Inherent steady-state availability¶
The long-run operating fraction of a repairable item calculated as MTBF divided by MTBF plus mean corrective repair time, excluding planned maintenance and support delays.
Core Idea¶
Inherent steady-state availability is a predicted long-run operating fraction for a repairable item when only corrective repair counts as downtime. Under a stated stationary failure-and-repair model, \(A_i=\mathrm{MTBF}/(\mathrm{MTBF}+\mathrm{MTTR})\). MTBF is the mean time the item operates between defined failures; MTTR is the matching mean time to correct a failure. NASA's displayed steady-state formula uses constant failure and repair rates as model assumptions.[^ref-4063f9f524df]
Inherent excludes planned maintenance and waits for parts or maintenance resources. Steady-state refers to the limiting fraction as time increases, not an observed percentage for one month. Achieved and operational availability include other kinds of downtime and can give different values for the same equipment.[^ref-4063f9f524df]
Scope of Application¶
Use this measure for a repairable item with defined operating and failed states, matching MTBF and corrective MTTR estimates, and a justified long-run model. A NASA radar-receiver design study compares three candidate configurations using the inherent formula.[^ref-3e72fb497f2e] An EPA wastewater report applies it to sewage pumps and motors and explicitly excludes administrative repair delay.[^ref-40edad52973f] These are calculations for a design or equipment model, not observed long-run histories.
A service calculation that includes preventive maintenance or supply and maintenance waiting has a different downtime boundary.[^ref-4063f9f524df] It may be valuable, but its percentage should not be called this exact inherent measure without changing the accounting.
Clarity¶
An availability number needs four labels: which item, what counts as a failure, which downtime counts, and whether the result is a long-run model or a finite observation. Without those, two reports of “availability” may appear to disagree while using different definitions. The corrective-only boundary separates the inherent calculation from achieved or operational availability.[^ref-4063f9f524df]
Manages Complexity¶
Repeated operating and repair episodes become two mean durations and one ratio. Longer MTBF raises the predicted fraction if repair time is fixed; shorter MTTR raises it if MTBF is fixed. The simple number remains useful only while the item, failure boundary, corrective-repair scope and model horizon stay attached to it.[^ref-4063f9f524df]
Abstract Reasoning¶
Start with the same repairable item and failure definition for both inputs. Confirm that MTTR counts corrective work rather than planned maintenance or support waits. Then divide MTBF by MTBF plus MTTR, with both durations in the same unit. For a fixed target, different MTBF/MTTR pairs can qualify; the target alone does not choose the least costly design. NASA's receiver study compares the estimated costs of alternatives that all reach about 0.990.[^ref-3e72fb497f2e]
The quotient alone does not establish operational availability when supply or maintenance waiting enters downtime.[^ref-4063f9f524df] It also does not prove that the assumed model rates were observed; that requires separate evidence.
Knowledge Transfer¶
The same calculation applies to unlike repairable equipment when the roles stay fixed. The NASA radar receiver and EPA pumps differ in technology, yet each has a specified item, mean operating interval, mean corrective repair and ratio.[ref-3e72fb497f2e][ref-40edad52973f] The arithmetic is broadly portable; the specialist name still requires corrective-only time accounting and a long-run horizon.
Examples¶
Radar receiver design. NASA CR-61382 compares proposed MTBF/MTTR pairs of 200/2.02, 300/3.03 and 350/3.54 hours. Each gives approximately 0.990 inherent availability. They are alternative designs, not three measured field records.[^ref-3e72fb497f2e] Here the receiver is the item, MTBF is its mean operating interval, MTTR its mean corrective repair, and the ratio the modeled long-run fraction.
Sewage pumps and motors. The EPA report uses 8760 hours MTBF and 13 hours MTTR and says administrative repair delay is excluded.[^ref-40edad52973f] The calculation from those inputs is \(8760/(8760+13)\approx0.998518\), or about 0.9985 to four decimal places. The report prints 0.9986.[^ref-40edad52973f] The value here is independently recalculated. Pumps and motors fill the same four roles in a different setting.
Relationships to Other Abstractions¶
Current abstraction Inherent steady-state availability Domain-specific
Parents (1) — more general patterns this builds on
-
Inherent steady-state availability is a kind of Ratio Prime
Inherent steady-state availability is the ordered MTBF-to-total-cycle-time ratio for a repairable item under corrective-only downtime accounting.
Hierarchy path (1) — routes to 1 parentless root
- Inherent steady-state availability → Ratio → Comparison
Neighborhood in Abstraction Space¶
Inherent steady-state availability sits in a sparse region of the domain-specific corpus (85th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Supply Chain & Inventory Management (29 abstractions)
Nearest neighbors
- Mean time between failures — 0.86
- Reset (military) — 0.82
- Failure Analysis — 0.80
- Backorder — 0.80
- Engineering Critical Assessment — 0.80
Computed from structural-signature embeddings · 2026-10-17
Not to Be Confused With¶
- MTBF alone: it is the numerator; repair time also affects availability.
- Achieved or operational availability: those count planned maintenance or additional support delays under different rules.[^ref-4063f9f524df]
- One month's uptime percentage: an observed period is not by itself a steady-state prediction.
- Generic high availability: IBM's service example shows the formula but does not establish this exact corrective-only long-run boundary.[^ref-f15da2e821ef]
- Any ratio: Ratio is the broad mathematical relation; this entry fixes failure, corrective repair and limiting-horizon roles.
References¶
[^ref-4063f9f524df]: Tim Adams, Availability: What Is It?, NASA Kennedy Space Center (June 2024), pp. 1–3, especially the type comparison on p. 2 and steady-state formulas and assumptions on p. 3. https://extapps.ksc.nasa.gov/Reliability/Documents/Availability_What_is_it.pdf [^ref-3e72fb497f2e]: James H. Nelson, Methodology for the Systems Engineering Process, Volume III: Operational Availability, NASA Contractor Report CR-61382 (March 1972), Appendix C, printed pp. 75–82, especially radar-receiver alternatives on p. 76 and cost comparison and decision on pp. 78–82. https://ntrs.nasa.gov/api/citations/19730003260/downloads/19730003260.pdf [^ref-40edad52973f]: Charles W. Mallory and Robert Waller, Application of Selected Industrial Engineering Techniques to Wastewater Treatment Plants, EPA-R2-73-176 (February 1973), section IX, printed p. 101 (sewage pump and motor inherent-availability calculation and administrative-delay boundary). https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=9100TFS3.TXT [^ref-f15da2e821ef]: IBM Cloud, “Understanding high availability,” formula and 30-day server illustration. https://cloud.ibm.com/docs/resiliency?topic=resiliency-understanding-ha