Rest-hour records that hold up when someone checks

Hours of rest is the compliance area most likely to be technically satisfied and evidentially weak at the same time. The crew rested. The watch rotation was sensible. And the record that proves it was assembled on a Thursday from memory, three weeks after the fact, by someone reconciling a paper sheet against a roster.

That gap between what happened and what can be shown is where the exposure sits — and it is also where the fatigue risk hides, because a record reconstructed after the fact cannot warn anyone while there is still time to act.

What the rules require

STCW sets minimum rest for anyone with watchkeeping duties: at least 10 hours of rest in any 24-hour period and at least 77 hours in any 7-day period. The 10 hours may be split into no more than two periods, one of which must be at least 6 hours long, and the interval between consecutive rest periods is capped. MLC 2006 covers the same ground from the employment side, with the alternative of a maximum-hours-of-work framing, and requires that records be kept and that seafarers receive a copy of their own.

The specifics matter less here than the shape. These are rolling limits, not daily ones. A day that looks fine in isolation can still break a 7-day total, and a split that looks reasonable can breach the minimum-block rule. That is not something a person scanning a sheet reliably catches, and it is not something a monthly reconciliation catches in time to matter.

"A rest-hour record produced after the voyage answers an auditor. A rest-hour record produced during the voyage answers the master, which is the one that changes an outcome."

Why reconstruction fails

Retrospective records fail in a predictable order. First they lose granularity — a call-out at 0300 becomes a rounded block. Then they lose the exceptions, which are precisely the entries an inspector is interested in. Then they start to converge on the limits, because a person filling a form from memory anchors on what the rule permits rather than on what occurred. The result reads as compliant and is worth very little as evidence.

The operational cost is larger than the compliance one. If the record only exists after the fact, nobody can see a 7-day total trending toward its limit while there is still a rotation decision available. Fatigue management becomes something you report rather than something you do.

Recording as it happens

The alternative is not more discipline; it is a shorter distance between the event and the entry. Rest is recorded onboard, by the person or the watch system that knows, at the time. Rolling totals are computed continuously rather than at month end, so an approaching breach is visible as a warning rather than discovered as a finding. Each seafarer can see their own record, which both satisfies the MLC copy requirement and creates the only reliable error-correction mechanism there is — the person who was there disagreeing with the entry.

This has to work without a connection, because the periods of highest workload are frequently the periods furthest from coverage. An entry made in open water and a roster change made ashore have to reconcile into one record rather than compete; that mechanism is the subject of how we keep data in sync across the fleet.

Done this way, the record stops being a document you produce for inspections and becomes an operational instrument you happen to be able to show. When an inspector asks, the answer is the same one the master has been using all voyage.

CruiseControl records time and rest against the crew record, computes the rolling totals, and keeps the entry available onboard whether or not there is bandwidth. The time-tracking and alerting modules cover the mechanics, and a demo is the fastest way to see it against your own watch patterns.