Ops Teams: Five Steps to an Equipment Maintenance Schedule That Works
How to build an equipment maintenance schedule people actually follow
An equipment maintenance schedule tells you which asset gets which task, how often, and who's responsible, with a next due date and a status for every line. That's the minimum: asset ID, task and frequency, owner, next due. If you've got none of it yet, don't start with a template. Start with a walk round the site, a roll of asset tags and a list of the machines that would stop production if they failed.
Then pilot the schedule on your ten to twenty most critical assets before you roll it out everywhere. Measure it for ninety days and adjust. A schedule nobody measures isn't a programme, it's a document.
Sort every task into a frequency band
Most working schedules sort tasks into bands, from per-shift checks up to annual overhauls. Mapping your equipment onto these bands is quicker than inventing intervals machine by machine.
Daily or per-shift checks belong to operators. Look for leaks and damage, check fluid levels, test the emergency stop, confirm guards are in place. None of it needs a technician, which is exactly why it can happen every shift. For some equipment it's the law: OSHA requires forklifts to be examined before they're put into service, at least daily, and after every shift where they run round the clock (29 CFR 1910.178(q)(7)).
Weekly and monthly work is hands-on: lubrication, filter changes, belt and chain tension, cleaning. A conveyor running three shifts a day may need weekly tension checks. The same conveyor on one shift might get by with monthly. Duty cycle sets the interval, not the calendar alone.
Quarterly and semi-annual tasks bring in condition monitoring: vibration analysis, thermal imaging on electrical panels, oil sampling, shaft alignment. You'll usually need a trained technician and kit such as a vibration analyser or an infrared camera.
Annual tasks are the heavy lifts: overhauls, instrument calibration, and rebuilds triggered by what the quarterly readings showed. A pump whose vibration crept up all year might move from "annual inspection" to "annual rebuild" on the strength of that data.
| Band | Typical tasks | Usually done by |
|---|---|---|
| Daily or per shift | Visual check, fluid levels, e-stop test, guards in place | Operator |
| Weekly or monthly | Lubrication, filters, belt and chain tension, cleaning | Maintenance technician |
| Quarterly or semi-annual | Vibration analysis, thermography, oil sampling, alignment | Reliability technician |
| Annual | Overhaul, calibration, condition-based rebuild | Senior technician or contractor |
Every task should point to a standard operating procedure. "Lubricate motor" tells a new technician nothing about grease type, quantity or which panel to open. "Lubricate motor per SOP-MTR-014" gives them all of it without a call to the supervisor.
Five steps to build the schedule from scratch
Do these in order. Skip one and you'll be rebuilding the whole thing in six months, once the gaps show up on the shop floor.
- Give every asset a unique ID and record its manufacturer, model, install date and, where it has one, the hour meter reading. It's tedious. It's also the step most teams skip, and the reason an auditor's question like "which pump is this, exactly?" ends in silence.
- Rank assets by criticality. Score each on production impact, safety risk and cost of failure. A backup compressor that sits idle most of the year ranks below the single chiller cooling your server room. Criticality decides how often and how tightly each asset gets maintained.
- Pull tasks from the OEM manual and your own failure history, then map each one to an SOP. The manual is a starting point. If a bearing has failed three times in two years while you followed the manual's interval, your history wins.
- Set starting frequencies, converting hour-based OEM intervals into calendar dates. A 500-hour service on a generator running 8 hours a day, every day, comes round about every two months. Get the maths wrong and you'll over-service idle kit or miss intervals on kit that never stops.
- Name an owner for every task, set the next due date, and decide how you'll measure compliance: PM tasks done on time divided by PM tasks scheduled. A task without a named owner is a task everyone assumes someone else has.
Turn OEM hours into calendar dates, then tune them
Manuals talk in running hours. Your schedule needs dates, because nobody wants to read an hour meter every morning to find out whether today's a service day.
Once you know the duty cycle, the maths is simple. Take a 500-hour interval on a machine running 16 hours a day, five days a week. That's 80 hours a week, so 500 hours arrives in about 44 days, roughly six weeks. Drop the same machine to 8 hours a day and the interval stretches to about three months. Same manual, very different dates. That's why a calendar-only schedule built without checking real runtime either wastes visits or misses them.
Environment shifts it again. A machine in a dusty foundry, a humid coastal plant or an outdoor yard in winter wears faster than the same model in a clean, heated building. Tighten intervals for those assets, then let the condition data tell you by how much.
When condition data disagrees with the calendar, the data wins. If vibration or oil analysis shows a bearing degrading faster than the interval assumes, bring the date forward. If readings stay flat well past the due date, that's a fair reason to stretch the interval instead of servicing a machine that doesn't need it yet.
Put an annual interval review on the operations calendar as a fixed date. Record the hours-to-days conversion in each asset record too. When an auditor asks how you got a due date six months from now, you want the working written down, not in one planner's head.
Some intervals aren't yours to tune
Statutory inspections sit on the schedule at a fixed maximum, whatever your condition data says. In the US, OSHA's lockout/tagout standard requires you to inspect each energy control procedure at least once a year (29 CFR 1910.147(c)(6)(i)). The forklift pre-use check above is another.
Mark these lines as fixed in your schedule so nobody stretches them during an interval review. Pressure systems, lifting equipment and fire equipment often carry their own legal or insurer intervals too, so check the rules that apply to your site and country before you set the dates.
Start in Excel, but with the right columns
A spreadsheet is a perfectly good place to start, and paying for software you don't need yet is wasted money. It only works if it has the right columns from day one.
| Column | Why it's there |
|---|---|
| Asset ID | Ties the line to the tag on the physical machine |
| Location | Building, line or zone, so the technician can find it |
| Manufacturer and model | For parts orders and the OEM manual |
| Task and frequency | What gets done and how often |
| Trigger type | Calendar, running hours or condition |
| Owner | A named person or role |
| Last done and next due | The two dates that make it a schedule |
| Status | On track, due soon or overdue |
| SOP link | The procedure for that exact task |
| Hours and parts | For labour planning and stock |
Next due and Status are the columns that matter most. Leave them out and you've got a task list, not a schedule.
Excel is fine for one production line or a few dozen assets looked after by one or two people. You've outgrown it when you run several sites, when enough technicians edit the file that nobody trusts which copy is current, or when a customer or regulator wants evidence a spreadsheet can't give you. At that point a CMMS that generates the work orders for you is worth the money.
Measure it monthly and keep the evidence
A schedule earns its keep when it turns into work orders with the right SOP and parts list attached, not when it sits in a shared drive.
Four numbers tell you whether it's working:
- PM compliance: tasks completed on time divided by tasks scheduled. Write down a target and make it stricter for critical assets.
- Overdue jobs: how many tasks are past their next due date right now.
- Mean time to repair (MTTR): how long a failure keeps a machine down.
- Mean time between failures (MTBF): how long a machine runs between breakdowns. Watch your backlog in days alongside it.
Critical assets get the stricter target because a missed task there costs far more. A late filter change on a spare compressor is an annoyance. On the only chiller, it's an outage.
Capture evidence as the work's done: a photo of the finished job, the technician's sign-off, and the readings logged against the SOP. An auditor doesn't want to hear the work happened. They want to see it.
Then close the loop each month. If a motor keeps failing between lubrication cycles, that isn't bad luck. The interval's wrong. Tighten it based on what failed, rather than leaving it alone and hoping the next breakdown isn't on a production day.
Worked example: one centrifugal pump
Here's how a single pump's record might look before it goes into a spreadsheet or a CMMS. The SOP numbers are placeholders for your own.
| Frequency | Task | Owner | SOP |
|---|---|---|---|
| Every shift | Check seal for leaks, listen for noise, check discharge pressure | Operator | SOP-PMP-001 |
| Weekly | Inspect coupling and guard | Maintenance technician | SOP-PMP-002 |
| Monthly | Check baseplate bolt torque, grease bearings if specified | Maintenance technician | SOP-PMP-003 |
| Quarterly | Vibration readings and oil sample | Reliability technician | SOP-PMP-004 |
| Annually | Strip, inspect, replace bearings if wear shows | Contractor | SOP-PMP-005 |
A few notes on reading it:
- The shift check takes a few minutes and needs no tools, which is why it can happen every shift.
- The quarterly readings decide whether the annual job is a routine inspection or a full bearing change.
- A pump handling corrosive fluid might need the strip-down every six months instead.
- Each SOP should open on a phone or tablet next to the pump, not live as a file on someone's desktop.
The same layout works for motors, compressors and conveyors. Only the tasks change.
Most schedules fail on admin, not engineering
You can build a schedule perfectly and still watch it fail, because the gap between "scheduled" and "done" is usually administrative. Someone forgets to open the spreadsheet. A due date passes quietly on a Friday afternoon. Nobody escalates until the machine stops.
The rollouts that fail rarely get the frequency maths wrong. They fail on quieter things: tasks with no SOP, so quality depends on whoever's free; tasks with no named owner; and intervals copied straight from the manual with no adjustment for duty cycle or environment.
Dated items are where this bites hardest: calibration certificates, statutory inspections, service contracts and the engineers' own tickets and licences. They often sit outside the CMMS, and they're the ones an auditor asks for first.
Next step: put your critical dates where someone will see them
Pick your ten most critical assets this week. Give every task an owner, an SOP and a next due date, and add the statutory inspections and calibration dates that go with them.
If those dates keep slipping, ExpiryEdge can hold the dated side of the programme: calibration certificates, inspection deadlines and service contracts. It reminds the owner by email, SMS, WhatsApp, Slack or Teams, escalates anything overdue to a supervisor, and keeps an audit trail you can export. It isn't a CMMS and won't raise work orders. There's a 14-day free trial, no credit card needed, so load your critical assets first and see what's due.
FAQ
What is the 80/20 rule in maintenance?
It's the Pareto idea applied to equipment: a small minority of your assets usually causes most of your downtime and repair cost. It's a rule of thumb, not a law, but it's why you rank assets by criticality before you set any frequencies.
What should a maintenance schedule include?
At minimum: an asset ID, the task and how often it runs, a named owner, and the next due date with a status. Better schedules also link the SOP, record the trigger type, and list labour hours and parts.
What is a typical maintenance schedule?
Daily operator checks, weekly or monthly lubrication and cleaning, quarterly or semi-annual condition monitoring such as vibration analysis, and annual overhauls or calibration. Duty cycle and environment decide where each task really sits.
Can you provide an example of a maintenance schedule?
For a centrifugal pump: a seal check every shift by the operator, a weekly coupling inspection and monthly bolt torque check by a technician, quarterly vibration and oil sampling by reliability staff, and an annual strip-down by a contractor.
How do I know if my equipment maintenance schedule is actually working?
Track PM compliance (tasks done on time divided by tasks scheduled) every month against a written target, stricter for critical assets. If compliance is high but breakdowns aren't falling, change the intervals or the task content.
Recommended
Frequently asked questions
At minimum: an asset ID, the task and how often it runs, a named owner, and the next due date with a status. Better schedules also link the SOP for each task, record the trigger type (calendar, hours or condition), and list estimated labour hours and parts so the planner can book people and stock in advance.
Most schedules group tasks by how often they run. Operators do quick checks every day or shift. Technicians handle lubrication, filters and belt tension weekly or monthly, while every quarter or half year someone takes vibration readings and oil samples. Once a year you overhaul and calibrate. How hard a machine works, and where it lives, decides where each task really sits.
Take a centrifugal pump. The operator checks the seal for leaks every shift. A technician inspects the coupling each week and checks bolt torque each month. Reliability staff take vibration readings and an oil sample every quarter. Once a year, a contractor strips it down and inspects it. Each line names an owner, an SOP and a next due date.
Track PM compliance: tasks done on time divided by tasks scheduled, reported monthly. Write down a target. For critical assets, make it stricter. Then watch two numbers, the overdue count and the time between failures. If compliance is high but breakdowns aren't falling, don't blame your crew's discipline. Your intervals or the tasks themselves need changing.
