Downtime Events

Costing Downtime Events: A Playbook for Funding the Right Fixes

Costing every downtime event within 24 hours builds the Pareto that funds real fixes. The template, the math, and the weekly review that makes it stick.

Costing downtime by the event, not just by the hour, is what turns a maintenance log into a capital plan. An hourly rate tells you downtime is expensive; an event cost tells you which specific failure to kill first and how much you can afford to spend killing it. A gearbox failure that costs 14,600 dollars per occurrence and happens three times a year is a 43,800 dollar annual problem that justifies a 30,000 dollar redesign without further debate. Plants that cost individual events find their repair versus replace, spares, and RCA decisions get faster and better, because every argument arrives with a dollar figure already attached.

The structure is duration times hourly loss, plus repair spend, plus ramp-up and penalty costs. Work one: a 3.5 hour outage on a line costing 2,800 dollars per hour is 9,800 dollars of production loss. Repair adds 1,800 dollars in parts and two technicians for 4 hours at 55 dollars, another 440. Ramp-up ran one hour at half rate, 1,400 dollars more. Expedited freight on the replacement part cost 1,200. Total: 14,640 dollars for one event the maintenance log recorded as gearbox failure, 3.5 hours. The Downtime Cost per Event calculator standardizes this so every event in the plant gets costed identically in under five minutes.

The tail costs are where most estimates fall short. Restart scrap typically runs 2 to 5 times steady state reject rates for the first 30 to 60 minutes on process-sensitive lines, and on food, coatings, or plastics a purge cycle can add 500 to 3,000 dollars in material alone. Missed shipments trigger customer penalties, expedited freight at 3 to 10 times standard rates, and in automotive supply chains line-down charges that reach thousands per minute. Overtime to recover the schedule adds a 50 percent labor premium. Ignore the tail and a 14,600 dollar event books as 2,200, and the redesign that would prevent it never clears the hurdle rate.

Event costing powers the Pareto that hourly totals hide. Sum a year of costed events by asset and by failure mode and the concentration is always striking: the top 10 events typically carry 60 to 70 percent of total downtime cost, and the top three assets often carry half. Separate chronic from sporadic: forty 20 minute jams costing 900 dollars each total 36,000 dollars a year and hide below the radar, while one dramatic 12 hour breakdown at 35,000 dollars gets all the meetings. The chronic problem is usually cheaper to fix and nobody is working on it. The costed Pareto surfaces both on the same axis: dollars.

The levers rank by which cost term they attack. Duration shrinks with MTTR work: diagnosis is commonly 40 to 60 percent of repair time, so fault-finding guides, better HMI diagnostics, and one-line troubleshooting trees cut hours off events without touching wrenches. Critical spares on the shelf remove freight waits that turn 2 hour repairs into 18 hour events. Repair kitting, staging parts, tools, and permits before wrench time, borrows SMED thinking and cuts 20 to 30 percent from planned portions of a repair. Ramp-up losses shrink with documented restart procedures and centerline settings. And frequency, the multiplier on everything, falls only through RCA and precision maintenance.

Watch the failure modes in the costing practice itself. Costing only spectacular events skews investment toward drama and away from chronic bleeders. Letting each supervisor estimate freely produces numbers that inflate over time until finance rejects the whole system; use standard rates per line and a fixed template. Waiting weeks to cost an event loses the details, especially the tail costs, which nobody remembers by month end. Double counting labor already in the hourly rate overstates events 10 to 20 percent. And costing events without feeding a Pareto and an action list turns the whole exercise into bookkeeping; the number exists to force a decision.

Run it on a tight cadence. Every event over 30 minutes gets costed within 24 hours by the supervisor using the standard template: duration, hourly rate, parts, labor, ramp-up, penalties, five minutes of work. Weekly, the top five events by dollars get reviewed in the operations meeting, each with a named owner and a due date for the countermeasure. Monthly, the rolling Pareto by asset and failure mode gets refreshed and the RCA queue reprioritized; any single event over 10,000 dollars or any asset over 25,000 dollars year to date triggers a formal RCA. Quarterly, verify the fixes: did the costed events actually stop recurring?

World class plants cost 95 percent of qualifying events within a day, hold event cost estimates to a standard template finance has blessed, and can show a rolling 12 month Pareto where last year's top three failure modes no longer appear. Their capital plans read like their downtime logs: every reliability project cites the events it kills and pays back in under 18 months. The trajectory is measurable: plants that adopt disciplined event costing typically cut total downtime cost 25 to 40 percent in two years, mostly by finally funding fixes for chronic events that hourly reporting had rendered invisible. Start by costing last week's worst event tonight.

Published 2026-07-02.