Innovative digitisation projects: Cedoon quality assurance in screw production as part of the FOTEC research project ISI

Digitalisation is continuously transforming the way we work and opening up exciting new opportunities in a wide range of industries. In the context of the ISI (Innovative Software Systems and their Implementation) research project at FOTEC, the research company of the University of Applied Sciences Wiener Neustadt, another remarkable initiative dedicated to the AI trend is the Cedoon project. Innovative technologies are being used to solve problems in quality assurance.

Cedoon's software solution has been specially developed for the screw industry and is the first of its kind to integrate the entire supply chain worldwide. It saves manufacturing companies and their suppliers time and money, supports their employees and drives production forward.

Details about the project

The quality of a drill screw is determined by the drill-down time. According to the standard, screws whose screw-in time exceeds one second are of poor quality. The necessary tests have been difficult to perform so far, as a screw must be screwed into a standardised sheet metal and the time measured.

The simpler solution is to create an AI model that classifies drill-down time using optical methods. Around 1,200 screws were tested and images were taken, with the screw-in times distributed as follows, additionally classified as good / medium / poor.

The following confusion matrix clearly shows the quality of the trained network, although there are still too few measurements in the area of medium and poor screws.

Particularly in the area of good screws (< 1 sec.), an accuracy of over 99% was achieved with the help of the AI model used.