August 6 marked the culmination of an incredible 3-month collaboration between our students at the Karlsruhe Institute of Technology (KIT) and industry leaders Trelleborg and IBM. Five ambitious projects tackled real-world challenges in data science and machine learning - and the results exceeded our expectations.

All five projects delivered impressive results, with two standout winners chosen for their exceptional performance:

1) AI-Powered Document Processing for Aerospace: Our students developed an intelligent system that automates feature extraction from complex engineering drawings and documents. They created a human-in-the-loop solution that could save Trelleborg up to 500 working hours annually - transforming a manual process that currently takes 2,000 hours per year.

2) Predictive Analytics for Manufacturing Quality: The team developed an ML-powered stiffness prediction tool for polymer manufacturing. Their approach shows potential to save ~80 days annually in setup time and reduce ~€200K in scrapping costs by helping predict test outcomes before expensive production runs.

These projects demonstrate how university-industry partnerships can accelerate innovation and solve real business challenges. Our students delivered production-ready solutions with measurable ROI in just 3 months.

Looking ahead: We’re excited to continue working with Trelleborg and IBM on scaling these solutions and exploring new opportunities in AI-driven manufacturing and document intelligence.

Thank you to all participating students for their exceptional work and professional collaboration with our industry partners.