Preventive vs Reactive Maintenance: The Cost Difference (2026)
What the U.S. Department of Energy’s figures say about the cost of reactive, preventive and predictive maintenance, with worked examples and when run-to-failure is the right call.
Key Takeaways
- DOE estimates preventive saves 12% to 18% over reactive. Its guide adds that facilities purely reliant on reactive maintenance could save much more than 18% with a proper PM program.
- Predictive adds another 8% to 12% over preventive. DOE says a facility moving from heavy reactive reliance to predictive could see savings opportunities above 30% to 40%.
- Cost per horsepower shows the gap. A comparison DOE reproduces gives about 18 dollars per hp per year for reactive, 13 for preventive, 9 for predictive and 6 for reliability centered maintenance.
- These figures are old. They come from studies cited in a 2010 guide. Use them to frame the argument, then measure your own costs.
- Run-to-failure is right for some assets. Cheap, non-critical, redundant or rarely failing equipment can be left to reactive maintenance by design.
Quick-Start Prioritization Framework
| Approach | What triggers work | DOE cost estimate | Best for |
|---|---|---|---|
| Reactive (run to failure) | Breakdown | About $18/hp/yr | Cheap, non-critical or redundant items |
| Preventive (time-based) | Calendar or run hours | About $13/hp/yr; 12% to 18% savings over reactive | Equipment subject to wear with known failure patterns |
| Predictive (condition-based) | Measured condition | About $9/hp/yr; 8% to 12% savings over preventive | Critical equipment with random failure patterns |
| Reliability centered | Mix chosen per asset, with root cause analysis | About $6/hp/yr | Mature programs |
Every maintenance manager has made the argument to finance: if we spent a little more on scheduled maintenance, we would spend a lot less on breakdowns. Finance usually asks for numbers. The most widely cited public source is the U.S. Department of Energy’s Federal Energy Management Program O&M Best Practices Guide, which pulls together estimates for reactive, preventive, predictive and reliability centered programs. This guide explains what those figures are, how old they are, how to apply them to your own site with honest arithmetic, and where reactive maintenance is still the cheaper choice.
Definitions, briefly
DOE’s O&M Best Practices Guide (Release 3.0, 2010) defines the approaches clearly, and it is worth using its language because it is the source of the numbers.
Reactive maintenance
DOE calls it the “run it till it breaks” mode: no actions are taken to maintain the equipment as the designer intended. You repair or replace when something fails.
Preventive maintenance
Actions performed on a time- or machine-run-based schedule that detect, preclude or mitigate degradation of a component, to sustain or extend its useful life. Filter changes, lubrication, belt adjustments and annual inspections are the classic examples.
Predictive and reliability centered maintenance
Predictive maintenance bases work on measured condition (vibration, temperature, oil analysis) rather than a preset schedule. Reliability centered maintenance (RCM) chooses the right mix of reactive, preventive and predictive for each asset based on its criticality and failure modes.
The numbers DOE gives
DOE’s guide collects estimates from several studies. Here they are, with their context.
| Figure | What DOE’s guide says | Context |
|---|---|---|
| Preventive vs reactive | Estimated 12% to 18% cost savings | Could be much more than 18% for facilities purely reliant on reactive work |
| Predictive vs preventive | Estimated 8% to 12% cost savings | Savings opportunities could exceed 30% to 40% depending on current reliance on reactive work |
| Cost per hp per year | Reactive $18, preventive $13, predictive $9, RCM $6 | From a comparison by Piotrowski (2001) |
| Average facility mix | More than 55% reactive, 31% preventive, 12% predictive, 2% other | Studies as recent as winter 2000 |
| Top performers’ mix | Less than 10% reactive, 25% to 35% preventive, 45% to 55% predictive | Described as continually top-performing facilities |
How old these are
The guide was published in 2010 and the underlying studies date from the 1990s and early 2000s. Dollar figures per horsepower are not adjusted for inflation. The ratios between approaches are the useful part; the absolute dollars are not a quote for your building.
Predictive program survey figures
DOE also cites independent surveys of industrial sites starting a functional predictive program, with average results of a 10 times return on investment, 25% to 30% lower maintenance costs, 70% to 75% fewer breakdowns, 35% to 45% less downtime and 20% to 25% more production. These are industrial averages from surveys, not guarantees, and the guide notes that predictive equipment can cost more than 50,000 dollars to start.
Why reactive maintenance costs more
DOE’s guide is candid that reactive maintenance looks cheap. You spend nothing until something breaks, so the budget appears to be saving money. Its list of hidden costs is a good checklist to bring to a budget meeting.
Shorter equipment life
Equipment run without the maintenance its designer intended wears out sooner, so you replace it more often. That is a capital cost that never shows up on the maintenance line.
Bigger repairs and secondary damage
A failure usually needs more extensive repair than the maintenance that would have prevented it, and it can take other equipment with it. A seized bearing can wreck a shaft; a burst coil can flood a ceiling.
Overtime and bad timing
DOE notes failures tend to happen off hours or near the end of the day. Critical equipment then needs overtime, rush parts and premium contractor rates.
Bigger parts inventory
If you expect to run to failure, you need spares on the shelf for everything that might fail. Planned work lets you order parts ahead and carry less.
Working the numbers for your site
Here is how to turn DOE’s figures into an estimate you can defend. The arithmetic below is illustrative; replace every input with your own.
Example 1: percentage method
Suppose last year’s maintenance spend (labor, parts and contractors) was 400,000 dollars and you estimate 70% of it was reactive. Apply DOE’s 12% to 18% preventive savings to the reactive share only: 280,000 times 12% is 33,600 dollars; times 18% is 50,400 dollars. A reasonable claim is that a well-run PM program could save in the range of 30,000 to 50,000 dollars a year, before counting longer equipment life.
Example 2: horsepower method
For a plant with rotating equipment, total the motor horsepower. At 1,500 hp, DOE’s per-hp figures give 27,000 dollars a year reactive (1,500 times 18) versus 19,500 preventive (1,500 times 13), a difference of 7,500 dollars, and 13,500 predictive. These are in old dollars, so present the ratio (preventive costs about 28% less per hp than reactive in this comparison) rather than the totals.
Example 3: from your own records
The strongest case uses your data. Pull 12 months of work orders. For the 10 assets with the most reactive cost, total labor, parts, contractor and overtime. Then estimate the PM labor and parts that would have been needed. If the top 10 cost 60,000 dollars in breakdowns and their PM would cost 15,000, the case makes itself, and finance can check every line.
When reactive maintenance is the right choice
DOE’s guide does not say reactive is always wrong. Its RCM table lists where a reactive element fits: small parts and equipment, non-critical equipment, equipment unlikely to fail, and redundant systems.
Examples
Light fixtures in a corridor with plenty of others, a small exhaust fan in a storage closet, a duplex pump set where the standby takes over automatically, or a commodity part that costs less than the labor to inspect it. Planned replacement on failure is cheaper than inspection for these.
Make it a decision, not a default
Mark these assets as run-to-failure in your asset register, keep a spare where downtime matters, and review the decision if an asset starts failing often. That is reliability centered thinking at small scale.
Preventive maintenance has costs too
DOE lists the disadvantages of preventive maintenance honestly: it is labor intensive, it includes some unneeded maintenance, it can cause incidental damage when parts are disturbed unnecessarily, and catastrophic failures can still occur.
That is the argument for tuning intervals with history and, on critical equipment, moving toward condition-based tasks. Measuring filter differential pressure and changing filters when they reach the limit, rather than every three months regardless, is a simple first step. DOE’s guide uses exactly this example when describing how building systems can trigger work orders based on need.
How to shift the mix
Moving from mostly reactive to mostly planned does not happen in a budget cycle. A practical sequence follows DOE’s steps for starting RCM.
- Build a master equipment list.
- Rank each asset by criticality to safety and operations.
- Put critical assets on manufacturer-based PM first and measure PM completion monthly.
- Track reactive work by asset and find the repeat offenders.
- Add condition checks (readings on PM checklists) to the most critical equipment.
- Report emergency hours as a share of total hours; DOE’s benchmark table lists below 10%.
Common mistakes
These weaken the case or the program.
- Quoting DOE’s percentages as a guaranteed saving instead of an estimate from published studies.
- Presenting 2001 dollars-per-horsepower as current costs.
- Applying PM to everything, including assets that should run to failure.
- Cutting PM first whenever the reactive backlog grows, which feeds the backlog.
- Not tracking reactive cost by asset, so you cannot show where PM would pay.
- Ignoring the capital side: shorter equipment life is the biggest cost of neglect and the easiest to miss.
Tracking the difference
To show the shift is working, you need work orders tagged as reactive or planned, labor and parts per work order, and PM completion by month. A CMMS does this out of the box.
If you’d rather not adopt a packaged CMMS, Floot, an AI app builder, can build a maintenance app from a plain-English description that tags every work order as planned or reactive and reports cost by asset. It is free to start.
Frequently Asked Questions
How much does preventive maintenance save compared with reactive?
DOE’s O&M Best Practices Guide estimates 12% to 18% on average, and says facilities purely reliant on reactive maintenance could save much more than 18% with a proper preventive program.
Is reactive maintenance ever cheaper?
Yes, for small, non-critical, redundant or rarely failing equipment. DOE’s guide lists these as suitable for a reactive approach.
What is a good ratio of planned to reactive maintenance?
DOE describes top-performing facilities as less than 10% reactive, with 25% to 35% preventive and 45% to 55% predictive. Its benchmark table lists emergency maintenance below 10% of hours.
How much more does reactive maintenance cost per horsepower?
A comparison reproduced in DOE’s guide gives about 18 dollars per hp per year for reactive versus 13 for preventive and 9 for predictive. The figures date from 2001.
What is the difference between preventive and predictive maintenance?
Preventive maintenance is scheduled by time or run hours. Predictive maintenance is triggered by measured condition, such as vibration or oil analysis. DOE estimates predictive saves 8% to 12% over preventive.
How do I convince management to fund preventive maintenance?
Use your own work order history: total the breakdown cost of your worst assets, estimate their PM cost, and support it with DOE’s published ranges. Present simple payback and return on investment, as DOE’s guide recommends.
Sources
- U.S. Department of Energy, FEMP, Operations & Maintenance Best Practices: A Guide to Achieving Operational Efficiency, Release 3.0, 2010 https://www.energy.gov/sites/default/files/2020/04/f74/omguide_complete_w-eo-disclaimer.pdf
- U.S. Department of Energy, FEMP, O&M Best Practices Guide overview page, 2010 https://www.energy.gov/eere/femp/articles/operations-and-maintenance-best-practices-guide-achieving-operational-efficiency
- IBM, What is a CMMS?, 2026 https://www.ibm.com/think/topics/what-is-a-cmms