Facility Energy Usage Anomaly Analysis Builder

Analyze facility energy usage anomalies across meters, sites, weather, occupancy, equipment schedules, and utility bills to find savings and risk.

Prompt Template

You are a facilities analytics consultant helping diagnose energy usage anomalies and cost spikes.

Facility type: [office, retail store, warehouse, clinic, school, restaurant, hotel, manufacturing site]
Sites or meters in scope: [single building, multi-site, meter IDs, submetering available]
Utility types: [electricity, gas, water, steam, chilled water]
Data available: [utility bills, interval meter data, building management system, weather, occupancy, production, operating hours]
Time period: [months/years, interval granularity]
Known issue: [bill spike, overnight load, weekend usage, seasonal jump, site outlier, demand charge, leak suspicion]
Variables to normalize: [weather, degree days, occupancy, opening hours, production volume, square footage]
Equipment context: [HVAC, refrigeration, lighting, kitchen equipment, compressors, pumps, EV chargers]
Rate structure: [flat, time-of-use, demand charges, peak penalties, unknown]
Stakeholders: [facilities, finance, operations, landlord, sustainability, site managers]
Constraints: [limited meter detail, leased site, no BMS access, comfort requirements, safety rules]
Decision needed: [investigate, tune schedule, repair equipment, renegotiate rate, prioritize audit]

Create:
1. Data quality and normalization checklist.
2. Baseline model approach using weather, occupancy, hours, and site size where available.
3. Anomaly detection plan for bills, interval load, overnight baseload, weekend use, and site outliers.
4. Visual dashboard recommendations with charts and thresholds.
5. Root-cause hypothesis table linking patterns to likely equipment or process causes.
6. Savings estimate framework with confidence levels.
7. Investigation checklist for facilities teams before calling vendors.
8. Prioritized action plan by cost impact, urgency, effort, and operational risk.
9. SQL, spreadsheet, or BI calculation examples for key metrics.
10. Caveats for incomplete meter data, weather shifts, tariff changes, and one-time events.

Keep the analysis practical and do not recommend unsafe equipment changes without qualified review.

Example Output

Initial Finding

Warehouse B uses 38% more electricity per square meter than similar sites after normalizing for operating hours and cooling degree days. The excess is concentrated between 11 PM and 5 AM, when occupancy is near zero.

Anomaly Table

| Signal | Evidence | Likely Cause | Next Check |

|---|---|---|---|

| High overnight baseload | 42 kW steady load | HVAC schedule or compressor cycling | Compare BMS schedule to lease hours |

| Monday demand spike | 212 kW peak at 8 AM | Simultaneous equipment startup | Stagger startup sequence |

| Water bill jump | +29% month over month | Leak or irrigation schedule | Inspect meters after close |

Dashboard Metrics

- kWh per square meter adjusted for degree days.

- Overnight baseload as percent of daytime average.

- Peak demand by hour and weekday.

- Site outlier ranking with confidence notes.

Tips for Best Results

  • 💡Normalize before blaming a site; weather, hours, and occupancy can explain legitimate usage changes.
  • 💡Separate energy consumption from demand charges because the savings levers are different.
  • 💡Use interval data when possible; monthly bills hide overnight and weekend waste.
  • 💡Validate anomalies with facilities context before sending vendors on expensive investigations.