AI System for Opening Energy Consumption Variances for Review

The problem

Total consumption may conceal a department-level variance in day-to-day operations. When meter consumption, production volume, shift, outdoor weather, department and maintenance records are scattered across sensor, document, site or team records, the issue may not be identified in time.

This fragmented data delays preparation of the consumption variance file and forces teams to search for supporting evidence again by hand. Before making a decision, the energy manager or energy engineer must complete missing records and verify each inconsistency individually.

A common exception is that an additional shift, weather conditions or a planned test may increase consumption. Incomplete or incorrect automation that does not check this context could lead to the wrong equipment being shut down or allow a genuine loss to continue, so no physical or operational action can be taken directly.

What we set out to improve

• Proportion of alerts confirmed by the engineer as genuine variances
• Proportion of outputs corrected or rejected by the energy manager or energy engineer
• Time between the source event and the review output

How the system works

1. 01 Define the review scope and the role of the authorised energy manager or energy engineer
2. 02 Retrieve the inputs: meter consumption, production volume, shift, outdoor weather, department and maintenance records
3. 03 Check the source identifier, date, calibration or version, and data recency
4. 04 Check the meter and department identifiers
5. 05 Compare consumption with production and weather context
6. 06 Check the exception: an additional shift, weather conditions or a planned test may increase consumption
7. 07 Place low-confidence or inconsistent records in a separate specialist queue
8. 08 Write the source identifier, rule, output and time to the audit trail
9. 09 Have the energy manager or energy engineer approve, correct, defer or reject the output

Methods we used

• Normalised time-series variance checks
• Source, format, time and required-field checks for meter consumption, production volume, shift, outdoor weather, department and maintenance records
• A missing-data warning instead of an estimate when inputs are incomplete or inconsistent
• A separate exception check outside the main rule
• Joint retention of the source, rule result, consumption variance file and human decision

Where people stay involved

Data and security

Meter consumption, production volume, shift, outdoor weather, department and maintenance records are accessed with the narrowest permissions required for the task. Commercial, location and employee information is masked where possible, and only the necessary portion of the data is sent to the model rather than the complete raw dataset. Access, outputs and the energy manager's or energy engineer's decision are recorded; production does not begin until the organisation approves the retention period and data location.

Who this suits

Frequently asked questions