Industrial Daily Reports Across Daylight Saving: Define the Day Before Comparing Totals
A daily energy total changes on the clock-change weekend although the operating load appears steady. Before treating the change as a plant fault, check what the report means by a day. Local calendar days and fixed 24-hour intervals can have different boundaries.

Original conceptual illustration; not a live ProSight Software screen, specific product or customer installation.
Key takeaways
Name the site's IANA time zone, not just a fixed UTC offset.
Convert each report boundary independently.
Show elapsed duration and completeness beside comparisons.
1. Specify the question the report answers
A local production day, a shift and a trailing 24-hour window are different reporting requirements. Write the start and end rules before selecting a chart aggregation. Include the site location and the treatment of shifts crossing midnight.
For a local-day report, a useful definition is local midnight inclusive to the next local midnight exclusive. The half-open boundary prevents the same boundary reading being assigned to two adjacent periods. How interval readings are allocated at that boundary must also be documented.
2. Keep instants and civil time distinct
An unambiguous timestamp identifies an instant. A local clock label describes that instant using a zone's rules. RFC 3339 defines timestamp offsets; it does not define the full changing civil-time rules for a location. IANA's time-zone database supplies location-based historical rules and is updated when those rules change.
Store or preserve unambiguous timing evidence and carry the named reporting zone separately. Australia/Adelaide is more useful for future local-day boundaries than permanently applying +09:30 or +10:30. A site's reporting zone should not silently change because an analyst views the report from another state.
3. Work through a constant-load example
For an illustrative Adelaide local-day calculation, 4 October 2026 has 23 elapsed hours between its midnight boundaries; 4 April 2027 has 25. At a hypothetical constant 10 kW, those durations correspond to 230 kWh and 250 kWh respectively. A normal 24-hour day at the same load gives 240 kWh.
The difference does not, by itself, indicate an efficiency change. Dividing each complete period's energy by its actual elapsed hours gives 10 kW in this example. Real equipment varies, so preserve production, weather and operating context rather than using duration normalisation as a complete explanation.
The dates and calculations are worked examples, not customer results. Verify the deployed time-zone data and current official rules when implementing a report, especially for sites in different jurisdictions.

Original worked-example diagram. Values are illustrative and assumptions are explained in the text; this is not a product specification or field result.
4. Resolve both boundaries independently
Construct the start and next end in the named local zone, then convert each to UTC for querying. Do not create the end simply by adding 86,400 seconds to the start when the requirement is the next local midnight.
On a backward clock change, some local labels occur twice. Retain the offset or other unambiguous instant information so those readings remain distinguishable. On a forward change, certain local clock labels do not occur. Do not manufacture readings merely to fill a visually regular hourly table.
5. Separate duration from missing coverage
A 23-hour local day can be completely observed. A 24-hour day with an outage can be incomplete. A report needs both the expected elapsed duration and the actual valid measurement coverage to distinguish them.
For interval energy, confirm whether each value belongs to the interval start or end and how partial boundary intervals are treated. For cumulative registers, independently check resets, missing boundary values and interpolation assumptions. Changing the time zone does not repair those measurement problems.
Avoid comparing a partial day with a completed day without a clear label. If you estimate a missing interval, keep the estimated result distinguishable from the observed total and state the method.
Record the named reporting zone and period definition.
Resolve each local boundary using current time-zone rules.
Record the UTC boundaries and elapsed seconds.
Check valid-data coverage and boundary allocation.
Reconcile the displayed result against an independent calculation.
6. Test before using the report operationally
Use a constant-load test and a separate time-varying test across ordinary, forward-change and backward-change dates. Confirm adjacent periods neither overlap nor leave a gap, and verify an export reproduces the same boundaries.
Agree the reporting contract for your actual ProSight Software deployment, including available aggregation, export and time-zone behaviour. This guide describes requirements to test, not a list of guaranteed platform features. Energy reporting is also separate from payroll treatment of clock-change shifts and from application-specific billing rules.
Related guide and project resources
Technical references
Need help with your application?
Send a representative dataset, equipment details and the calculation or reporting rule you need to verify. Confirm the supported functions and integration requirements for your actual deployment.

.png)
Comments