Definition
EEACS measures the frequency and severity of conflicting outputs between the energy optimization system and the environmental monitoring system over a defined period. A high EEACS indicates the two systems are coherent: energy optimization decisions do not trigger environmental monitoring alerts. A low EEACS indicates the systems are working at cross-purposes, with energy savings recommendations actively degrading the environmental conditions that the monitoring system is designed to protect.
The Measurement Gap
Measurement Gap and Decision
Decision Question: Are the energy optimization AI and the environmental monitoring AI producing conflicting outputs. With one reducing HVAC capacity for efficiency while the other alerts on the resulting environmental degradation?
What Goes Unmeasured Today:
Healthcare facilities increasingly deploy both energy optimization AI and environmental monitoring AI. These systems often operate independently. The energy optimizer reduces supply air temperature, adjusts damper positions, or modifies chiller staging to minimize energy consumption. The environmental monitoring system detects the resulting changes in temperature, humidity, or air pressure differentials and may generate alerts. Neither system knows what the other is doing. The energy manager sees energy cost declining. The facilities engineer sees environmental alert volume increasing. Nobody connects the two trends because the systems report to different dashboards and different stakeholders. In the worst case, the energy AI causes an environmental condition change, the monitoring AI alerts on it, a technician investigates, finds the HVAC system operating as the energy AI commanded, and closes the work order as "no issue found", because from the HVAC perspective, the system is doing what it was told to do. The conflict is invisible because no metric tracks whether one AI's outputs trigger the other AI's alerts.
Invisible Failure Mode:
Two AI systems optimize independently toward objectives that can conflict. Energy optimization degrades environmental conditions; environmental monitoring detects the degradation; neither system attributes the cause to the other. The organization pays twice (energy savings and alert investigation costs) while the conflict persists undetected.
Formula
Primary Formula:
EEACS = 1 − (Conflict_Events / Total_Optimization_Actions)
Variable Definitions:
| Variable | Definition | Source |
|---|---|---|
| Conflict_Events | Instances where an energy optimization action (setpoint change, staging change, load shed) was followed within 60 minutes by an environmental monitoring alert in a space served by the affected system | BAS optimization logs + environmental monitoring alerts (time-correlated) |
| Total_Optimization_Actions | All energy optimization actions executed during the measurement period | Energy optimization platform / BAS logs |
| 60-minute window | The temporal correlation window for associating optimization actions with environmental alerts | Configurable (default 60 minutes) |
Data Requirements:
Energy optimization action logs with timestamps, affected systems, and parameter changes. Environmental monitoring alert logs with timestamps and affected spaces. Building system mapping connecting optimization targets to monitored spaces. Time-correlation engine to match optimization actions to subsequent alerts.
Edge Cases and Limitations:
The correlation window (default 60 minutes) must account for HVAC system response time. Large buildings have thermal lag that may push environmental effects beyond the window. Environmental alerts that coincide with but are not caused by optimization actions (concurrent sensor drift) will inflate conflict counts; cross-reference with ESSDR to filter sensor-related alerts. Seasonal transitions naturally produce more environmental variability and may temporarily lower EEACS.
Proposed formula, pending validation pilots.