The Steel Institute at the „1. WZL/IMA Industrial Internet of Things Hackathon“
In the age of industry 4.0 new topics arise which can only be successfully solved by applying innovative methods such as artificial intelligence. Meanwhile, besides science, industry is also working on questions concerning smart and optimised production chains.
In order to generate the widest possible range of solutions for current problems, research institutions and companies organise so-called hackathons. These are competitions in which teams have to solve defined tasks within a fixed period of a few days by programming and using algorithms. Those interested in programming have the opportunity to use their skills to generate a joint solution contribution for computer-based questions of industry and science. While the best programming groups can enjoy prize money, the organizers receive interesting approaches to their tasks. Due to this win for all, hackathons are arousing increasing interest worldwide, including the interest of RWTH Aachen University.
Under the slogan "1st WZL/IMA Industrial Internet of Things Hackathon", a hackathon for machine learning in production technology took place in Aachen from 19 to 21 January 2018. This event, at which almost 60 highly motivated participants jointly programmed over 48 hours, was organized by the Friends of the WZL Machine Tool Laboratory and the Cybermetics Lab IMA/ZLW & IfU of RWTH Aachen University. The prizes were sponsored by Feintool and the VDI Aachen.
One of the four tasks of the hackathon was assigned by Prof. Münstermann (Integrity of Materials and Structures). It consisted of recognizing, analysing and evaluating voids on pictures of ductile fracture surfaces using artificial intelligence. Dr. Brinnel, head of the working group for Structural Integrity, represented the IEHK in the jury, which decided on the awarding of the prize money.
A total of five teams of young programmers worked on this task during the weekend. In the end, they were able to present impressive approaches and solutions. The algorithms developed can now be used to quantitatively determine modeling input variables for damage mechanics models.
Professor Münstermann and the members of the working group for Structural Integrity would like to thank all participants of the teams "Gyminformatics","Supervisioners","Deep Unicorn","Cooky Factory" and "digitalPriests" for the informative discussions and the great results.
At this point, we would like to congratulate the teams "DigitalPriest" and "Gymformatiker" for their strong third place at the 1st WZL/IMA Industrial Internet of Things Hackathon and the junior award, respectively.