Operational Availability Calculator
Ao, Ai and Aa from MTBF and the pieces of downtime that actually dominate it. The gap between inherent and operational availability is usually logistics, not reliability — and it's where sustainment KPPs are lost.
Inputs
Reliability
Mean time between failures. If you have a failure rate λ in failures/hour instead, MTBF = 1/λ.
Downtime
MLDT is waiting for a part. MADT is waiting for a decision, a form, or a person. Both are usually larger than the repair itself.
Preventive maintenance (optional)
Result
More than half of your downtime is waiting, not fixing. Improving MTBF barely moves Ao from here — spares positioning and approval turnaround will.
Need this rolled into a sustainment KPP, an RBD, or an FMECA tied to your requirements?
Sign up →Frequently asked questions
What's the difference between Ai, Aa and Ao?
Ai = MTBF/(MTBF+MTTR) counts only corrective repair — it's what a design organisation can control, and it's the number vendors quote. Aa = MTBM/(MTBM+M̄) adds preventive maintenance. Ao = MTBF/(MTBF+MDT) adds logistics and administrative delay, and is the only one that describes what the operator actually experiences.
Why is Ao usually so much worse than Ai?
Because MDT is dominated by waiting. A two-hour repair that waits six days for a part is a 146-hour downtime event. This is why sustainment KPPs are written against Ao and why a program can meet every reliability requirement and still fail its availability requirement.
Where does the "nines" language come from?
Availability expressed as a percentage: 99% is roughly 88 hours of downtime a year, 99.9% is 8.8 hours, 99.99% is 53 minutes. Each additional nine costs roughly an order of magnitude more, which is why specifying one more nine than the mission needs is an expensive habit.
What does this tool NOT do?
Reliability block diagrams, redundancy and k-of-n configurations, Weibull or non-constant failure rates, sparing optimisation, FMECA, or mission-phase-dependent availability. Those are in the Defense pack.
Upgrade for full RAM analysis
- Reliability block diagrams with redundancy and k-of-n
- FMECA linked to requirements and risks
- Sparing analysis and provisioning models
- Sustainment KPP tracking against milestone gates
- Availability feeding the review package directly