Reliability-Based Spare Parts Provisioning Calculator

Estimate service-level stock for independent constant-rate spare demand across a fleet, planning horizon, and procurement lead time.

Engineering screening tool—verify inputs and results independently. Use at your own risk.

Fleet and planning inputs

Provisioning result

Recommended stock s*—usable interchangeable spares

Additional spares Δs—

Expected failures μ—

Achieved service level—

Current service level—

Current stockout risk—

The recommendation is the smallest whole-number stock count meeting the selected Poisson service probability. Procurement approval still requires cost, repair, pipeline, shelf-life, criticality, and event-scenario review.

Probability of demand by stock count

Bar height is Poisson probability; the blue bar marks recommended stock.

Probability (%)

Demand during coverage period (failures)

Stock s	Service level P(X ≤ s)	Stockout risk P(X > s)

Worked example

Consider 25 continuously exposed items with MTBF 50,000 operating hours, a 2-year planning horizon, and 6-month lead time.
Use a 95% service target and 8 usable spares on hand. The model gives μ = 10.95 expected failures.
The minimum recommendation is 17 spares, achieving 96.90% service. Current stock covers only 23.65%, so modeled stockout risk is 76.35% and the screen indicates 9 additional spares.
Interpret this as an independent constant-rate demand baseline, not a purchase authorization. Add repair returns, open orders, costs, criticality, and common-cause scenarios.

How to use

Select MTBF for a defensible useful-life value in operating hours, or Annual failure rate for an exposure-adjusted rate.
Enter the positive whole-number installed population and the active reliability input in its displayed unit.
Enter planning horizon in years and procurement lead time in months; both contribute to demand exposure.
Enter a target strictly between 0% and 100% and current usable stock as a whole number.
Select independent failures only when common-cause and clustered demand are negligible. The alternate choice displays a model-limit warning.
Select Calculate to update the single recommendation, supporting metrics, probability plot, and sensitivity table.
Use Reset example to restore the numbered example.
Read achieved service with stockout risk; discrete stock means achieved service can exceed the target.
Correct red validation errors and investigate amber warnings before using any output.
Do not use this model for wear-out, batch/common-cause failures, changing fleets, repair-pool optimization, safety assurance, or procurement approval without independent analysis.

Theory and method

Equations used

Let N be installed population, λ annual failures per item, H planning horizon in years, L lead time in months, C target probability, s stock, and X cumulative demand. MTBF mode uses λ = 8760/MTBF for continuous annual exposure.

μ = Nλ(H + L/12)

P(X = k) = e^−μ μ^k/k!

P(X ≤ s) = Σ_k=0^s e^−μ μ^k/k!

s* = min{s ≥ 0 : P(X ≤ s) ≥ C}; Δs = max(0, s* − s₀)

Counts are non-negative and dimensionless; λ is failures/(item·year), exposure is years, and μ is expected failures. The direct recurrence computes the exact Poisson CDF for μ ≤ 700.

Assumptions and limits

The model assumes a fixed population of statistically similar items, independent failures, constant useful-life hazard, and immediate one-for-one consumption. It excludes repair returns, condemnation fraction, cannibalization, pipeline orders, shelf degradation, obsolescence, varying utilization, changing population, intermittent demand, wear-out, and cost optimization. Clustered/common-cause failures can materially understate stockout risk; use scenarios, event trees, empirical overdispersion, or a validated negative-binomial model instead.

References

Continue in TriboSolver to refine maintenance scenarios and independently document component reliability assumptions where an applicable workflow is available.