Optimal maintenance interval calculator
Find the replacement interval with the lowest cost per operating hour, from the Weibull shape and scale of the failure mode, the cost of a planned replacement and the cost of a failure. Free, with the method shown.
Maintenance interval calculator
Runs in your browser. Nothing you enter is sent to a server or stored.
How to use it
Enter the Weibull shape (beta) and scale (eta, in hours) of the failure mode, from the Weibull calculator; the cost of a planned replacement; and the cost of an unplanned failure, including lost production and any secondary damage. The calculator compares age replacement (replace at a set age or on failure, whichever comes first) with inspection (enter the cost of an inspection and the detection start p: a failure at age t is caught if an inspection falls between p times t and t). It also shows the cost of running to failure. Use the long-term basis for a population replaced again and again, and the short-term basis for the expected cost of a single cycle.
How to read the result
The curve is the cost per operating hour against the replacement age, and its lowest point is the optimal interval. A flat bottom means the exact interval matters little; a sharp one means it matters a lot. Planned replacement only pays when beta is above 1, so failures rise with age, and a failure costs more than a planned replacement. With beta at or below 1 there is no minimum, and the answer is to inspect or run to failure.
Why not replace at the MTBF?
Because for a wear-out failure mode about half the units fail before they reach the MTBF: with beta = 2.5, 52% do. When a failure costs much more than a planned replacement, the cost-optimal interval sits well short of the MTBF.
What to do with the numbers
How to calculate the optimal maintenance interval
The method step by step, with the cost curve.
Maintenance intervals: balancing cost, reliability and risk
What to do when the cheapest interval leaves reliability too low.
Maintenance strategy
Intervals for every failure mode on an asset, from your data.
Questions people ask
What if beta is close to 1?
Then failures are random in time, and replacing on a schedule does not reduce them. Inspect for early signs or run to failure instead.
Which costs should I include?
For a failure: the repair, the production lost during the downtime and any secondary damage. For a planned replacement: the parts, the labour and any production lost to the planned stop.
Where do beta and eta come from?
From a Weibull analysis of the failure mode’s history: paste the times to failure into the Weibull calculator. At least five failures give a fit worth using.
Who built it
The tool was built by Pardus Consulting and uses the same methods the firm applies on client work. When the question is bigger than a calculator, the analysis is run on your own CMMS data and signed by a registered professional engineer.