A compiled reference of 62 verified statistics on manufacturing and industrial downtime: what an hour costs by source and sector ($2.3 million in a large automotive plant and $36,000 in consumer goods per Siemens; a $124,669 median across 3,215 industrial businesses per ABB; about $25,000 per facility per MaintainX), how often plants go down and for how long (27 hours a month per large plant, 69 percent with a monthly outage on critical equipment), what causes it, how much maintenance is still reactive (21 to 67 percent depending on the survey and how it defines reactive), what preventive and predictive programs are claimed to save (12 to 18 percent and 30 to 40 percent per the Department of Energy; 52.7 percent less unplanned downtime for less-reactive manufacturers per NIST), the $18.4 billion government estimate of U.S. unplanned downtime losses, and the injury, OSHA citation and FDA Form 483 data that describe the same machines from the regulator's side.
Key statistics
7 highlights from this report
Key statistics
Key takeaways
The hourly cost of downtime is whatever the survey sample makes it, from $25,000 to $2.3 million, and the useful numbers are the ones with a stated sample and a published distribution. What has not changed across the surveys is the reactive share: half or more of facilities still use run-to-failure in every survey that asked whether it is used at all (2020, 2024, 2025).
An hour of unplanned downtime runs from about $25,000 at an average facility to $2.3 million in a large automotive plant; the spread is the sample, not a disagreement.
ABB's $124,669 median across 3,215 industrial businesses is the citable 'typical plant' figure; 16 percent of its respondents are above $500,000 an hour.
Large plants lose 27 hours a month, 326 hours a year; incidents are down 41 percent since 2019 but each one takes 81 minutes to recover from, up from 49.
NIST's $18.4 billion a year is the government figure for U.S. unplanned downtime losses; the '$50 billion' that circulates has no traceable primary.
Half or more of facilities still use run-to-failure in every survey that asked whether it is used at all (2020, 2024, 2025); only 46 percent of manufacturers acting on downtime report measurable improvement, and 72 percent say undocumented fixes drive recurring downtime (L2L).
Predictive programs are credited with 35 to 45 percent less downtime by DOE, 30 to 50 percent by McKinsey and 50 percent by Siemens' own customers; NIST's measured 52.7 percent gap between reactive-light and reactive-heavy plants sits just above that band.
The maintenance standards OSHA cites most, lockout/tagout and machine guarding, drew 4,060 citations in FY2025, and FDA cited drug equipment cleaning and maintenance under 21 CFR 211.67(a) and (b) 165 times.
How we built this report
Every figure was compiled in September 2026 from the publisher's own report, database, regulation text or spreadsheet and verified against the original before publishing.
- Primary publishers only
Cost and frequency figures from Siemens, ABB, MaintainX, L2L, Fluke and Plant Engineering's own reports and releases; savings ranges from the Department of Energy's FEMP guide and the NIST survey; injury rates from BLS tables; citation counts from OSHA's final FY2025 data; FDA counts from the agency's own spreadsheets.
- Surveyed separated from extrapolated
Each vendor survey is labeled as the vendor's own, with its sample size; Siemens' Fortune 500 total and Fluke's weekly dollar totals are extrapolations and are named as such or left out.
- Samples kept apart
Siemens' large-plant sample, ABB's industrial mix and MaintainX's cross-industry sample answer different questions, so their figures sit side by side on a labeled ladder rather than averaged, and Fluke's weekly incident counts are not charted against Siemens' monthly ones.
- Roundup numbers excluded
The $50 billion, $260,000 an hour, 82 percent and 80 percent-equipment-failure figures that rank for this query could not be traced to a primary and appear only in the unverified section.
- Arithmetic labeled
Monthly and annual bills, lost-shift costs, the share above $500,000 an hour and downtime's share of the NIST total are RapidEye arithmetic and are marked Computed with their inputs.
- Independent review
Written by one co-founder, reviewed by the other before publishing.
Scope caveat: none of the cost surveys is a probability sample of U.S. manufacturing. Siemens covers large plants in four sectors, ABB covers industrial businesses across eleven sectors, MaintainX covers facilities of all sizes and industries, and L2L and Fluke are vendor surveys with their own screens. The Department of Energy savings ranges date from 2010 and summarize older studies. NIST's totals are in 2016 dollars. Regulator counts (BLS, OSHA, FDA) are administrative or government statistical records rather than vendor samples.
What an hour of unplanned downtime costs, by source
Eight published per-hour figures, each labeled by publisher and year; bar length is on a log scaleThe 92-to-1 spread between the top and bottom bars is a sample difference, not a disagreement: Siemens surveyed large automotive, heavy-industry, oil and gas and consumer-goods plants; ABB's median covers 3,215 industrial businesses across eleven sectors; MaintainX's average covers 1,165 facilities of all sizes, 31 percent of them manufacturing. Fluke's per-hour figures are omitted because the vendor published no method and its two releases disagree.
The reactive share, 2010 to 2025
Five sources, five definitions of "reactive"; read the wording under each numberof maintenance resources and activities still reactive
U.S. DOE FEMP guide2010of industrial facilities use a run-to-failure method
Plant Engineering survey2020run a run-to-fail strategy as their primary approach
ABB survey, 3,215 respondents2023of facilities still rely on run-to-failure (5% as their only program)
MaintainX survey, 1,165 respondents2024take a reactive approach to maintenance
L2L survey, 600+ U.S. manufacturers2025ABB's three shares sum to 100, so its 21 percent is the share whose main strategy is run-to-fail; Plant Engineering's and MaintainX's shares exceed 100 together, so theirs count any use of run-to-failure. L2L asked about the overall approach, and the DOE figure is a share of resources, not of facilities. None of the five is comparable to the others, and they show the same thing: half or more of facilities still use run-to-failure in every survey that asked whether it is used at all (2020, 2024, 2025).
What each maintenance strategy is claimed to save
Every published range on this page, with who published it and what kind of evidence it is| Comparison | Claimed effect | Publisher, year | Kind of evidence |
|---|---|---|---|
| Preventive vs. purely reactive | 12% to 18% cost savings | U.S. DOE FEMP guide, 2010 | Estimate summarizing older studies |
| Predictive vs. preventive alone | 8% to 12% cost savings | U.S. DOE FEMP guide, 2010 | Estimate summarizing older studies |
| Predictive vs. reactive-heavy facility | 30% to 40% or more | U.S. DOE FEMP guide, 2010 | Depends on reliance on reactive work |
| Functional predictive program, industrial average | 35% to 45% less downtime; 70% to 75% fewer breakdowns; 25% to 30% lower maintenance cost; 10x ROI | U.S. DOE FEMP guide, 2010 | Guide cites unnamed independent surveys |
| Less reactive vs. more reactive manufacturers | 52.7% less unplanned downtime; 78.5% fewer defects; 81.7% more direct maintenance spend | NIST (Thomas and Weiss), 2021 | Survey split; authors call it anecdotal |
| More predictive vs. more preventive (proactive group) | 18.5% less unplanned downtime; 87.3% fewer defects | NIST (Thomas and Weiss), 2021 | Smaller subgroups |
| Well-executed predictive maintenance | 5% to 20% less facility downtime; 10% to 30% lower inventory; 5% to 20% more labor productivity | Deloitte infographic, 2023 | Consulting estimate, unsourced |
| Predictive maintenance, typical | 30% to 50% less machine downtime; 20% to 40% longer machine life | McKinsey article, 2018 | Consulting estimate, chemicals context |
| Condition-based vs. time-based users | 42% vs. 35% reported uptime increase in the past year | ABB survey, 2023 | Self-reported, 3,215 respondents |
| Siemens' own Senseye customers | 50% less unplanned machine downtime; 40% lower maintenance cost; payback within three months | Siemens, 2024 | Vendor deployment data, not survey |
Only the NIST rows and the ABB row rest on a stated sample; the DOE guide summarizes studies it does not name, Deloitte and McKinsey state ranges without sources, and Siemens' row is its own customers. The ranges still cluster: predictive programs are credited with 35 to 45 percent less downtime by DOE, 30 to 50 percent by McKinsey and 50 percent by Siemens' own customers; NIST's measured 52.7 percent gap between reactive-light and reactive-heavy plants sits just above that band.
Manufacturing downtime, by the numbers
All 68 figures, 62 verified and 6 computed, grouped by theme, each quoted from the publisher's own report, database or regulation text and independently verified, followed by the figures that circulate and could not be verified.
Rates the sources imply
The six figures below are RapidEye arithmetic on the verified sources in the groups that follow: a large automotive plant's monthly and annual bill at Siemens' hourly rate, the price of one lost shift at three different hourly figures, the share of ABB's respondents above $500,000 an hour, downtime's share of NIST's total maintenance bill, the order-of-magnitude comparison of hours per year between two vendor surveys, and the year-over-year rise in FDA Form 483s. Each names its inputs.
Statistic 1
Siemens' 27 lost hours a month at its $2.3 million automotive hourly rate is about $62.1 million a month, or about $745 million a year, consistent with Siemens' own rounded 'approaching $750 million a year' on another page of the same report; its stated annual figure is $695 million.
RapidEye Research, computed from Siemens' 2024 downtime cost report (27 hours x $2.3 million x 12)
Statistic 2
An eight-hour shift lost at ABB's $124,669 median hourly cost is $997,352; the same shift at MaintainX's $25,000 average is $200,000; at Siemens' $2.3 million automotive rate it is $18.4 million. That is the same 92-to-1 spread as the hourly figures.
RapidEye Research, computed from ABB's 2023 survey median, MaintainX's 2024 survey average and Siemens' 2024 automotive figure (x 8 hours)
Statistic 3
In ABB's 2023 distribution, 16 percent of industrial businesses put an hour of unplanned downtime above $500,000 (7 + 6 + 3 percent), while 23 percent put it at $50,000 or less (10 + 13 percent).
RapidEye Research, computed from the hourly-cost distribution in ABB's 2023 reliability survey report
Statistic 4
NIST's $18.4 billion in annual unplanned downtime losses is 8.3 percent of its $222.0 billion estimate of all U.S. manufacturing maintenance costs and losses; the untraceable '$50 billion a year' figure that circulates is 2.7 times the government estimate.
RapidEye Research, computed from the NIST 2021 maintenance cost survey and analysis ($18.4 billion / $222.0 billion)
Statistic 5
L2L's 30 hours a month is 360 hours a year of all downtime, more than half of it unplanned, so at least 180 unplanned hours; Siemens' large-plant figure is 27 unplanned hours a month, 324 a year (Siemens reports 326). The two surveys agree on the order of magnitude, not the number: their definitions and samples differ.
RapidEye Research, computed from L2L's 2025 survey press release and Siemens' 2024 downtime cost report
Statistic 6
FDA's system-generated Form 483 count rose from 4,056 in FY2024 to 4,862 in FY2025, an increase of 19.9 percent; drug 483s rose 27 percent, device 483s 18 percent and food 483s 18 percent.
RapidEye Research, computed from the FDA inspection observations summary tabs, FY2024 and FY2025
What this means: The sources publish hourly rates and monthly hours separately; multiplying them out is what turns a survey headline into a number a plant manager can compare against a maintenance contract or a spare-parts budget.
What an hour of unplanned downtime costs at a large plant
According to Siemens' 2024 downtime report (siemens.com), which draws on Siemens' own survey of 181 maintenance, engineering and IT professionals at large industrial organizations, an hour of unplanned downtime now costs a large automotive plant $2.3 million, more than $600 a second, while the low end is $36,000 an hour in fast-moving consumer goods. The same Siemens report puts the annual cost at $695 million for a large automotive plant and $253 million for the average large plant in its sectors, and says plants lose 27 hours a month to unplanned downtime, down from 39 in 2019, while the average time to restore production rose from 49 to 81 minutes. Siemens' $1.4 trillion figure for the world's 500 biggest companies is an extrapolation, not a survey result, and its Senseye customer gains are the vendor's own deployment data.
Statistic 7
An hour of unplanned downtime costs a large automotive plant $2.3 million, more than $600 a second.
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 8
At the low end, an hour of downtime costs $36,000 in Fast Moving Consumer Goods.
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 9
The per-hour cost of downtime in automotive is 2x its 2019 level; in heavy industry it is 4x (a 113% and 319% rise respectively, against 19% US price inflation over 2019-23).
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 10
Annual downtime cost for a large automotive plant is $695 million (1.5x 2019); for a heavy-industry plant, $59 million (1.6x 2019).
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 11
The average large plant in the surveyed sectors loses $253 million a year to unplanned downtime.
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 12
Unplanned downtime costs the world's 500 biggest companies almost $1.4 trillion a year, 11% of their revenues (down 6% since 2022).
Siemens, 2024 downtime cost report (Siemens extrapolation to the Fortune Global 500, not survey data)
Statistic 13
Plants average 25 unplanned downtime incidents a month, down from 42 in 2019 (41% fewer).
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 14
A large plant loses an average of 27 hours a month (326 hours a year) to unplanned downtime, down from 39 hours a month in 2019.
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 15
Average time to restore production after a downtime incident rose from 49 minutes (2019) to 81 minutes.
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 16
In automotive and heavy industry, hours lost to unplanned downtime have halved over five years; FMCG is the only sector where downtime hours increased.
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 17
SME manufacturers' downtime costs reach $150,000 an hour at the top end.
Siemens, 2024 downtime cost report (Siemens assertion; not from its survey sample, no method given)
Statistic 18
Almost half of surveyed manufacturers now have a dedicated predictive-maintenance team, twice the 2019 share; nine in ten do some condition monitoring.
Siemens, 2024 downtime cost report (Senseye predictive maintenance; own survey of 181 maintenance, engineering and IT professionals at large industrial organizations)
Statistic 19
Siemens reports that its own Senseye predictive maintenance customers achieved a 50% reduction in unplanned machine downtime, 55% increase in maintenance staff productivity, 40% reduction in maintenance costs, and 85% improvement in downtime forecasting accuracy, with payback within three months.
Siemens, 2024 downtime cost report (results from Siemens' own Senseye customer deployments, not survey data)
What this means: Siemens' hourly figures are the highest in the literature because its sample is large automotive, heavy-industry, oil and gas and consumer-goods plants; they are the right benchmark for a 1,000-person plant and the wrong one for a 60-person job shop. The trend inside the report matters as much as the level: incidents are down 41 percent and hours down about a third since 2019, but each incident now takes 32 minutes longer to recover from.
What maintenance strategy saves, per the Department of Energy
According to the Department of Energy's Federal Energy Management Program guide to operations and maintenance (energy.gov), more than 55 percent of maintenance resources and activities at an average facility were still reactive as of 2010, against 31 percent preventive and 12 percent predictive. The same DOE guide estimates that a preventive program saves 12 to 18 percent over a purely reactive one, that a functioning predictive program saves a further 8 to 12 percent over preventive alone and 30 to 40 percent or more against a reactive-heavy facility, and it cites unnamed industrial surveys for a tenfold return on investment, 70 to 75 percent fewer breakdowns and 35 to 45 percent less downtime.
Statistic 20
More than 55% of maintenance resources and activities at an average facility are still reactive; 31% preventive, 12% predictive, 2% other.
U.S. Department of Energy, Federal Energy Management Program, operations and maintenance best-practices guide, 2010
Statistic 21
A preventive maintenance program saves an estimated 12% to 18% over a purely reactive program.
U.S. Department of Energy, Federal Energy Management Program, operations and maintenance best-practices guide, 2010
Statistic 22
A properly functioning predictive maintenance program saves 8% to 12% over preventive maintenance alone, and 30% to 40% or more versus a reactive-heavy facility.
U.S. Department of Energy, Federal Energy Management Program, operations and maintenance best-practices guide, 2010
Statistic 23
The DOE FEMP guide puts the industrial-average results of a functional predictive maintenance program (from surveys it cites but does not name) at: 10x return on investment, 25-30% lower maintenance costs, 70-75% fewer breakdowns, 35-45% less downtime, 20-25% more production.
U.S. Department of Energy, Federal Energy Management Program, operations and maintenance best-practices guide, 2010
What this means: These are the most-copied maintenance numbers on the internet and almost nobody links the page they come from. They are a federal-facility guide's summary of older studies, dated 2010, not a measurement; treat them as the planning range a contractor or CMMS vendor is quoting, and check them against the NIST survey figures below, which are newer and sourced to a stated sample.
The typical industrial business: ABB's 3,215-respondent survey
According to ABB's 2023 reliability survey report (abb.com), a survey ABB commissioned from Sapio Research across 3,215 industrial respondents in energy, oil and gas, food and beverage, metals, chemicals and other sectors, unplanned outages cost the typical industrial business a median of $124,669 an hour, with sector medians from $179,056 in energy and power down to $84,681 in food and beverage and $71,047 in metals. The same ABB report found 69 percent of plants suffer an unplanned outage on critical equipment at least monthly and 8 percent every day, that 21 percent still run a run-to-fail strategy, and that condition-based users reported a 42 percent uptime gain against 35 percent for time-based users.
Statistic 24
Unplanned outages cost the typical industrial business a median of $124,669 per hour (rounded to "$125,000").
ABB, 2023 reliability and maintenance survey report (ABB's own commissioned survey, Sapio Research, 3,215 industrial respondents)
Statistic 25
Median downtime cost per hour by sector: energy/power $179,056; plastics/rubber $178,941; oil and gas $144,452; chemical $127,563; wind $126,000; rail $115,000; marine $101,000; water/wastewater $98,222; utilities $96,536; food and beverage $84,681; metals $71,047.
ABB, 2023 reliability and maintenance survey report (ABB's own commissioned survey, Sapio Research, 3,215 industrial respondents)
Statistic 26
Distribution of reported hourly downtime cost: 10% up to $10K; 13% $10K-$50K; 10% $51K-$75K; 10% $76K-$100K; 11% $101K-$150K; 9% $151K-$200K; 8% $201K-$300K; 9% $301K-$500K; 7% $501K-$750K; 6% $751K-$1M; 3% $1M+; 3% don't know.
ABB, 2023 reliability and maintenance survey report (ABB's own commissioned survey, Sapio Research, 3,215 industrial respondents)
Statistic 27
69% of plants experience unplanned outages on critical equipment at least once a month; 8% every day; 14% several times a week; 14% once a week; 19% a few times a month; 13% once a month; 14% every quarter; 8% yearly; 8% less than once a year; 1% never.
ABB, 2023 reliability and maintenance survey report (ABB's own commissioned survey, Sapio Research, 3,215 industrial respondents)
Statistic 28
Among businesses using a run-to-fail strategy, 80% experience unplanned outages at least monthly, versus 69% overall.
ABB, 2023 reliability and maintenance survey report (ABB's own commissioned survey, Sapio Research, 3,215 industrial respondents)
Statistic 29
Maintenance strategy mix: 45% time-based preventive, 33% condition-based (predictive), 21% run-to-fail.
ABB, 2023 reliability and maintenance survey report (ABB's own commissioned survey, Sapio Research, 3,215 industrial respondents)
Statistic 30
92% said maintenance increased their uptime in the last year; 38% by more than a quarter; condition-based users reported a 42% uptime increase versus 35% for time-based users.
ABB, 2023 reliability and maintenance survey report (ABB's own commissioned survey, Sapio Research, 3,215 industrial respondents)
Statistic 31
60% plan to increase reliability and maintenance investment over the next three years; a third by more than 10%; 43% find it difficult to recruit maintenance staff; average age of maintenance staff surveyed is 37.
ABB, 2023 reliability and maintenance survey report (ABB's own commissioned survey, Sapio Research, 3,215 industrial respondents)
What this means: ABB's median is the citable 'typical plant' number because it has a stated sample size and a published distribution: the distribution shows why averages mislead here, with 23 percent of respondents at or under $50,000 an hour and 16 percent above $500,000. The sample is industrial rather than purely manufacturing, so quote it as 'industrial businesses'.
Causes, frequency and strategy mix in four industry surveys
According to Plant Engineering's 2020 maintenance survey (plantengineering.com), 88 percent of industrial facilities follow a preventive maintenance strategy, 51 percent still use run-to-failure, 88 percent outsource some or all maintenance, and aging equipment is the leading cause of unscheduled downtime, followed by mechanical failure, operator error and lack of training. According to MaintainX's 2024 industrial maintenance survey (getmaintainx.com), the CMMS vendor's own survey of 1,165 MRO professionals, the average cost of an hour of unplanned downtime per facility is about $25,000, 57 percent of facilities still use run-to-failure and 72 percent of those whose downtime costs rose blamed parts and shipping prices. According to L2L's 2025 downtime survey release (l2l.com), U.S. facilities average 30 hours of downtime a month, more than half of it unplanned, six in ten spend more than $250,000 a year on it and 67 percent take a reactive approach. According to Fluke's 2025 survey releases (fluke.com), 55 percent of U.S. manufacturers were hit by unplanned downtime in the past year.
Statistic 32
88% of industrial facilities follow a preventive maintenance strategy, 52% have a CMMS, and 51% use a run-to-failure method.
Plant Engineering (WTWH Media), 2020 maintenance survey article, published 2021, sponsored by Advanced Technology Services
Statistic 33
The average facility spends 33 hours a week on scheduled maintenance; 46% allocate up to 10% of annual operating costs to maintenance and 41% more than 10%.
Plant Engineering (WTWH Media), 2020 maintenance survey article, published 2021, sponsored by Advanced Technology Services
Statistic 34
88% of facilities outsource some or all maintenance; the average facility outsources 23% of maintenance operations.
Plant Engineering (WTWH Media), 2020 maintenance survey article, published 2021, sponsored by Advanced Technology Services
Statistic 35
The leading cause of unscheduled downtime is aging equipment, followed by mechanical failure, operator error and lack of proper training; more than half of facilities plan equipment upgrades to reduce it.
Plant Engineering (WTWH Media), 2020 maintenance survey article, published 2021, sponsored by Advanced Technology Services
Statistic 36
The average cost of an hour of unplanned downtime per facility is about $25,000 and can exceed $500,000 for larger organizations.
MaintainX, 2024 industrial maintenance survey report (the CMMS vendor's own survey of 1,165 MRO professionals; 31 percent manufacturing)
Statistic 37
85.2% reported unplanned downtime incidents stabilized or decreased in the past year, 45% reported a decrease, yet 29.4% reported the cost of those incidents rose.
MaintainX, 2024 industrial maintenance survey report (the CMMS vendor's own survey of 1,165 MRO professionals; 31 percent manufacturing)
Statistic 38
Drivers of rising downtime cost: rising cost of parts and shipping 72%; labor cost 48%; wear and tear on critical assets 41%; lack of capacity to make up lost production 24%.
MaintainX, 2024 industrial maintenance survey report (the CMMS vendor's own survey of 1,165 MRO professionals; 31 percent manufacturing)
Statistic 39
What reduced unplanned downtime: evolving maintenance strategy 65%, replacing aging equipment 43%, better training 27%.
MaintainX, 2024 industrial maintenance survey report (the CMMS vendor's own survey of 1,165 MRO professionals; 31 percent manufacturing)
Statistic 40
The anticipated leading cause of unplanned downtime in the next 12 months: aging equipment 29%, equipment failure 22%, operator error 12%, labor shortages 9%. Among facilities whose downtime rose, 65.7% blamed labor shortages and skill gaps.
MaintainX, 2024 industrial maintenance survey report (the CMMS vendor's own survey of 1,165 MRO professionals; 31 percent manufacturing)
Statistic 41
86.8% of facilities run preventive maintenance, 57% still use run-to-failure (5% as their only program), 30.2% use predictive maintenance; 59% spend less than half of maintenance time on planned work.
MaintainX, 2024 industrial maintenance survey report (the CMMS vendor's own survey of 1,165 MRO professionals; 31 percent manufacturing)
Statistic 42
64.4% of facilities allocate 5-20% of annual operating budget to maintenance; about one in five allocate more than 20%.
MaintainX, 2024 industrial maintenance survey report (the CMMS vendor's own survey of 1,165 MRO professionals; 31 percent manufacturing)
Statistic 43
US manufacturing facilities average 30 hours of downtime a month, more than half of it unplanned.
L2L, 2025 manufacturing downtime survey press release (the software vendor's own survey of 600+ U.S. manufacturing leaders)
Statistic 44
Six in ten manufacturers say downtime costs them more than $250,000 a year.
L2L, 2025 manufacturing downtime survey press release (the software vendor's own survey of 600+ U.S. manufacturing leaders)
Statistic 45
67% take a reactive approach to maintenance; 72% say undocumented fixes contribute to recurring downtime; 73% say downtime hurts product quality; 52% say it prevents meeting production or shipping targets.
L2L, 2025 manufacturing downtime survey press release (the software vendor's own survey of 600+ U.S. manufacturing leaders)
Statistic 46
93% of manufacturers are taking action on downtime but only 46% report measurable improvement.
L2L, 2025 manufacturing downtime survey press release (the software vendor's own survey of 600+ U.S. manufacturing leaders)
Statistic 47
55% of US manufacturers were hit by unplanned downtime in the past year (61% across the three-country sample).
Fluke Corporation, 2025 downtime survey press releases (the vendor's own Censuswide survey of 600+ decision-makers in the U.S., U.K. and Germany)
Statistic 48
Among affected manufacturers, 48% report 6-10 downtime incidents a week and 19% report 11-20; 45% say outages last up to 12 hours and 15% up to 72 hours.
Fluke Corporation, 2025 downtime survey press releases (the vendor's own Censuswide survey of 600+ decision-makers in the U.S., U.K. and Germany)
What this means: Every survey in this group was run by a vendor or a sponsored trade magazine and is primary only for its own findings; the samples, definitions of 'incident' and industries differ, which is why Fluke's 6 to 10 incidents a week and Siemens' 25 a month are not put on the same chart. What they agree on: half or more of facilities still use run-to-failure in every survey that asked whether it is used at all (2020, 2024, 2025), aging equipment leads the cause list, and the reactive share has not moved much in 15 years.
What maintenance strategy is worth in dollars: NIST, Deloitte, McKinsey and DOE assessments
According to NIST researchers Thomas and Weiss, in a 2021 peer-reviewed paper (open-access manuscript on PMC, nih.gov), U.S. manufacturing's annual maintenance costs and losses total an estimated $222.0 billion in 2016 dollars, of which unplanned downtime accounts for about $18.4 billion, and manufacturers that relied less on reactive maintenance had 52.7 percent less unplanned downtime and 78.5 percent fewer defects while spending 81.7 percent more on direct maintenance. According to Deloitte's 2023 predictive maintenance infographic (deloitte.com), well-executed predictive maintenance cuts facility downtime 5 to 20 percent. According to McKinsey's 2018 chemicals analytics article (mckinsey.com), predictive maintenance typically reduces machine downtime 30 to 50 percent. According to the Department of Energy's Industrial Assessment Centers database (itac.university), 22,916 no-cost assessments of small and medium manufacturers have produced recommendations averaging $144,000 a year in savings per plant, 49.9 percent of which (of those with a known status) were implemented and 41.8 percent of which (of those with sector codes) pay back in under six months; these are energy and productivity recommendations, not maintenance-only.
Statistic 49
NIST estimates total annual US manufacturing costs and losses associated with maintenance at $222.0 billion (Monte Carlo average), of which unplanned downtime accounts for about $18.4 billion.
Thomas and Weiss, NIST, peer-reviewed maintenance cost survey and analysis, International Journal of Prognostics and Health Management, 2021 (open-access manuscript at PMC)
Statistic 50
Manufacturers relying less on reactive maintenance had 52.7% less unplanned downtime, 78.5% fewer defects and 51.8% lower fault/failure costs relative to shipments, while spending 81.7% more on direct maintenance.
Thomas and Weiss, NIST, peer-reviewed maintenance cost survey and analysis, International Journal of Prognostics and Health Management, 2021 (open-access manuscript at PMC)
Statistic 51
Among proactive maintainers, those leaning predictive rather than preventive had 18.5% less unplanned downtime and 87.3% fewer defects.
Thomas and Weiss, NIST, peer-reviewed maintenance cost survey and analysis, International Journal of Prognostics and Health Management, 2021 (open-access manuscript at PMC)
Statistic 52
Deloitte reports well-executed predictive maintenance delivers a 5-20% reduction in facility downtime, 10-30% reduction in inventory levels, 5-20% increase in labor productivity, and 3-5% reduction in new equipment costs.
Deloitte, smart manufacturing predictive maintenance infographic, 2023 (citing Deloitte's 2022 predictive maintenance analysis)
Statistic 53
McKinsey: predictive maintenance typically reduces machine downtime by 30 to 50 percent and increases machine life by 20 to 40 percent.
McKinsey and Company, chemicals advanced-analytics article, February 2018 (McKinsey's own estimate, chemicals context)
Statistic 54
DOE Industrial Assessment Centers have completed 22,916 no-cost assessments of small and medium manufacturers, producing 170,024 recommendations (7.4 per assessment) with average recommended savings of $144,000 per plant per year.
U.S. Department of Energy, Industrial Assessment Centers database, live statistics page, data through 2024
Statistic 55
Of IAC recommendations with known implementation status, 49.9% were implemented (77,088 of 154,627); implemented recommendations average 3.5 per plant and $49,000 a year in savings.
U.S. Department of Energy, Industrial Assessment Centers database, implementation-rates and statistics pages
Statistic 56
41.8% of IAC recommendations pay back in under six months and 56.5% within a year; only 15.6% take more than three years.
U.S. Department of Energy, Industrial Assessment Centers database, payback page (90,883 recommendations with NAICS codes)
What this means: NIST's $18.4 billion is the defensible government answer to 'what does downtime cost U.S. manufacturing'; it is less than half the '$50 billion' that Deloitte's own infographic repeats from a sponsored trade piece. The consulting ranges (Deloitte 5 to 20 percent, McKinsey 30 to 50 percent) are unsourced assertions inside larger documents; the IAC database is the one place a plant can see real, implemented recommendations with a payback column.
Injuries, citations and what an inspection checks
According to the Bureau of Labor Statistics' 2024 Survey of Occupational Injuries and Illnesses (bls.gov), manufacturing recorded 2.7 total recordable cases per 100 full-time workers against 2.3 for all private industry, with motor vehicle manufacturing at 5.6 and food manufacturing at 3.3; manufacturing had 326,400 nonfatal injuries in 2023. According to OSHA's final FY2025 citation data as published by Safety+Health magazine (safetyandhealthmagazine.com), lockout/tagout under 29 CFR 1910.147 was the fourth most-cited standard with 2,562 violations and machine guarding under 29 CFR 1910.212 was tenth with 1,498. According to the regulation text as published by Cornell's Legal Information Institute (cornell.edu), 29 CFR 1910.147 covers servicing and maintenance where unexpected energization or release of stored energy could injure employees, and 29 CFR 1910.212 requires guarding of the point of operation, ingoing nip points, rotating parts and flying chips and sparks.
Statistic 57
Manufacturing recorded 2.7 total recordable injury and illness cases per 100 full-time workers in 2024, versus 2.3 for all private industry; food manufacturing 3.3, fabricated metal products 3.2, transportation equipment 3.2, motor vehicle manufacturing 5.6, chemicals 1.6.
U.S. Bureau of Labor Statistics, Survey of Occupational Injuries and Illnesses, Table 1, 2024 data, released 2025
Statistic 58
Private industry reported 2.5 million nonfatal injuries and illnesses in 2024 (2,488,400: 2,340,400 injuries, 148,000 illnesses), a TRC rate of 2.3 per 100 FTE.
U.S. Bureau of Labor Statistics, The Economics Daily, 2026, and the 2024 SOII news release
Statistic 59
Manufacturing had 326,400 nonfatal injuries in 2023 (rate 2.6 per 100 FTE), down from 347,800 (2.8) in 2022.
U.S. Bureau of Labor Statistics, SOII news release, Table 4 (injuries by industry, 2022 to 2023)
Statistic 60
Lockout/Tagout (29 CFR 1910.147) was OSHA's 4th most-cited standard in FY2025 with 2,562 violations; Machine Guarding (1910.212) was 10th with 1,498.
OSHA final FY2025 top-10 citation data (October 1, 2024 to September 30, 2025), as published by Safety+Health magazine (National Safety Council), April 2026; ranking confirmed on OSHA's own top-10 page
Statistic 61
OSHA's lockout/tagout standard applies to "servicing and maintenance of machines and equipment in which the unexpected energization or start up of the machines or equipment, or release of stored energy could cause injury to employees."
29 CFR 1910.147(a)(1)(i), as published by the Legal Information Institute, Cornell Law School
Statistic 62
OSHA's machine guarding rule requires guarding for "point of operation, ingoing nip points, rotating parts, flying chips and sparks" and lists guillotine cutters, shears, power presses, milling machines, power saws, jointers, portable power tools, and forming rolls as machines that usually require point-of-operation guarding.
29 CFR 1910.212(a)(1) and (a)(3)(iv), as published by the Legal Information Institute, Cornell Law School
Statistic 63
OSHA's lockout/tagout standard requires the employer to inspect each energy control procedure at least once a year, using an authorized employee who is not the one carrying out that procedure, to correct any deviations found, and to certify each inspection with the machine, the date, the employees covered and the inspector.
29 CFR 1910.147(c)(6), as published by the Legal Information Institute, Cornell Law School
What this means: The two OSHA standards most tied to maintenance work, lockout/tagout and machine guarding, together drew more than 4,000 citations in one fiscal year; both are inspected by looking at the machine and the procedure, which is the same evidence a downtime investigation needs.
Equipment maintenance as an FDA inspection finding
According to the FDA's inspection observations spreadsheets (fda.gov), the agency issued 4,862 system-generated Form 483s in FY2025, up from 4,056 in FY2024. In drug manufacturing, equipment cleaning, sanitizing and maintenance under 21 CFR 211.67(a) was the fifth most-cited observation with 95 citations, 21 CFR 211.67(b) drew 70 for missing written procedures and 21 CFR 211.68(a) drew 38 for calibration and inspection not done to a written program. In food, failure to keep the plant in adequate repair under 21 CFR 117.35(a) was the fourth most-cited observation with 211 citations. In devices, 21 CFR 820.72(a), lacking procedures to routinely calibrate, inspect and maintain equipment, drew 30 observations in FY2025.
Statistic 64
FDA issued 4,862 system-generated Form 483s in FY2025 (713 drugs, 791 devices, 2,719 foods), up from 4,056 in FY2024 (561 drugs, 672 devices, 2,307 foods).
U.S. FDA, inspection observations spreadsheets, FY2025 and FY2024 summary tabs
Statistic 65
Equipment cleaning, sanitizing and maintenance (21 CFR 211.67(a)) was the 5th most-cited drug GMP observation in FY2025 with 95 citations; 211.67(b) (no written cleaning/maintenance procedures) drew 70 and 211.63 (equipment design/size/location) 81.
U.S. FDA, inspection observations spreadsheet, FY2025, drugs tab
Statistic 66
Routine calibration/inspection/checking of equipment not performed per a written program (21 CFR 211.68(a)) drew 38 drug observations in FY2025 (30 in FY2024).
U.S. FDA, inspection observations spreadsheets, FY2025 and FY2024, drugs tab
Statistic 67
"Plant not maintained in a clean and sanitary condition or in adequate repair" (21 CFR 117.35(a)) was the 4th most-cited food observation in FY2025 with 211 citations; equipment and utensils not designed or maintained to be adequately cleaned (117.40) drew 120.
U.S. FDA, inspection observations spreadsheet, FY2025, foods tab
Statistic 68
For medical devices, lack of procedures to ensure equipment is routinely calibrated, inspected, checked and maintained (21 CFR 820.72(a)) drew 37 observations in FY2024 and 30 in FY2025.
U.S. FDA, inspection observations spreadsheets, FY2024 and FY2025, devices tab
What this means: For a regulated manufacturer, an unmaintained machine is not only a downtime risk but a citable inspection finding with its own CFR section; the counts are read from the citation rows of each program-area tab in the agency's own spreadsheets, and FDA notes the files omit manually prepared 483s.
The numbers that rank for this query and could not be verified
These figures are widely repeated online. Each was traced as far as the trail went and is listed here so a writer does not re-chase it.
Unverified claim
$50 billion a year cost of unplanned downtime. Deloitte's infographic sources it to a sponsored trade piece, which in turn traces to an unpublished analyst estimate; NIST's $18.4 billion (unplanned downtime) and $222 billion (all maintenance costs and losses) are the traceable replacements.
Unverified claim
$260,000 or $540,000 per hour of downtime. Attributed to 2016 and 2017 analyst reports for software vendors; every live copy is a vendor restatement and the originals are gated or gone.
Unverified claim
82 percent of companies had unplanned downtime in three years, $2 million per incident. Same 2017 vendor-commissioned trail; Fluke 2025 (55 to 61 percent in the past year) and ABB 2023 (69 percent monthly) are the live replacements.
Unverified claim
Equipment failure accounts for 80 percent of unplanned downtime. Repeated across vendor roundups with no primary; the nearest sourced figures are Plant Engineering's cause ranking (no percentages) and MaintainX's 22 percent naming equipment failure as the anticipated leading cause.
Unverified claim
Aging equipment 34 percent, mechanical failure 20 percent, operator error 11 percent. Attributed to Plant Engineering's 2020 survey, but the open article gives only the ranking; the percentages sit inside a registration-gated PDF.
Unverified claim
Nearly 80 percent of manufacturers cannot calculate the cost of downtime; $864 billion or 8 percent of revenue. Usually attributed to Siemens' 2022 edition; neither figure is in the 2024 report, and the 2022 PDF was not opened.
Unverified claim
Unplanned downtime costs 15 times planned downtime; up to 800 hours a year; $3 million per incident. Roundup numbers with no primary anywhere; Siemens' 326 hours a year per large plant is the sourced replacement for the hours claim.
Unverified claim
Fluke's $1.7 million per hour and $852 million per week. In Fluke's own press releases, but with no published method, and the U.S. release gives both $400,000 and $1.7 million per hour for the same sample; only Fluke's percentages are used here.
Cite this study
Academic or press use: copy a ready-made reference. RapidEye is the publisher.
Quick FAQ
How much does unplanned downtime cost a manufacturer per hour?
It depends entirely on the plant. Siemens' 2024 report puts an hour at $2.3 million in a large automotive plant and $36,000 at the low end in fast-moving consumer goods, from its survey of large industrial organizations. ABB's 2023 survey of 3,215 industrial businesses found a median of $124,669 an hour, with sector medians from $179,056 in energy and power to $71,047 in metals. MaintainX's 2024 survey of 1,165 facilities of all sizes puts the average at about $25,000 an hour, rising above $500,000 for larger organizations.
How many hours of downtime does a manufacturing plant have?
Siemens' 2024 report says a large plant loses an average of 27 hours a month, 326 hours a year, to unplanned downtime, across 25 incidents a month. L2L's 2025 survey of U.S. manufacturers found 30 hours of downtime a month, more than half of it unplanned. ABB's 2023 survey found 69 percent of plants have an unplanned outage on critical equipment at least monthly and 8 percent every day.
What does downtime cost U.S. manufacturing as a whole?
The most defensible figure is NIST's: about $18.4 billion a year in unplanned downtime losses, inside $222.0 billion of total maintenance costs and losses, in 2016 dollars (Thomas and Weiss, 2021). The '$50 billion a year' figure repeated on vendor pages traces to a sponsored trade piece and an unpublished analyst estimate, and could not be verified.
What causes most unplanned downtime in manufacturing?
Aging equipment leads both surveys that asked. Plant Engineering's 2020 survey ranked aging equipment first, followed by mechanical failure, operator error and lack of training. MaintainX's 2024 survey found 29 percent expect aging equipment to be the leading cause in the next year, 22 percent equipment failure, 12 percent operator error and 9 percent labor shortages; among facilities whose downtime rose, 65.7 percent blamed labor shortages and skill gaps. The '80 percent of downtime is equipment failure' claim has no primary source.
How much does preventive or predictive maintenance reduce downtime?
The Department of Energy's FEMP guide estimates preventive maintenance saves 12 to 18 percent over a purely reactive program and predictive maintenance a further 8 to 12 percent, or 30 to 40 percent or more against a reactive-heavy facility, with 35 to 45 percent less downtime. NIST's 2021 survey found manufacturers relying less on reactive maintenance had 52.7 percent less unplanned downtime. McKinsey's estimate is 30 to 50 percent less machine downtime; Deloitte's is 5 to 20 percent less facility downtime.
Data sources
Every figure on this page traces to one of these publishers' own documents, each checked against the original before publishing. Sources are named at the publisher level and shown by root domain.
U.S. Department of Energy, Federal Energy Management ProgramFederal operations and maintenance best-practices guide (2010)energy.gov
MaintainX2024 industrial maintenance survey report (1,165 respondents)getmaintainx.com
National Library of Medicine (PMC)Open-access manuscript of the NIST maintenance cost survey and analysis, 2021nih.gov
DeloittePredictive maintenance benefits infographic (2023)deloitte.com
U.S. Bureau of Labor StatisticsSurvey of Occupational Injuries and Illnesses, 2024 tables, news release and The Economics Dailybls.gov
Occupational Safety and Health AdministrationTop 10 most frequently cited standards pageosha.gov
Safety+Health magazine (National Safety Council)OSHA final FY2025 top-10 cited standards with counts (April 2026)safetyandhealthmagazine.com
Legal Information Institute, Cornell Law School29 CFR 1910.147 and 29 CFR 1910.212 as publishedcornell.edu
U.S. Food and Drug AdministrationInspection observations spreadsheets, FY2024 and FY2025fda.gov