Predictive Maintenance
EDP Redes España needed to speed up electricity-network inspections, detect defects more accurately, and reduce the cost and environmental impact of manual review processes.
Reported impact
EDP Redes España cut inspection cycles dramatically, improved defect-detection accuracy, increased inspection frequency, and created a more scalable and lower-impact way to maintain network quality as electricity demand grows.
EDP Redes España and Capgemini replaced slow manual power-line inspections with an AI inspection stack built around computer vision, deep learning, and MLOps. The project began with a detailed review of inspection workflows, then introduced an MLOps platform that could train models from relatively small labeled datasets, plus an image-processing and inspection layer that attached geolocation data to network assets. The team used LandingAI and AWS technologies, first validating the approach in a proof of concept across three use cases and then extending it to nine more use cases in three countries. Helicopter images flow through the processing pipeline, AI models flag likely defects, and the inspection platform delivers prioritized results to field teams. Capgemini and EDP also ran change-management work so inspectors could adopt the new operating model without disrupting regular network maintenance.
Helicopter inspection images are ingested and processed, geolocation is attached to the relevant assets, MLOps-managed computer-vision models analyze the imagery for defects, and the inspection platform returns prioritized findings to network inspection teams.
"EDP Networks collaborated with Capgemini to implement a comprehensive inspection platform, leveraging deep learning, MLOps and computer vision technologies to optimize network inspections."
"The team opted for LandingAI and AWS technologies for the solution and began development with a proof of concept (PoC) that evaluated the AI models in three different use cases."
"The time required was drastically reduced from one year to just one month"
"This effort encouraged the company to implement a machine learning operations (MLOps) platform capable of training AI models with small datasets, as well as an image processing solution to support these models."
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
Snowflake Sample Modified · Aug 7, 2026 Uniper Volatile energy markets demanded faster access to trading, risk, and operational data than Uniper's aging reporting platform and fragmented systems could provide. 18.5 gigawatts of power-generating capacity Snowflake · Snowflake CoCo · Snowflake CoWork 
In-house Sample Est. · Jul 10, 2026 Aramco Aramco needs to turn extremely large and complex exploration, drilling, refining, logistics, and market datasets into faster, safer, and more reliable operational decisions while reducing uncertainty, downtime, and risk… five facilities recognized by the World Economic Forum as Lighthouse sites Dammam-7 · Ghawar-1 · Operations Co-Pilot 
Bain Sample Archived · May 26, 2026 TCL Photovoltaic Technology TCL Photovoltaic Technology's growth created more dispersed projects, more complex operating processes, and higher policy risk, so the company needed AI to improve efficiency, consistency, and control as it scaled. 10%-15% auditing process cost reduction AI service group · AI-powered recognition · Intelligent inspection 
