How to Set PM Intervals in Days, Hours, or Cycles (and When to Use Each)

The Interval Unit Question That Trips Up Every Asset Register

You've just decided to formalize PM on a dozen assets — good. The first thing the spreadsheet asks is: when is it next due? Simple enough for the HVAC unit that gets a filter change every 90 days. Harder for the CNC machining center whose spindle lubrication is specified at every 500 runtime hours. Harder still for the hydraulic press that wears its seals in proportion to how many strokes it completes, not how long the clock runs.

Choose the wrong unit and the maintenance interval becomes fiction. Set a 500-hour lube task in calendar days on a machine running six hours per day and you'll trigger it 23 days early, every cycle, accumulating unnecessary labor and parts cost. Set it on a machine running 14 hours per day and you'll miss it by 11 days every time — and a missed lube interval on a high-speed spindle is the kind of event that turns a $40 maintenance task into a $6,000 bearing replacement.

This guide explains what PM interval units actually mean, how to calculate a next-due date from each one, and how to match the right unit to the right asset class. By the end you'll have a clear decision rule you can apply to every piece of equipment in your facility.

---

What a PM Interval Unit Actually Means

A PM interval is how often a maintenance task is due — expressed in a unit that reflects what causes wear on that asset. The unit is not cosmetic; it determines the trigger.

Three units cover almost every industrial and light-manufacturing asset:

• Calendar days — the interval resets on the clock, regardless of whether the asset ran. A 90-day interval is due 90 days after the last PM, period.
• Runtime hours — the interval counts only the hours the asset actually operated. A 500-hour interval is due after 500 hours of running time have accumulated since the last PM.
• Cycles — the interval counts discrete mechanical events: strokes, shots, lifts, weld events, print cycles. A 50,000-cycle interval is due when the asset has completed 50,000 operations since the last PM.

Each unit maps to a different wear mechanism. Choosing the right one is not a preference — it's an engineering question, and the answer usually lives in the OEM manual.

PM interval (days/hours/cycles): the elapsed quantity of the chosen unit — time, runtime, or discrete operations — between successive preventive maintenance tasks on a given asset.

Always confirm specific intervals against the equipment OEM's documentation and any applicable recognized standards (ASHRAE for HVAC/R, NFPA 70B for electrical distribution, OSHA standards for powered industrial trucks). Intervals and requirements vary by equipment model, duty cycle, environment, and jurisdiction and change over time. The unit types described here are general starting points.

---

Calendar Days: When the Clock Is the Right Trigger

Calendar-based maintenance is the right choice when degradation is primarily time-driven rather than use-driven — when an asset deteriorates whether it runs or not.

Best fit:

• Lubricants and fluids that oxidize, absorb moisture, or degrade with age (compressor oil, gearbox oil, hydraulic fluid with long idle periods)
• Filters in environments with continuous ambient contamination (HVAC supply-air filters, paint-booth exhaust filters)
• Belts, hoses, and seals subject to dry-rot or ozone cracking
• Annual inspections mandated by regulation, insurance, or OEM warranty terms
• Assets that run continuously at near-constant duty (a boiler, a chilled-water pump, a conveyor that runs all three shifts)

When calendar days mislead: if an asset runs intermittently — four hours some weeks, twelve others — a calendar interval treats a light week and a heavy week identically. For those assets, runtime hours will track wear more accurately.

Worked example — next-due date in calendar days:

Input	Value
Last PM date	March 1
PM interval	90 days
Next PM due	May 30 (March 1 + 90 days)

Nothing more complex than addition. In a spreadsheet this is =A2+90. The number is reliable as long as the asset's wear actually correlates with elapsed time.

---

Runtime Hours: When Usage Is the Right Trigger

Runtime-based maintenance counts only the hours the machine actually operates. It is the standard OEM interval unit for rotating and reciprocating equipment, because wear in bearings, gears, seals, and blades is proportional to how long the machine runs — not how long it sits on the floor.

Best fit:

• CNC machining centers, lathes, mills (spindle lube, filter changes, way lubrication)
• Air compressors (valve inspection, separator element, belt tension)
• Diesel generators and engines
• Forklifts and lift trucks (engine hours — OSHA 29 CFR 1910.178 mandates regular inspections; confirm the applicable inspection interval and frequency requirement with OSHA or qualified counsel)
• Industrial fans and blowers running variable schedules
• Any asset whose OEM manual specifies "every X hours of operation"

The shift-pattern problem: runtime hours are only accurate if you actually track runtime. On assets without hour meters, this means estimation — typically average daily runtime × operating days. The accuracy of the interval is only as good as the accuracy of the hours estimate. See how to track runtime hours without IoT sensors for practical approaches.

Worked example — next-due date in runtime hours:

Suppose a CNC machining center requires spindle lubrication every 500 runtime hours.

Input	Value
Hours at last PM	4,200 hr
PM interval	500 hr
Target odometer at next PM	4,700 hr
Avg daily runtime	10 hr/day
Days until next PM	50 days (500 hr ÷ 10 hr/day)
Approximate calendar date	~50 days from last PM date

Now run the same machine at 6 hr/day vs 14 hr/day:

Daily runtime	Days to next PM
6 hr/day	~83 days
10 hr/day	~50 days
14 hr/day	~36 days

Three identical machines, three very different calendar schedules. A fixed-date approach would over-maintain the light-use machine and under-maintain the heavy-use one. For a deeper look at how next-due dates propagate across an asset register, see next PM due-date calculation explained.

---

Cycles: When Each Operation Is the Right Trigger

Cycle-based maintenance counts discrete mechanical events rather than elapsed time or runtime hours. It applies when the primary wear mechanism is load-per-event — the stress placed on a part each time a specific mechanical action completes.

Best fit:

• Hydraulic and mechanical presses (seal wear per stroke)
• Injection molding machines (barrel, screw, and mold maintenance per shot)
• Stamping and punching equipment (die maintenance per hit)
• Welding equipment (contact tip and liner life per weld event)
• Conveyor systems measured in load-cycles rather than hours
• Any asset whose OEM manual specifies "every X strokes / shots / cycles"

Cycle tracking without automation: like runtime hours, cycles require a counter. Modern presses and molding machines typically have built-in cycle counters on the PLC or HMI. Older equipment may need a low-cost mechanical counter or operator log. If neither is available, an average-cycles-per-shift estimate can serve as a proxy — with the same caveat about accuracy.

Worked example — next-due date in cycles:

A hydraulic press requires seal inspection every 50,000 strokes.

Input	Value
Counter at last PM	210,000 strokes
PM interval	50,000 strokes
Target counter at next PM	260,000 strokes
Avg strokes per shift (8 hr)	2,500
Shifts per day	2
Daily cycle rate	5,000 strokes/day
Days until next PM	10 days (50,000 ÷ 5,000)

Change the throughput — say the line runs slower during a product changeover at 3,000 strokes/day — and next PM slides to approximately 17 days. The cycle counter stays authoritative; the calendar date is just an approximation derived from it.

---

Choosing the Right Unit: A Decision Rule

Match the unit to the primary wear mechanism for each task on each asset. One asset can have multiple tasks with different units — a compressor might have a 90-day oil change (calendar, because oil oxidizes with age) and a 2,000-hour valve inspection (runtime, because valve wear is proportional to use).

Primary wear driver	Correct interval unit
Time/age (oxidation, corrosion, dry-rot, regulation)	Calendar days
Runtime (friction, heat, rotating wear)	Runtime hours
Load events (stroke, shot, hit, weld)	Cycles
Combination (age and use)	Use whichever comes first — dual-trigger logic

"Whichever comes first" logic is the conservative approach for high-criticality assets: set both a calendar threshold and a runtime/cycle threshold, and trigger PM at whichever limit is reached first. This is common OEM language ("every 6 months or 500 hours, whichever comes first"). When you implement this in a register or planner, you carry two interval values and check both at each review. For a broader comparison of calendar-based and usage-based approaches, see calendar-based vs. usage-based maintenance.

---

How the Unit Choice Affects Projected Annual Cost

The interval unit affects not only when PM is due — it also affects how many PM events occur per year, which directly drives the per-asset annual maintenance cost estimate:

Annual PM cost (single task) = (Annual occurrences) × (Labor hours per event × Labor rate + Parts cost per event)

Annual occurrences depend on the unit:

• Days: 365 ÷ interval days (e.g., 365 ÷ 90 = 4.06 PM events/year)
• Hours: (Annual runtime hours) ÷ interval hours (e.g., 2,500 hr/yr ÷ 500 hr = 5.0 PM events/year)
• Cycles: (Annual cycles) ÷ interval cycles (e.g., 900,000 strokes/yr ÷ 50,000 = 18 PM events/year)

Set the unit wrong and the annual cost estimate is wrong — often substantially. For a five-task asset, a miscounted unit on two tasks can shift the annual cost projection by 30%–60%, which has direct consequences for budget planning and for understanding your fleet's maintenance cost as a percentage of replacement asset value (MC/RAV) — the standard fleet-cost KPI.

Getting the unit right for every asset in a large register is where spreadsheets start to struggle. A persistent, multi-asset PM interval calculator tracks units, accumulates runtime or cycles between PMs, and recalculates next-due dates and annual cost estimates across the whole fleet automatically — a materially different capability from a one-time calculator widget or a tab-per-asset spreadsheet. See our features page to understand how the Maintenance Cost and Interval Planner handles all three interval units. For full guidance on translating intervals into a cost forecast, see the preventive maintenance interval and cost guide.

---

Building Your Asset-Level Interval Register

A workable asset-level register for PM intervals contains, at minimum:

1. Asset name and ID
2. PM task description (one row per task per asset)
3. Interval unit (days / runtime hours / cycles)
4. Interval quantity (e.g., 90, 500, 50,000)
5. Date or counter reading at last PM
6. Next PM due (calculated from the above)
7. Estimated labor hours per event
8. Estimated labor rate (enter your facility's actual rate — BLS OES median rates for general maintenance workers were $23.38/hr as of May 2024, but your rate will differ by region, skill level, and benefit loading; use your own number)
9. Estimated parts cost per event
10. Annual PM cost estimate (occurrences × unit cost)

If you are building this for the first time, the Annual PM Schedule Template gives you a pre-structured Excel workbook with columns for all three interval units, next-due-date formulas, and a fleet-level annual cost rollup — so you're not building the skeleton from scratch. You can also review when and how to deviate from OEM-specified intervals in PM interval override: when to deviate from OEM.

---

Start with the OEM Manual, Then Choose Your Unit

The practical sequence for setting PM intervals:

1. Pull the OEM manual for each asset. Locate every PM task and its specified interval. Note the unit the OEM uses — it is almost always deliberate.
2. Apply the decision rule. If the OEM specifies hours but your asset runs continuously at near-constant load, a calendar approximation may be acceptable. If duty cycles vary significantly, honor the runtime or cycle unit.
3. Set up your interval register with the correct unit per task, last-PM date or counter reading, and next-due calculation.
4. Estimate occurrences per year from each interval, then multiply by per-event labor and parts cost to get annual PM cost per task.
5. Sum across tasks and assets to produce your fleet-level annual PM cost estimate — the number your budget process actually needs.

Getting this right on paper is the prerequisite to knowing whether your maintenance spend is reasonable relative to your asset base, and to building a PM schedule you can defend to operations and finance alike.

The Annual PM Schedule Template is structured for exactly this workflow: open it, populate one row per task, and the next-due dates and annual cost estimates calculate automatically.