Hour-meter preventive maintenance: how to build the plan

In peak season, a wheel loader works two shifts a day. Its oil and filter change was scheduled “every three months”, which in a normal season meant about 250 hours. This time, after three months it had logged almost 500 hours. The filter clogged, the hydraulic pump suffered and the machine sat idle for a week, right when it was needed most. The plan existed; it was just measuring time with the wrong clock.

This article explains how to build a preventive plan based on the machine’s actual usage, how to calculate when each task is due and which indicators tell you whether the plan works. All of it can be done in a spreadsheet.

By usage or by calendar: when each one fits

Wear on an engine, an undercarriage or a hydraulic system depends on how much the machine works, not on how many days go by. But some things degrade over time even when the machine is idle: batteries, hoses, stored lubricants, fire extinguishers, technical inspections.

Trigger	When to use it	Examples
By usage (hours or kilometers)	Wear depends on operation and usage varies a lot from month to month	Engine oil, filters, bearings, blades, tires
By calendar	The component ages even when unused, or there is a fixed-date inspection	Batteries, hoses, safety inspections, paperwork reviews
Whichever comes first	The machine may sit idle for some seasons and run flat out in others	250 h or 3-month service, whichever arrives first

Rule of thumb: if a machine’s usage changes by more than 30% between seasons, the calendar alone won’t do. Use the hour meter and keep the date as a backstop for when the machine works little.

Step by step to build the plan

1. List your machines and their criticality. High if a stoppage halts operations, medium if there is a backup, low if it has no impact. Start the plan with the critical ones.
2. Break them down into components. Engine, transmission, hydraulics, undercarriage. Tasks and costs make more sense per component than per whole machine.
3. Define tasks and intervals. Start from the manufacturer’s manual and adjust with your own history: dust, altitude, load and fuel quality all change wear.
4. Choose each task’s trigger: hours, date or whichever comes first.
5. Set a tolerance. For example, ±10% of the interval. Outside that window, the task counts as late.
6. Build each task’s parts kit. Filters, oil, gaskets. If the kit isn’t in the warehouse when the task comes due, the plan falls apart.
7. Decide who takes the readings and how often. Without reliable readings there is no hour-meter plan.
8. Review the plan every quarter with the indicators below.

How to calculate the next due point

The basic formula is:

Next due = reading at which the task was due + interval

Note the detail: you add from the reading at which the task was due, not from the reading at which it was done. Suppose a 250 h service was due at 5,000 h and was done at 5,040 h. The next one is due at 5,250 h, not 5,290 h. If you add from the actual completion, every delay carries over and the plan drifts without anyone noticing.

To know when it falls on the calendar:

Days remaining = (next due − current reading) ÷ usage hours per day

If the machine reads 5,130 h and works 16 h a day on average, there are (5,250 − 5,130) ÷ 16 = 7.5 days left. With that figure you can order the parts kit in time and schedule the stoppage with operations.

Readings: the part that usually fails

An hour-meter plan is only as good as its readings. Three simple rules:

• A reading never goes down. If someone writes 4,980 after 5,130, it’s a typo. The only exception is an hour-meter replacement, and it must be recorded as such, with the old meter’s final reading and the new one’s initial reading.
• Fixed frequency. Daily for critical machines, weekly for the rest.
• One person responsible per machine. The operator records and the supervisor checks.

MTBF and MTTR with an example

Two indicators tell you whether preventive maintenance is working:

• MTBF (mean time between failures) = operating time ÷ number of failures. Higher is better.
• MTTR (mean time to repair) = total repair time ÷ number of failures. Lower is better.
• Availability = MTBF ÷ (MTBF + MTTR).

Suppose a truck had 1,500 planned operating hours in a quarter. It had 5 unplanned failures and racked up 40 hours of repair.

• Operating time = 1,500 − 40 = 1,460 h
• MTBF = 1,460 ÷ 5 = 292 h
• MTTR = 40 ÷ 5 = 8 h
• Availability = 292 ÷ (292 + 8) = 97.3%

How to read them: if MTBF rises quarter after quarter, preventive work is preventing failures. If MTTR rises, the problem is usually waiting for parts or technicians, not the repair itself. Count only unplanned failures; planned stoppages are not failures.

Maintenance cost per machine

Maintenance cost = parts + labor (hours × rate) + outside services

And to compare machines that work differently:

Cost per operating hour = maintenance cost ÷ hours operated

Suppose three machines over the same half-year:

Machine	Parts	Labor	Outside	Total	Hours operated	Bs per hour
Loader	18,000	4,000	2,000	24,000	1,200	20
Truck 1	9,000	3,000	0	12,000	1,500	8
Truck 2	21,000	5,000	4,000	30,000	1,000	30

Both trucks are the same model, but the second costs almost four times as much per hour. That’s the question to ask: does it have a recurring failure, a different operator, a harsher route, or is it at the end of its useful life? This figure, together with the machine’s book value, is the basis for deciding between repairing and replacing.

For the calculation to be useful, every part that leaves the warehouse must be charged to a specific machine, not to “general maintenance”.

Corrective vs. preventive

% preventive = preventive maintenance hours ÷ total maintenance hours × 100

Preventive compliance = tasks done within tolerance ÷ tasks due in the period × 100

If most of the shop’s hours go to corrective work, the team is firefighting. If compliance is low, the plan exists only on paper. Look at both together: high compliance with many failures means the tasks or intervals are wrong.

To tackle recurring failures, record the failure mode (what happened), the cause and the action on every corrective job. Sorted from most to least frequent, a few causes usually explain most stoppages.

Common mistakes

• Scheduling everything by calendar because it’s easier to remember.
• Calculating the next due point from the date the task was done.
• Accepting readings that go backward without looking into it.
• Taking parts from the warehouse without charging them to a machine.
• Counting planned stoppages as failures and distorting MTBF.
• Never reviewing intervals against real history.

Checklist

• List of machines with criticality.
• Main components of each critical machine.
• Tasks with interval, trigger and tolerance.
• Parts kit per task, with stock available.
• Person responsible and reading frequency per machine.
• Rule for hour-meter replacements.
• Every corrective job with mode, cause and action.
• MTBF, MTTR, compliance and cost per hour reviewed every quarter.

If you want to automate it

A spreadsheet works as long as someone keeps it up to date. As the fleet grows, an ERP like MKA can run plans by calendar, by hour meter or both, reject readings that go backward, deduct each order’s parts from the warehouse and show each machine’s reliability and cost in its reports (see fleet and heavy equipment).