Case Study: Fagor Ederlan achieves real-time defect detection and improved quality monitoring with Microsoft Power BI

Developers use cloud power to boost quality of automotive parts manufacturing

Fagor Ederlan, a global automotive-parts manufacturer in Spain, needed to detect defects in aluminum molding parts much earlier than the typical one-month customer inspection delay. The challenge was compounded by legacy and proprietary machines, diverse communication protocols, huge millisecond-level data volumes, and low-bandwidth factory networks. Working with Microsoft Power BI and Azure services (Azure IoT Hub, Azure Functions, Azure Stream Analytics and Azure Machine Learning), the developer team prototyped a cloud-based solution to ingest and analyze production data securely and reliably.

Working with Microsoft Power BI, the team deployed a lightweight Windows Service to compress and send CSVs to Azure IoT Hub, two Azure Functions to decompress and extract features, Stream Analytics to join streams and invoke Azure Machine Learning models, and a Power BI dashboard for real-time monitoring. The solution cut defect-detection latency from the prior one-month timeline to near real-time, conserved network bandwidth with zip-file strategies and automated uploads, and produced a scalable, repeatable system that can be applied across Fagor Ederlan’s factories.

Fagor Ederlan