top of page
Image by Rob Lambert


Graph neural network for shape recognition in the aluminium extrusion sector


NEED for AI:

AI can enable more accurate and timely analysis of complex data sets, automate processes, and identify patterns and insights that would be difficult or impossible to detect manually. By leveraging AI, companies can enhance their competitiveness, optimize their operations, and drive innovation in their industries.



Developement of a tool that automates the creation of graphs in combination with a Graph Neural Network. The tool takes an Euclidean dataset as input and creates a graph dynamically with relaxed geometrical rules. The GNN works as a grader, rewarding edges that lead to correct results.



  • A novel tool for data space preparation in the aluminum industry which includes a data model, library, and validated use-case in aluminum extrusion

  • A graph-structured data space for the experiment using open-source geometry databases

  • An open-source database for the market and research entities interested in digitalization for MTO/ETO companies' pre-production phase

Intellico is a digital company focused on providing AI-powered solutions for smart factories in Italy and Europe. Focus of the experiment GRAPHO is to build a new tool using cutting-edge techniques to improve Data Space preparation in the aluminum industry, including a data model and library, and to create a Data Space and validated use-case in the aluminum extrusion industry. The goal is to increase similarity detection and it will be tested on multiple databases gathered by Intellico.
This experiment aligns with Vanguard Region and AI Regio objectives, providing a novel tool based on Graph Neural Networks and Graph generation for a Data Space preparation in the aluminum industry, which will deliver a new way to organize Data Spaces for Manufacturing.

bottom of page