How to Set Up a Preventive Maintenance Schedule (2026)
A step-by-step method for building a preventive maintenance schedule: list your equipment, rank it, pick tasks and intervals, balance the workload and keep it honest.
Key Takeaways
- Start with a master equipment list. You cannot schedule maintenance on equipment you have not written down. The list comes first, then everything else hangs off it.
- Rank before you schedule. Critical, safety-related and expensive-to-fail equipment gets PM first. Small, cheap or redundant items can reasonably be left to run to failure.
- Take intervals from the manufacturer, then adjust with history. DOE’s guide defers to manufacturer recommendations first, then machinery history, root cause findings and engineering judgment.
- Balance the hours, not just the dates. Add up the labor each month. A schedule that puts 60 hours of PM in the first week of every month will be skipped.
- Measure PM completion every month. DOE cites a benchmark of more than 90% of scheduled PM actions completed. If you fall well below it, fix the schedule before adding to it.
Quick-Start Prioritization Framework
| Step | What you produce | Typical time for one building |
|---|---|---|
| 1. List equipment | Master equipment list with location and nameplate data | 1 to 3 days of walk-downs |
| 2. Rank criticality | Each asset marked critical, important or run-to-failure | Half a day with the team |
| 3. Pick tasks and intervals | Task list per asset with frequency and trigger | 1 to 2 days with manuals |
| 4. Estimate labor | Hours per task and per month | A few hours |
| 5. Spread the load | Calendar with even monthly hours | A few hours |
| 6. Launch and measure | Work orders created ahead of due dates; monthly PM completion figure | Ongoing |
Most maintenance departments already do some preventive maintenance. The filters get changed, the generator gets run, the fire extinguishers get tagged. What they often lack is a schedule: one list that says which task happens on which asset, how often, who does it, and when it is late. Without that list, PM lives in people’s heads and quietly slips whenever the reactive work piles up. This guide walks through building a schedule from nothing, using the method the U.S. Department of Energy’s Federal Energy Management Program lays out in its O&M Best Practices Guide, plus the practical details that decide whether a schedule survives its first busy month.
What a preventive maintenance schedule actually is
DOE’s O&M Best Practices Guide defines preventive maintenance as actions performed on a time- or machine-run-based schedule that detect, preclude or mitigate degradation of a component, with the aim of sustaining or extending its useful life. The schedule is the document that turns that definition into work on a Tuesday morning.
In practice, a PM schedule is a table. Each row is one task on one asset (or one group of identical assets) with a trigger, a checklist, an estimated duration and an owner. The trigger is either a calendar interval (every month, every year) or a usage meter (every 250 run hours, every 10,000 cycles).
Why bother writing it down
DOE’s guide estimates that a preventive program saves 12% to 18% over a purely reactive one, and says facilities relying entirely on reactive maintenance could save more than that. Those savings come from equipment that lasts closer to its design life, runs more efficiently and fails less often. None of it happens if the PM only gets done when someone remembers.
What it is not
A PM schedule is not a promise to service everything. DOE is explicit that preventive maintenance includes some unneeded work and that catastrophic failures can still happen. The goal is to put scheduled effort where it pays and consciously leave the rest alone.
Step 1: Build the master equipment list
DOE’s steps for moving away from reactive maintenance (adapted from NASA) start the same way: develop a master equipment list identifying the equipment in your facility. If you already have an asset register, check it against reality before trusting it.
Walk the building
Go room by room with a phone camera. Photograph every nameplate. Mechanical rooms, roofs, ceiling spaces above corridors, electrical rooms and the parking structure are where the forgotten equipment lives: exhaust fans, unit heaters, sump pumps, backflow preventers.
Capture the fields you will need for PM
For each asset: a unique ID, description, location, make, model, serial number, install date if known, and the equipment it serves. Add a field for the meter if it has one (run hours on a generator, cycles on a compressor). Our guide to building an asset register for maintenance goes deeper on naming and hierarchy.
Collect the manuals
Find the O&M manual for each model. Many are on the manufacturer’s website by model number. Store the PDF against the asset so the technician can open it on site.
Step 2: Rank equipment by criticality
DOE’s second step is to prioritize the listed components based on importance or criticality to operation, process or mission. A simple three-tier ranking is enough for most buildings.
| Tier | What qualifies | Maintenance approach |
|---|---|---|
| Critical | Life safety, regulatory, or failure stops the building or process (fire pumps, generators, main air handlers, elevators, boilers in winter) | Full PM program; consider condition monitoring |
| Important | Failure causes discomfort, complaints or moderate cost (zone units, domestic water pumps, exhaust fans) | Time-based PM at manufacturer intervals |
| Run to failure | Cheap, non-critical, redundant or unlikely to fail (small parts, a spare pump in a duplex set, light fixtures) | No PM; keep spares; fix on failure |
Run-to-failure is a legitimate choice
DOE’s reliability centered maintenance table lists small parts and equipment, non-critical equipment, equipment unlikely to fail and redundant systems as suitable for a reactive approach. Writing that decision down is what separates a strategy from neglect.
Ask what failure costs, not what the asset costs
A 400 dollar condensate pump that floods a server room is critical. A 20,000 dollar rooftop unit serving a storage area might only be important. Rank by consequence.
Step 3: Choose tasks and intervals
DOE’s guide says to determine the type, number and periodicity of maintenance activities using manufacturer technical manuals, machinery history, root cause analysis findings and good engineering judgment. Its own equipment checklists are presented as a supplement to the manufacturer’s recommendations, not a replacement.
An example: air handling units
DOE’s air handling system checklist (drawn from Better Bricks) gives a useful pattern. Daily: an overall visual inspection that equipment is running and safety systems are in place. Monthly: check filter condition, inspect for leakage at connections and access doors, and cycle dampers fully open to fully closed to check actuators and linkage. Annually: inspect the filter rack, pressure differential gauge and tubing, and inspect and clean coils. That becomes three PM tasks on each air handler: a daily round item, a monthly PM and an annual PM.
Calendar or meter
Use a calendar trigger when wear tracks time or when the requirement is regulatory (annual inspections). Use a meter trigger when wear tracks use: a standby generator that runs 30 hours a year and a prime generator that runs 3,000 hours a year should not share an oil change interval.
Fixed or floating
A fixed schedule counts the next due date from the last due date. A floating schedule counts from the last completion. Compliance inspections should be fixed, so a late inspection does not push every future one later. Wear items such as belts and filters can float.
Write the checklist as steps with readings
“Service AHU-3” is not a task. “Record filter differential pressure in inches of water column; replace filters if above the manufacturer’s limit; check belt tension; cycle outside air damper; note any abnormal noise” is. Numeric readings entered as numbers let you trend them later.
Step 4: Estimate labor and check it against your staff
DOE’s list includes a step to assess the size of the maintenance staff. This is where many schedules fail: they are written as if the team had nothing else to do.
Add up the hours
Give every task an estimated duration, then total by month. A worked example: 24 air handlers with a 45-minute monthly PM is 18 hours. Add 6 boilers and chillers at 2 hours monthly (12 hours), 40 exhaust fans quarterly at 20 minutes (about 13 hours in each quarter month) and annual coil cleaning on the 24 air handlers at 3 hours each (72 hours, which you should spread across the shoulder seasons). Without spreading, the busiest month lands well over 100 hours.
Compare to available hours
Two technicians at about 140 productive hours each a month gives 280 hours. If reactive work historically takes 180 of them, you have roughly 100 hours for PM. If the schedule needs 150, either cut low-value tasks, extend intervals where the manufacturer allows, contract some work out, or make the case for another technician with these numbers.
Let operators do the simple checks
DOE suggests identifying tasks that operations staff can perform. Daily visual rounds, reading gauges and reporting unusual noise are good candidates, freeing technicians for skilled work.
Step 5: Spread the load across the calendar
Once you know the hours, place the tasks so each week carries a similar amount.
Stagger identical assets
Instead of servicing all 24 air handlers in the first week, split them into four groups of six, one group a week. The monthly interval is the same; the workload is flat.
Respect the seasons
Do cooling-tower and chiller work before cooling season and boiler work before heating season. Put annual roof and drain inspections before the rainy months. Avoid scheduling big PMs during known busy periods such as move-in week in student housing.
Group by route
DOE’s step 3 is to assign components into logical groupings. Group tasks by location so a technician going to the roof does all the roof work in one trip.
Step 6: Launch, generate work orders and measure
A schedule only works when it produces work orders before they are due, assigned to a person, with the checklist attached.
Create work orders ahead of the due date
Set a lead time per task, for example 7 days for a monthly PM and 30 days for an annual one that needs a contractor or a shutdown. The technician then has a window rather than a single day.
Track PM completion
DOE’s table of industry O&M benchmarks (from NASA) gives preventive maintenance completion percentage as PM actions completed divided by PM actions scheduled, with a benchmark above 90%. If you scheduled 120 PMs in October and completed 102 on time, that is 85%. Look at which ones slipped and why before the next month.
Review intervals quarterly
If an asset still fails between PMs, shorten the interval or add a task that catches that failure mode. If three years of PM show nothing found, consider extending it where the manufacturer allows. The schedule is a living document.
Common mistakes
These are the ones that come up again and again when a new schedule stalls.
- Trying to schedule every asset on day one. Start with the critical tier and add the rest over a few months.
- Copying a generic template without checking the manufacturer’s manual for the actual model.
- Putting all monthly tasks on the 1st of the month.
- Vague checklists with no readings, so nothing can be trended.
- Letting PM be the first thing cut when reactive work piles up, without recording that it was deferred.
- Floating intervals on compliance inspections, so a late inspection pushes every future one later.
- Never retiring PMs for equipment that has been removed.
Keeping the schedule in a system
A spreadsheet can hold a first schedule for a small building. It struggles once you need work orders generated automatically, checklists filled in on a phone and a monthly completion figure without manual counting. That is the point where most teams move to a CMMS.
If you’d rather not fit your process into an off-the-shelf tool, Floot, an AI app builder, can build a maintenance app from a plain-English description, with its own database and scheduled jobs that create PM work orders ahead of their due dates. It is free to start.
Frequently Asked Questions
How often should preventive maintenance be done?
As often as the manufacturer recommends for that model, adjusted by your own failure history. DOE’s O&M guide defers to the manufacturer first. Common patterns are daily visual checks, monthly filter and belt checks, and annual deep inspections and cleaning.
What is a good PM completion rate?
DOE’s O&M Best Practices Guide lists a benchmark of more than 90% for preventive maintenance actions completed versus scheduled. Measure it monthly.
Should every piece of equipment have a PM?
No. DOE’s guide lists small parts, non-critical equipment, equipment unlikely to fail and redundant systems as reasonable candidates for a reactive, run-to-failure approach. Decide deliberately and write it down.
What is the difference between time-based and meter-based PM?
Time-based PM is triggered by the calendar, such as every month. Meter-based PM is triggered by use, such as every 250 run hours. Use meters when equipment usage varies a lot between assets or seasons.
How far in advance should PM work orders be created?
Long enough to plan the work. A week ahead suits most monthly tasks; a month or more suits annual work that needs parts, a contractor or a shutdown window.
Can I run a PM schedule from a spreadsheet?
For a handful of assets, yes. Once you need automatic work orders, mobile checklists and completion reporting, a CMMS or a purpose-built app is usually less work than maintaining the spreadsheet.
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