Overtime fatigue risk: the hidden staffing shortfall in your man-hour plan

Ask a hangar supervisor how the summer went and you'll hear: "We coped — the lads did the hours." Now ask the quality manager what the overtime report says. Those extra hours are not resilience. They are an unbudgeted staffing shortfall, and they come with a fatigue risk that no amount of goodwill cancels. Here is the math that converts an overtime report into the headcount number your man-hour plan should have contained.

01Overtime is headcount in disguise

A full-time maintenance engineer delivers roughly 150–160 productive hours per month once you strip leave, training, sickness and non-productive time. That is the denominator your AMC1 145.A.30(d) man-hour plan already uses. So every month of overtime has a direct translation:

Worked example from a composite AW139 base operation, Q3 figures:

Read the table the way a surveyor will: the plan said 8 engineers, but the operation actually consumed the hours of 10.3. That is a +19% planned-vs-actual deviation — approaching the 25% revision trigger — and it never appeared as a staffing decision. It appeared as payslips.

02The fatigue curve you bought with those hours

Overtime is not a linear purchase of capacity. Human-factors data in maintenance consistently shows the same pattern past a normal duty day:

• Hours 9–12: measurable drop in inspection reliability — missed-fault rates climb as vigilance degrades. This is why IOGP-690 §25C caps offshore duty at 12 hours, and why your SMS should treat routine 12-hour hangar days the same way.
• Beyond 12 hours: error likelihood roughly doubles; judgment and memory tasks — exactly what certifying a task card demands — degrade first.
• Chronic load: six-day weeks and short rest gaps accumulate sleep debt that single days off do not repay. The risk compounds across the roster period, not the day.

And the fatigue concentrates on exactly the wrong people: your certifying staff. Support mechanics can be rotated; the B1/B2 licence holders signing the critical-path tasks are the ones staying late, because their signature is the bottleneck. You are applying the highest workload to the highest-consequence role.

03The three numbers to report every quarter

Fold this into the quarterly review your man-hour plan already requires:

• Overtime ratio: OT hours ÷ planned hours, per month. Above 10% sustained, the plan is understating demand — flag it before the deviation review forces the issue.
• Hidden FTE gap: the formula above, per license category. An overall 2-FTE gap that is actually 2 B2s is a different problem than 2 mechanics.
• Peak duty days: count of duties exceeding 10 hours, by individual. If three certifying staff carry 80% of the long days, you have a single-point-of-failure with a signature on it.

Do this and the next hiring conversation changes shape. It stops being "can we afford another engineer?" and becomes "we are already paying for 2.3 engineers in overtime premium — at 1.5× rate — and receiving fatigued hours instead of fresh ones." At a 1.5× premium, 1,050 quarterly OT hours cost the equivalent of 1,575 straight-time hours: more than the fully loaded cost of two permanent hires, with none of the authorisation depth.

04What good looks like

A healthy plan holds sustained overtime under 5% of planned hours, treats anything above 10% as a revision trigger, and models the peak two-week windows — check overlaps, vacation season — where overtime always originates. The plan should absorb the peak with rostered capacity; overtime should be the exception that handles unscheduled defects, not the mechanism that delivers scheduled checks.