AMC1 145.A.30(d): what a compliant man-hour plan actually contains

Ask ten Part-145 quality managers where the man-hour plan requirement lives and most will point at 145.A.30(d). Ask what the plan must contain and the answers get vague. Yet the acceptable means of compliance is short and concrete — and the organisations that fail it usually fail on arithmetic, not intent. Here is what the AMC expects, and the waterfall method I teach for building a plan that survives a surveyor's first five minutes.

01What the AMC actually demands

AMC1 145.A.30(d) expects your man-hour plan to demonstrate that you have enough personnel, in the right categories, for the maintenance you have committed to. In practice that means four things:

• Built from real data. The starting point is last year's actual maintenance man-hours — not a guess, not headcount divided by aircraft.
• Adjusted for expected change. Fleet growth, new contracts, a type leaving the fleet — applied as an explicit factor.
• Net of non-productive time. Leave, training, sickness and meetings come off the top before you divide hours by people.
• Revised on deviation. When actual workload drifts significantly from the plan — the 25% figure used across this blog — the plan gets re-run, not annotated.

02The availability waterfall

The single biggest compliance error is dividing required hours by contractual hours. An engineer on a 1,840-hour contract does not produce 1,840 maintenance hours. Worked example, per engineer per year:

That is a 29% availability loss before a single task card is opened. If your local leave entitlement or training load is heavier — offshore rotations, for example, cut much deeper — your waterfall number will be lower. Use your number, not mine.

03From hours to headcount

Now the demand side. Say your mixed AW139/AW189 operation logged 41,600 productive man-hours last year across line and base work, and the commercial team is forecasting 10% fleet growth:

Two more checks make this a plan rather than a number. First, the category split: if only 60% of those 36 hold certification authorisation on type, you have ~22 signatories — run that against your peak two-week window, because averages hide the pinch. Second, the monthly profile: 45,760 hours spread evenly is 3,813 per month, but check clustering means some months run 30–40% hotter. The AMC plan that passes is the one that shows both.

04Where organisations fail it

• Headcount plans masquerading as man-hour plans. "We have 30 engineers for 8 aircraft" contains no hours and demonstrates nothing.
• Contractual-hour division. 45,760 ÷ 1,840 says 25 engineers. You would be eleven short on day one — this is the finding that keeps coming back in audits.
• No revision trigger. A plan dated January, untouched while hours run 30% over forecast, is evidence against you.
• Missing scope. Line stations, away-from-base work and subcontract oversight hours belong in the total. Surveyors ask for them specifically.

Do the waterfall once a year with honest numbers, profile it monthly, and tie it to a deviation trigger — that is AMC1 145.A.30(d) compliance in three sentences, and it is exactly what the calculator below automates.