Popular searches
Not sure where to start?Talk to our team about running sovereign AI in your environment.
↑ ↓ Navigate↵ OpenEsc Close
AES built an AI practice from zero to 150+ production models, cutting maintenance repair costs by two-thirds and saving $1M a year on avoided technician trips.
AES, a leading renewable-energy company, set out on a two-year AI transformation to predict equipment failures, optimize renewable output, and improve load distribution. The catch: it had no data scientists on payroll when it started. AES needed a platform to build a practice fast and scale models across the business.
When the two-year initiative began, AES had no data scientists on payroll. To move from a standing start to production AI, the company needed a way to build both a team and a platform that could turn a handful of high-value use cases into repeatable, scaled results.
AES built its practice on H2O AI Cloud, using it to develop and deploy models across the business. The team monitored model performance and bias with H2O MLOps, put models in front of business users through H2O Wave applications, and drove reuse with H2O AI Feature Store so features discovered for one project could power the next.
Rather than boil the ocean, AES started with a few targeted use cases, proved value, then expanded. Wind turbine failure prediction, smart-meter analysis, and predictive maintenance became early wins that seeded a wider program.
The predictive maintenance program cut the cost of crane-assisted repairs by two-thirds, from about $100k to $30k each, and delivered millions of dollars in savings. A smart-meter project eliminated 3,000 non-essential technician trips, saving roughly $1M a year. Reliability improved too, with a 10% reduction in Customer Average Interruption Duration Index (CAIDI).
AES now runs 35 production models for wind turbine failure prediction alone, all exceeding 90% accuracy, and more than 150 models in production supporting 85 AI use cases across the organization.
“We didn't have a single data scientist on payroll when we started our AI transformation. We started with a few high value use cases, built a team, and found a platform to accelerate our results.”Sean OttoDirector and Head of AI at AES
AES is the largest US-based global power company and a leading developer, owner, and operator of renewable, thermal, LNG, and battery storage facilities. Headquartered in Arlington, Virginia, the company is a top supplier of clean energy to corporations worldwide, working to accelerate the transition to cleaner energy through renewables, storage, and grid modernization.
In 2024, AES generated enough electricity to power 17 million people, employing roughly 9,100 people across 12 countries. Learn more at https://www.aes.com/.