Per-well allocation,
compared to the field.
Real Volve data.
Allocation drift is silent. A producing well over-reports for months before metering catches up. Cross-well comparison sees it the day it happens. Six real Volve wells, six years of daily production records, every day where a well’s reported share diverges from its peer-expected share by more than ten percentage points.
No cd commands.
No clone. Just click.
The question every allocation analyst asks.
For each producing well on each day:
- Reported share — the well's reported barrel count divided by the field's reported barrel count.
- Expected share — the well's on-stream hours divided by the field's on-stream hours.
- Drift — the difference between the two, in percentage points. A drift greater than 10 percentage points flags the day as a deviation event.
The demo runs this scan across the full Volve archive and returns the top 8 most significant drift events plus a per-well summary. Honest read: not every drift event is a meter error or an over-report. Some are physically explainable. The demo surfaces them; an allocation analyst confirms them.
Five roles, five wins.
Allocation analyst
A daily scan that flags the wells worth re-checking against test-separator runs. Less time hunting; more time confirming.
Production engineer
Operating drift surfaces by date and well. Pair it with maintenance records to spot meter degradation events.
Reservoir engineer
Anomalies in reported vs expected share often indicate communication or coning effects between wells — reservoir-relevant signals.
CFO / commercial
Each percentage point of mis-allocation on a large field rolls up to material revenue mis-attribution. Earlier detection narrows the gap.
Joint-venture partner
Continuous third-party-verifiable comparison of every well’s reported share against the rest of the field, by construction.
What changes for you
Today an allocation discrepancy surfaces during the monthly close, when test-separator runs disagree with day-to-day reported numbers. By then the drift has run for weeks. The cleanup is a manual journal entry against the production accountant's notebook.
This scan runs every day, on every well, against every peer. The drift events surface on day one, not on day twenty-seven. The conversation with the test-separator schedule becomes a confirmation, not an investigation.
Same data your historian already collects. Same peer field already producing alongside the well in question. Different question shape, asked every day instead of every month.
A drift event is a flag, not a finding. Real allocation comes from test separators, well tests, and metering audits. What this demo proves is that the question can be asked daily, at scale, on commodity CPU, with the field's existing daily production data. The conversation it enables is the value.
Volve Field daily production data · Equinor 2018 · equinor.com/energy/volve-data-sharing · CC BY 4.0.
7 producing wells · daily reports from 2008-2016 · the same archive Demo 05 (V.L. on Volve ESP) reads.