AI Decision Engine

Every decision balances resident care, comfort and electricity cost.

Patient / Resident

  • Wearable device

    Temperature trend · activity · presence

  • RTLS

    Current location · room occupancy · movement

  • Care requirements

    Temperature preference · mobility · environmental restrictions

Environment

  • Location

    Room · facility zone · orientation

  • Weather

    External temperature · humidity · sunlight · forecast

Building

  • HVAC status

    Current modes · setpoints · indoor temperature

  • Smart meters

    Readings · electricity demand · peak proximity

  • Device availability

    Blinds · fans · gateway health

AI Decision Engine

Vendor-agnostic · continuously re-evaluated

Outputs

Air conditioning
Curtains / blinds
Fans

Priority hierarchy

  1. 1Resident safety
  2. 2Care requirements
  3. 3Comfort
  4. 4Energy cost optimisation

Energy optimisation can never override safety or care requirements.

Decision example

Margaret Wilson — Room 203

Inputs

Resident

Margaret Wilson

Location

Room 203

Room temperature

25.8°C

Outside

33°C

Solar load

High

Resident status

Resting

Preferred range

24–26°C

Electricity demand

Near peak

AI decision

  • • Close west-facing blind to 75%
  • • Keep AC at 25°C
  • • Fan ON — Low
  • • Avoid reducing AC setpoint to 23°C

Maintains resident comfort while avoiding approximately 0.6 kW of additional cooling demand.

Estimated saving

$0.19 / hour

Confidence

92%

Human approval queue

Routine adjustments apply automatically. Changes over 2°C, multi-room actions, care-sensitive zones and amenity changes are held here.

Automation level

Cautious

Small single-room adjustments run automatically. High-impact changes wait.

Care-critical changes always stop for a human decision, at every level.

3 changes waiting for your approval

Small adjustments run automatically. Large or care-sensitive changes stop here first.

Raise setpoints across South Wing by 3°C

High impact

20 rooms · 20 residents

24°C → 27°C for 90 minutes

Forecast demand 96 kW against a 90 kW network peak threshold at 15:00.

Multi-room change greater than 2°C — requires facility manager approval.

≈ 6.4 kW peak avoided · ≈ $38 AUD saved todayConfidence 89%4 min ago

Reduce cooling in Clinical Area

Critical change

Clinical Area · 4 residents

23.8°C → 25.5°C

Low occupancy detected and clinical rounds finished 20 minutes ago.

Care-sensitive zone — energy optimisation cannot proceed without care staff sign-off.

≈ 1.8 kW cooling demand avoidedConfidence 74%11 min ago

Close all west façade blinds to 100%

High impact

West Wing · 18 blinds

70% → 100% closed until 17:30

Extreme solar load forecast, outdoor peak 34°C at 15:00.

Full blackout of resident rooms — daylight amenity change needs approval.

≈ 4.2 kW cooling demand avoidedConfidence 83%23 min ago

Recent AI actions

Room 203 switched to Eco Mode

Auto

Resident left the room 18 minutes ago (RTLS + wristband presence).

AC setpoint adjusted 24°C → 27°C

≈ $0.84 / hour savedConfidence 96%2 min ago

West Wing blinds closed to 70%

Auto

High afternoon solar load detected on west-facing façade.

12 motorised blinds moved to 70% closed

≈ 3.1 kW cooling demand avoidedConfidence 91%9 min ago

Dining Area pre-cooling

Auto

31 residents expected to arrive within 20 minutes (activity schedule + RTLS trend).

AC pre-cool started ahead of occupancy peak

Peak demand shifted, ≈ 2.4 kW avoided at peakConfidence 88%14 min ago