In many small factories, maintenance means fixing things when they break. It feels economical: no time spent on machines that are working, no parts replaced before they fail, and a skilled technician who can usually get things running again. The cost only becomes visible when a critical machine stops in the middle of a big order, the part is not in stock, the repair takes three times longer than it would have on a planned day, and the same failure happens again a month later.
A breakdown is almost always more expensive than the same repair done at a planned time. It arrives when production is busiest, it often damages surrounding parts, the right spare may not be on hand, people are pulled off other work, and the restart takes longer. The goal is not to maintain everything as much as possible. It is to decide, machine by machine, how much and what kind of maintenance makes sense, and to stop the same failures recurring.
This article explains the main maintenance strategies, how to rank equipment by how critical it is, how operators can take on daily care, how to plan maintenance windows, how to learn from repeat failures, and what to measure. It is general information for owners and managers of small manufacturing and processing businesses. Follow manufacturers’ instructions and safety requirements, including isolation and lockout procedures, for all maintenance work.
Four maintenance strategies
| Strategy | What it means | Suits |
|---|---|---|
| Run to failure | Fix it when it breaks | Cheap, non-critical items with spares on hand and no safety impact |
| Time-based preventive | Service or replace at fixed intervals or usage | Wear parts with predictable lives, such as filters, belts and lubricants |
| Condition-based | Check condition and act when it starts to deteriorate | Items whose failure develops gradually and can be detected, such as bearings, hoses and electrical connections |
| Predictive | Use trends in measurements, such as vibration, temperature or oil analysis, to forecast failure | Critical equipment where early warning is worth the cost of monitoring |
Run to failure is a legitimate choice for the right items. The mistake is making it the default for everything, including the machines the business cannot afford to lose. The underlying idea, sometimes called reliability-centred maintenance, is that maintenance requirements depend on each asset’s function and operating context, so the strategy should be chosen deliberately rather than inherited.
Rank equipment by how critical it is
Not every machine deserves the same attention. A simple ranking uses a few questions:
- What happens to output if it stops? Is it a bottleneck, or is there a backup?
- What does a failure cost, in lost contribution, repair, scrap and late deliveries?
- Is there a safety or environmental consequence if it fails?
- How long would a repair take, including getting parts?
- How often has it failed in the past year?
Sort equipment into three groups: critical, important and minor. Critical equipment gets condition checks, planned maintenance, critical spares on hand and, where justified, predictive monitoring. Minor equipment may reasonably run to failure. The putting a dollar value on equipment losses article covers valuing downtime so the ranking reflects real money.
Operators as the first line of maintenance
The people who run a machine every day are best placed to notice that something has changed: a new noise, a leak, a hotter motor, a slower cycle. Total productive maintenance, or TPM, builds on this by giving operators simple daily care tasks, often summarised as clean, inspect, lubricate and tighten. A short daily checklist might include:
- Look for leaks, loose fasteners and damaged guards.
- Check fluid levels and gauges against marked normal ranges.
- Clean the machine and its surroundings, which makes problems visible.
- Listen and feel for unusual noise, vibration or heat.
- Record anything abnormal on a tag or log, so maintenance can act.
Operators need training, clear limits on what they may adjust, and confidence that reported problems will be dealt with. If reports go nowhere, they stop. Mark normal ranges directly on gauges and sight glasses, so anyone can see at a glance when something is outside them.
Plan maintenance windows
Planned maintenance works best when it is grouped into scheduled windows rather than squeezed in whenever a machine is free. Bundling tasks means one shutdown covers several jobs, parts and people are ready in advance, and the restart is planned. For each critical machine, keep a simple list of planned tasks, their intervals, the parts and tools needed, and the time each takes.
When a condition check shows a problem developing, schedule the repair into the next window rather than waiting for failure. A repair done on a planned day, with the part already on the shelf, typically takes a fraction of the time of the same repair during an unplanned stoppage.
Keep the right spares
Spares are a balance. Too many ties up cash and space; too few turns a two-hour repair into a two-week wait. For critical equipment, identify the parts whose failure would stop the machine and that take time to obtain, and keep them on hand, checked and labelled. For other items, know where to get parts quickly. Record which spares were used, so stock levels reflect real failure rates. The buying for the whole life of equipment article covers planning spares and support before a machine is bought.
Learn from repeat failures
The most valuable maintenance records are failure records. For each breakdown, note the machine, the failed part, the symptoms, the cause if known, the downtime and the fix. Every month, look for patterns: the same failure on the same machine, or the same part failing across several machines. A repeat failure is a problem with a cause that has not been found.
For repeat failures, ask why several times until you reach something you can change: a missing filter, a lubrication task that is skipped, an operating setting outside the design range, a part of the wrong specification. Then change the cause, not just the part.
Give lubrication its own routine
A large share of mechanical failures traces back to lubrication: the wrong lubricant, too much or too little, contamination with dust or water, or a task that is simply skipped. Lubrication is cheap, so it is often treated casually. Keep a simple lubrication schedule listing each point, the lubricant, the quantity and the interval. Label containers and grease guns so products are not mixed, store lubricants clean and sealed, and record each lubrication task as done. Where a point is hard to reach, it will eventually be missed, so consider making access easier.
Design and buy for maintainability
Maintenance effort is largely decided when equipment is bought or modified. Ask, before purchase, how easily wear parts can be reached and replaced, whether parts are standard or proprietary, where condition checks such as temperature or vibration readings can be taken, and what training and documentation come with the machine. Involve whoever maintains the equipment in these decisions. A machine that is cheap to buy but slow to service costs more over its life.
If you use maintenance contractors
Many small businesses rely on outside technicians. Contractors work better when they have your equipment ranking, the failure history and the planned task lists. Agree response times for critical machines in advance, ask for a short report after each visit covering what was found, what was done and what is likely to need attention next, and keep those records yourself rather than relying on the contractor’s files. Your knowledge of your equipment should not leave when the contractor does.
What to measure
A few measures show whether maintenance is improving:
- Unplanned downtime on critical equipment, in hours per month.
- The share of maintenance work that is planned rather than reactive.
- Repeat failures: how many breakdowns are the same failure as before.
- Time between failures on critical machines.
- Planned tasks completed on time.
Overall equipment effectiveness combines availability, speed and quality losses into one figure; the measuring productivity and equipment effectiveness article explains how to calculate and use it.
A worked example
This is an illustration. A plastics injection moulding business runs eight moulding machines. Two of them produce most of the parts for its largest customer. Maintenance is mostly reactive, and unplanned downtime on those two machines averages about 30 hours a month, mainly from hydraulic hose failures, heater band failures and chiller trips.
Ranking. The owner and the maintenance technician rank the machines. The two machines serving the main customer are critical; three are important; three older machines making low-volume parts are minor and stay largely run-to-failure.
Operator care. Operators on the critical machines get a five-minute start-of-shift checklist: look for oil leaks, check hydraulic oil level and temperature against marked ranges, check heater band indicators and confirm chiller water flow. Abnormal findings go on a tag hung on the machine and into a shared log.
Planned windows. A three-hour planned maintenance window is scheduled every second Saturday morning, when production is light. Hose inspections, filter changes and electrical checks are bundled into it.
Condition checks. The technician uses a handheld thermal camera on heater bands and electrical cabinets once a month, and hydraulic oil is sampled every quarter. Two heater bands showing hot spots are replaced in a planned window instead of failing mid-run.
Spares. Critical spares for the two machines, including hoses, heater bands and thermocouples, are now held on site.
Repeat failures. Chiller trips keep recurring. Asking why leads to a condenser clogged with dust from a nearby grinding area. Regular condenser cleaning is added to the planned schedule, and a dust screen is fitted. The trips stop.
After four months, unplanned downtime on the two critical machines falls to about 12 hours a month, while planned maintenance takes about 6 hours a month, mostly outside production time. At an illustrative lost contribution of around $120 per machine-hour, the 18 fewer hours of unplanned downtime are worth roughly $2,000 a month, before counting fewer late deliveries to the main customer.
How this applies to a small Australian business
- Rank your equipment into critical, important and minor.
- Choose a strategy for each, rather than one default.
- Give operators simple daily checks and act on what they report.
- Group planned work into scheduled windows.
- Hold critical spares for critical equipment.
- Record every breakdown, and review for repeats monthly.
- Fix causes, not just failed parts.
- Name one owner for the maintenance routine.
- Follow safe isolation procedures and manufacturers’ instructions.
- Keep a lubrication schedule with points, products and intervals.
Signals worth watching
- Breakdowns during the busiest periods.
- Long waits for parts.
- The same failure on the same machine again and again.
- Operators who have stopped reporting problems.
- Planned tasks regularly postponed.
- Contractor visits with no written report.
- No record of what failed and why.
- Unlabelled lubricants and skipped lubrication points.
Common mistakes
- Running everything to failure by default.
- Maintaining everything the same way, regardless of how critical it is.
- Replacing parts without finding why they failed.
- Ignoring operators’ observations.
- Overstocking minor spares while missing critical ones.
- Letting production pressure cancel planned maintenance every time.
- Treating lubrication casually because it is cheap.
Frequently asked questions
Do we need maintenance software? Not at first. A spreadsheet with the equipment list, ranking, planned tasks and a failure log is enough for most small factories. Software helps when the list grows.
How often should we service each machine? Start with the manufacturer’s recommendations, then adjust based on your own failure records and operating conditions.
Is predictive maintenance worth it for a small business? For critical machines, simple tools such as thermal cameras, oil analysis or vibration checks can be very cost effective. Start with the equipment where a breakdown hurts most.
What should operators not do? Anything beyond their training and the limits you set, especially work requiring isolation, electrical work or adjustments to safety devices.
Who should own maintenance in a small business? One named person, even part-time, who keeps the equipment list, schedules and failure log current. Without an owner, planned maintenance quietly becomes reactive again.
How do we protect planned maintenance from production pressure? Schedule it, give it an owner and treat a cancelled window as a decision that someone must approve and record.
Questions to ask
- Which machines can we least afford to lose?
- What strategy have we deliberately chosen for each?
- Who owns our equipment list, schedules and failure log?
- What do operators check each day, and what happens when they report a problem?
- Which failures have happened more than once this year?
- Do we hold the spares that would stop a critical machine?
- How much of our maintenance work is planned?
Bringing it together
Breakdowns cost more than planned repairs, but maintaining everything heavily wastes money too. Rank equipment by how critical it is, and choose a strategy for each: run to failure for minor items, scheduled servicing for predictable wear, condition checks and monitoring for critical equipment. Give operators simple daily care, group planned work into windows, hold the right spares and treat every repeat failure as a cause still to be found. Give one person ownership of the routine. Measured over a few months, fewer and shorter breakdowns usually repay the effort many times over.
Source: KEVOS notes, drawing on earlier KEVOS engineering handbooks on maintenance excellence, total productive maintenance and reliability-centred maintenance, and on overall equipment effectiveness and loss analysis. Examples and figures in this article are illustrations. This article is general information.
