The flexibility utilities already own, put to work.
Hydro, controllable loads, curtailable generation: utilities and grid operators have long held flexibility. Orvio shows the quarter-hours where deploying it pays – and the load position it is deployed from.
Swiss utilities and grid operators own more flexibility than they monetize: storage hydro, controllable loads, curtailable generation. Under the single-price regime, that flexibility has a market every quarter-hour – but deploying it blind risks working against the system. Two things make the difference: knowing where the system is heading, and knowing the group's own position when it gets there.
How Orvio fits
The system forecast delivers direction and the imbalance energy price per quarter-hour, refreshed every 15 minutes. The load forecast supplies the group's own position alongside it: trained on the portfolio's metered data, retrained daily, and aggregated along the portfolio's structure – from meter group to balance group.
Portfolio management brings the two together: the expected imbalance per quarter-hour with its cost, the specific move that improves it, and a backtest that demonstrates the net impact on realized data before it is relied on live.
82% directional hit-rate when Orvio forecasts the system at least 100 MW short or long.
- 68%
- Any signal
- 76%
- Forecast ≥ 50 MW
- 82%
- Forecast ≥ 100 MW
- 90%
- Forecast ≥ 200 MW
T-15 system-imbalance forecast vs realized Swissgrid data, last 90 days (about 2,240 quarter-hours at the 100 MW threshold), against a 54% naive baseline. The stronger the forecast signal, the more reliable the direction.
What the accuracy is worth
Modelled on the last 90 days of realized Swiss prices, curtailing feed-in when the system is long was worth about €116 per curtailed MWh on average – the fattest side of the market – while two-sided flexibility such as hydro earns on both. That average is concentrated: the strongly negative-priced quarter-hours carried roughly nine-tenths of the curtailment value in the window, which is the case for forecasting those windows rather than curtailing on a schedule. Each additional point of T-15 hit-rate is worth on the order of €2,500–6,000 per MW per year, depending on whether the asset moves one way or both.
These are illustrative, gross figures – modelled from Orvio's forecast against realized prices, not executed trades. They apply in the clear-signal quarter-hours – when the system runs at least 100 MW short or long, about a quarter of all quarter-hours – not around the clock. They sit before asset and opportunity cost and assume a price-taker position, so the per-MW value falls as the position grows, which matters most at utility scale. The 90-day window shifts with the season. Orvio delivers the signal and the measured impact; dispatch and its risk stay with the operator.
What changes in practice
For balance-group and dispatch teams, the work shifts from settling deviations after the fact to positioning ahead of them: a known load position per quarter-hour, a validated system signal, and the windows where the flexibility on hand earns instead of costs.
See what flexibility could earn.
Book a demo and see the signal on live data, applied to real assets.
Pricing scales with the portfolio. Book a demo for a quote.
FAQ
What data does Orvio need from us?
The portfolio's metered consumption history, uploaded directly into Orvio (CSV, Excel, Parquet, or ZIP). Weather is sourced automatically; the first load forecast stands within days.
Does it fit our systems?
Yes. Platform for the teams, REST API for integration – both deliver the same timestamped, scored values.
How do the parts fit together?
The system signal is the core: the direction and price of Swiss imbalance. Load forecasting adds the group's own position, and portfolio management turns the two into the move worth making. Each stands on its own or runs as one platform.
How accurate is the system forecast?
82% directional hit-rate over the last 90 days when the signal flagged a clear position of at least 100 MW – scored against realized Swissgrid data.