About us

LOG XY S.R.L.S. is an Italian technology company based in Como. We build artificial intelligence, blockchain and web-based software that turn business challenges into concrete, measurable advantages, and we follow each project from the first consultation to the delivered product. Our team has more than 20 years of experience.

What we do

  • Artificial intelligence: machine learning and AI agents that analyse data, automate repetitive tasks and work with existing CRM, ERP and databases.
  • Blockchain and Web3: traceability of physical products, with a verifiable digital history along the supply chain.
  • Web-based software: cloud platforms that run in the browser on any device, from complete business management to specialised solutions for sport, with dashboards, KPIs and real-time data.
  • DIA, the Digital Intelligent Assistant: voice interaction between operators and metal 3D printers. Integrated with EOS machines for laser powder bed fusion (LPBF), DIA lets operators talk to the system and receive spoken notifications in real time.

What we did in DIAMOND

DIAMOND (Digital Intelligent Assistant for Manufacturing) is an experiment of the second Open Call of WASABI, the European project that develops a white-label shop for digital intelligent assistants in manufacturing. LOG XY developed DIA and integrated it into the WASABI White Label Shop, together with two partners:

  • f3nice Srl, the manufacturing SME that designs and runs the LPBF experiments and validates DIA in its industrial additive manufacturing workflows, on an EOS M 290;
  • Confindustria Emilia-Romagna Ricerca, the Digital Innovation Hub of the experiment.

What the experiment delivered:

  • Voice access to the machine. Operators ask DIA for the machine state and the process values of the EOS M 290, read from EOSCONNECT Core, and hear the answer. DIA runs on Open Voice OS with a Rasa assistant.
  • Defect detection, layer by layer. An AI pipeline finds keyhole and lack-of-fusion defects in the layer images of the build. On 12 test specimens it reached 96.8% per-layer accuracy, and 95.5% on shapes it had not seen in training.
  • Early warning on a real build. Replaying the data of a production build on the EOS M 290 at f3nice that later failed, DIA raised an overheating warning 1 hour 43 minutes before the machine stopped.
  • No changes to the machine. DIA uses the data that already exist: read-only access to the machine interfaces, the layer images and the job file, with no new sensors.
  • This shop. The first DIA skill is published here, in LOG XY's instance of the WASABI White Label Shop.

The aim is fewer failed builds, less scrap, better documented quality and a deeper understanding of the process.

Funding

Funded by the European Union. DIAMOND has received funding from the WASABI project, funded by the European Union's Horizon Europe research and innovation programme under grant agreement No 101092176. Views and opinions expressed are those of the author(s) only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them.

Contact

Write to admin@logxy.it, use the contact form or visit www.logxy.it.