
The app not only provides step-by-step instructions and sends user-specific, targeted nudges in case of any issue - for instance, "careful, in the last shifts this step led to problems" - but monitors the whole process so that, if a sub task is not completed correctly, the user is not able to proceed further.
"Most companies have databases of non-conformities, i.e. errors that occur at a specific point in time. We use this information combined with the standard operating instructions, to identify if a specific part of the task has problems and then we bring this sub process into an interactive format," DFKI's Associate Head of the Educational Technology Lab, Carsten Ullrich, explains, explains.
DFKI is leading the activity. Other partners of the initiative include London-based Digital Catapult organization, which will bring in its expertise about Internet of Things connectivity solutions, and business champion NEOCOSMO, a German company that works on micro-learning environments.
The Airbus corporation is also participating, as pilot customer. Development and commercialization of the solution will follow two tracks. Using Airbus' facilities as test-bed, researchers will learn about the problems and obstacles occurring when applying the technology in a real environment.
At the same time, they will set up a demonstrator which mirrors the original process in more controlled situation, in order to show potential customers the benefits of the system.