Back to industries
Energy
Wind production forecasting & market arbitrage
Client · Renewable energy producer · 4.5 GW capacity
A 72-hour wind production forecasting system combined with an arbitrage engine across spot, intraday and ancillary services markets — generating €12M/year additional market revenue.
01
Challenge
- Forecast errors causing significant balancing costs on the regulated market.
- Public weather models insufficient for turbine-by-turbine forecasting (terrain, wake effects).
- Historically deterministic market placement, not capturing intraday market value.
- Increased electricity price volatility requiring stronger operational reactivity.
02
Solution
- Hybrid forecasting model: weather ensemble (ECMWF, AROME) + ML model (Temporal Fusion Transformer) correcting biases turbine-by-turbine.
- Backtesting platform allowing traders to simulate placement strategies on 5 years of history.
- Stochastic optimization engine recommending the optimal split between D-1 spot, intraday and reserves.
- Integration with trading systems (ETRM) for a forecast → recommendation → order cycle in under 15 minutes.
03
Business impact
- −28% MAE on 24h forecast vs. baseline.
- +€12M/year additional market revenue on the 4.5 GW portfolio.
- −35% balancing cost vs. historical strategy.
- Platform extended to solar PV in year 2.
Other industries
Banking
Real-time transaction fraud detection
Insurance
Health underwriting automation with GenAI
Retail
Local assortment personalization & dynamic pricing
Manufacturing
Predictive maintenance on F&B production lines
Logistics
Route optimization & demand forecasting
Healthcare
Oncology patient pathway optimization
Let's identify where AI can create value in your business.
Book a 30-minute introductory call with a founder.