If you are a vehicle manufacturer dealing with defects in battery housings — this project developed a data fusion system that optimizes welding parameters to reduce scrap and improve quality.
AI-Driven Digital Twins to Reduce Scrap and Costs in Metal Casting and Welding
Imagine having a perfect digital mirror of your factory line that predicts mistakes before they happen. This system watches how metal is poured and welded in real-time, adjusting the settings automatically to avoid waste. It's like a smart GPS for manufacturing that ensures every part is perfect, even when using recycled materials.
What needed solving
Metal manufacturers struggle with high scrap rates and long time-to-market due to the inability to effectively use process data. This is worsened when using recycled materials which vary in quality.
What was built
A digital twin control system and a data fusion tool for casting and welding lines. It includes a material characterization system to track raw material properties.
Who needs this
Who can put this to work
If you are a casting plant dealing with out-of-specification parts — this project developed a digital twin setup that uses sensor data to automatically adjust process parameters and lower waste.
If you are a metal processor dealing with inconsistent quality in recycled raw materials — this project developed a material characterization system that ensures production resilience and stability.
Quick answers
What is the cost or pricing for this system?
Based on available project data, specific pricing is not mentioned, but the project aims to define a corresponding business model for market uptake.
Is this system ready for industrial scale?
Yes, the project is validating the system in realistic conditions using industrial use cases, including high pressure die casting parts and vehicle battery housings.
How is the IP and licensing handled?
Based on available project data, specific licensing terms are not provided, though the consortium includes 6 market leaders cooperating to maintain their market dominance.
How does this integrate with existing factory sensors?
The project develops a data fusion system specifically designed to integrate different sensors across production lines for cooperative analysis.
What is the timeline for deployment?
The project runs from 2023-01-01 to 2025-12-31, with the final phase focusing on market uptake and business model definition.
Who built it
The consortium is heavily weighted toward industrial application, with a 50% industry ratio consisting of 14 partners. The presence of 6 market leaders (including Fronius, Nemak, and Benteler) indicates a high level of commercial intent and ensures that the developed tools are tailored for large-scale enterprise deployment rather than just academic research.
Contact the Katholieke Universiteit Leuven research office regarding MetaFacturing
Talk to the team behind this work.
Contact us to connect with the MetaFacturing industrial partners for pilot implementation.