Real-time Analytics
Siemens Gamesa needed to cut blade-manufacturing errors and reduce rework because manual inspection and a partially automated laser-grid system could not deliver the precision, throughput, or standardization needed across expanding factories.
Reported impact
Siemens Gamesa reduced rework risk, improved manufacturing consistency, and created a faster path for scaling best practices across new and existing turbine-blade factories.
Siemens Gamesa and IBM Consulting turned a visual quality-control concept into a production-ready computer-vision system for turbine-blade manufacturing. IBM helped the Digital Ventures Lab design a camera array above each manufacturing station, process video on Microsoft Azure IoT Edge, and apply machine-learning models in real time so the laser grid system could validate fiberglass placement and later detect foreign objects in the mold. Microsoft Azure Machine Learning and Azure DevOps supported the cloud engineering and model lifecycle, while IBM Garage methodology accelerated iterative delivery from design to MVP in Aalborg. The architecture kept feedback close to the production line: images are analyzed with minimal latency at the edge, then the resulting guidance is projected back into the manufacturing process so technicians can act immediately. By standardizing the digital workflow on Azure and scaling it plant by plant, Siemens Gamesa created a reusable manufacturing-intelligence platform rather than a one-off inspection tool.
Cameras mounted above the blade stations capture video, Azure IoT Edge runs machine-learning inference on the footage, the laser-grid system returns placement guidance and defect alerts to technicians, and Azure services manage the broader model and delivery lifecycle.
"By processing video on the Microsoft Azure IoT Edge platform"
"We expect a payback period of about two and a half years for our Azure-based production system."
"the solution detects any errors or abnormalities in the surface of the blade"
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
IBM Sample Jul 29, 2025 Matrix Renewables Matrix Renewables needed to onboard and monitor a rapidly growing portfolio of solar assets without the delays, data gaps, and scaling limits caused by decentralized monitoring across multiple SCADA systems. 10% reduction in downtime IBM Maximo Renewables 
NVIDIA Sample Sep 30, 2026 EPRI EPRI needed to run advanced AI for sensitive utility documents and weather-planning workflows locally, because energy-sector research data could not leave controlled environments and shared compute delays limited how… 150 tokens/sec aggregate across 10 concurrent Power Chat users DGX Spark · Dell Pro Max with GB10 · NVIDIA CBottle IBM Sample Sep 28, 2026 OnticWorks.io OnticWorks.io needed its Elmer agentic platform to reason within the physical and operational context where work happens instead of relying on disconnected dashboards, emails, sensor feeds, and simulations that users… 12-month acceleration IBM Bob · IBM watsonx.ai · Red Hat OpenShift 