How to Calculate Unplanned Downtime Cost Without Double Counting
A practical manufacturing framework for separating theoretical revenue exposure, realizable production loss, incident costs, and improvement break-even.
Read guideDowntime economics · Analyzer
Estimate realizable economic loss from an unplanned equipment stoppage, test uncertainty, and calculate the downtime reduction needed to justify an improvement.
Deterministic methodology · Explicit assumptions · No raw plant inputs sent to analytics
Most downtime calculators multiply a revenue rate by stopped time. That is useful as a theoretical exposure, but it can overstate the cash or contribution actually at risk.
This analyzer keeps theoretical lost revenue separate from realizable economic loss. The realizable layer uses expected good output, contribution or another documented avoidable value, plus non-overlapping incremental incident costs.
At 100 units/hour, 95% expected yield, and $12 contribution per good unit, two hours of downtime represents $2,280 of forgone contribution. Adding $600 idle labor, $500 recovery overtime, $250 startup scrap, and $400 of non-overlapping downstream consequence produces $4,030 of realizable economic loss. The same event at $20 revenue per good unit shows $3,800 of theoretical lost revenue, but that revenue figure is deliberately not added to the $4,030 realizable loss.
For an economic decision, contribution or another documented avoidable production value is usually more defensible than full revenue. Revenue can still be useful as theoretical exposure, which is why this tool reports it separately.
Treat each category as a distinct consequence. If lost contribution already captures the economic effect of missed production, do not add another downstream amount for the same missed units. Add only incremental labor, scrap, logistics, disruption, or other costs that are genuinely separate.
Use the demonstrated sustainable rate of the constrained or actually affected process during the relevant production context. Avoid an unconstrained nameplate rate unless it is demonstrably achievable and economically relevant.
No. It is a deterministic sensitivity range based on the uncertainty percentages you enter. It does not assign probabilities to the low, base, or high cases.
Compare the required avoided downtime with what the proposed improvement can realistically prevent over the same decision horizon. For recurring failures, use observed annual incident frequency instead of assuming a generic industry rate.