evidoria

← Back to browse

Good practice Imported

SOLIDA — SISAB and Schneider Electric's AI-Optimised HVAC Cuts Energy Use Across 87 Stockholm School Properties

Sweden · Stockholm · See the Sweden profile · See the Stockholm profile

Evidence: Observational / pre–post Top 89% 27/100 · Ask Evidence Copilot about this practice

SISAB, Stockholm's municipal school-property company, ran Schneider Electric's SOLIDA AI system across 87 school buildings for four years, cutting electricity use 8.9% and district heating 3.1%, a net 250-tonne CO2 reduction, now scaled to about 125 schools.

87 properties
School properties covered (2019-2024)
8.9 %
Electricity consumption reduction
8.7 %
Average per-building electricity reduction
3.1 %
District heating reduction
2.8 %
Average per-building heating reduction
250 tonnes CO2-eq
Net CO2 reduction
125 schools
Schools scaled up to after pilot
~3 GWh
Annual energy savings after scale-up
SOLIDA — SISAB and Schneider Electric's AI-Optimised HVAC Cuts Energy Use Across 87 Stockholm School Properties

Details

Maturity
Scaling
Promoter
SISAB (Skolfastigheter i Stockholm AB) / Schneider Electric / Myrspoven
Period
2019-2024
Region (NUTS)
SE11
Keywords
K-12 facilities management, school buildings, energy efficiency, sustainability

Context

SISAB (Skolfastigheter i Stockholm AB), Stockholm's municipal school-property company managing roughly 600 school and preschool buildings on a 340-million-Swedish-krona annual energy budget covering about 250 GWh of demand, partnered with Schneider Electric and Swedish AI firm Myrspoven to test whether machine-learning HVAC control could cut energy use without harming indoor comfort.

Activities

Built on Schneider Electric's EcoStruxure Building Operation platform and marketed as SOLIDA, the system used more than 9,900 sensors to continuously adjust heating, ventilation and airflow across 87 school properties over a four-year evaluation period, with a predictive layer forecasting demand up to 48 hours ahead.

Results

Across the 87 properties, electricity consumption fell 8.9% overall (8.7% average per building) and district heating fell 3.1% (2.8% average per building), for a net reduction of about 250 tonnes of CO2-equivalent once infrastructure emissions were subtracted. Following the pilot, SISAB extended the system to roughly 125 schools and preschools, reporting close to 3 GWh in annual energy savings.

Conclusions

The results come from a joint SISAB-Schneider Electric report rather than an independent third-party audit, and carbon savings would be substantially higher in grids more carbon-intensive than Sweden's, meaning the CO2 figures are specific to Sweden's largely fossil-free power mix and don't directly generalise to other countries.

Implementation
Indicative cost
High (€500k–€5M) — Sensor deployment (9,900+ units) and platform licensing across dozens of properties, funded from SISAB's own ~340M SEK annual energy budget.
Time to results
Long (> 3 years) — Four-year pilot evaluation (2019-2024) followed by scale-up to roughly 125 schools.
Staffing & skills
SISAB (municipal school-property company, implementing body), Schneider Electric (technology vendor, EcoStruxure platform), Myrspoven (Swedish AI firm, machine-learning modelling)

Conditions for success

  • Dense sensor network (9,900+ sensors) for granular building data
  • Predictive forecasting layer (48-hour horizon) to smooth heating peaks
  • Multi-year evaluation before scale-up decision

Common failure modes

  • Results reported by the operator/vendor rather than independently audited

Commonly funded by

Own resources / municipal budget

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful