In brief
Moving from corrective to planned maintenance starts by classifying equipment by criticality — the consequence of failure for safety, production, quality and cost — and assigning each class a different strategy. RCM, whose evaluation criteria are set out in SAE JA1011, decides for each failure mode whether a preventive task, a condition-based task, a failure-finding task or a redesign is warranted. TPM supplies the other half: the operator takes on basic inspection and cleaning, and the plant measures itself with overall equipment effectiveness. The transition is not made in one step: start with critical equipment, measure the ratio of planned to unplanned work, and expand as that indicator improves.
Almost no plant decides to be reactive. It becomes reactive. The maintenance team spends the day firefighting, planned work is pushed aside to deal with the urgent, and that postponement creates the next urgency. It is a cycle that closes on itself, and working more hours does not break it.
Breaking it means changing the rule by which time is allocated. And that requires first knowing which equipment deserves which strategy.
Criticality analysis: not everything deserves the same
Treating all equipment alike is expensive twice over: too much is spent on what does not matter and too little on what does. Criticality analysis ranks the asset base by the consequence of failure, not by the cost of the machine.
| Criterion | Question | Typical weight |
|---|---|---|
| Safety and environment | Can the failure injure someone or cause an environmental incident? | Highest: decides on its own |
| Production | Does the failure stop the line? Is there standby equipment? | High |
| Quality | Does the failure produce out-of-spec product? | High in continuous process |
| Repair cost and lead time | How much does it cost and how long does the spare take? | Medium, but it sets the inventory |
The result is a classification into three or four levels. Critical equipment gets condition monitoring and a detailed plan; intermediate equipment gets a standard preventive plan; low-criticality equipment can be deliberately run to failure, which is a legitimate decision where the consequence is low and the spare is on hand.
The P-F curve: why predictive maintenance exists
Between the moment a failure becomes detectable — point P — and the moment the equipment stops performing its function — point F — there is an interval. That interval is what makes timely intervention possible, and its length depends on the detection technique used.
- Vibration analysis: detects incipient bearing and gear damage weeks or months ahead.
- Thermography: reveals hot spots in electrical connections, couplings and overloaded bearings.
- Lubricant condition monitoring: identifies contamination, water and wear before there is any mechanical symptom.
- Structured sensory inspection: an operator who knows the machine notices changes in noise, smell or temperature that no installed sensor covers.
Inspection frequency has to be shorter than the P-F interval of the chosen technique. Measuring vibration once a year on equipment whose P-F interval is six weeks is equivalent to not measuring it: the failure happens between two readings.
RCM: the right question for each failure mode
Reliability centred maintenance is not a template or a piece of software: it is a decision method. SAE JA1011 sets the criteria a process must meet to be called RCM, and the core is to analyse each function of the equipment, how it can fail, what the consequence of that failure is and what task — if any — is worth doing.
That last qualifier is what changes practice most: in RCM a preventive task is justified only if it is technically feasible and worth doing. Replacing a component on the calendar when its failure is not age-related adds no reliability; it consumes hours and spares, and in some cases introduces infant mortality through the intervention itself.
| Failure consequence | Strategy that usually applies |
|---|---|
| Safety or environment | Mandatory task; if no effective task exists, redesign |
| Operational: stops production | Condition-based or preventive task depending on the failure pattern |
| Non-operational: repair cost only | Compare the cost of the task against the cost of the failure |
| Hidden failure (standby or protection) | Failure-finding task at a defined interval |
TPM: the part that depends on production
Total productive maintenance starts from an idea that is uncomfortable for both departments: the equipment does not belong to maintenance, it belongs to whoever operates it. Autonomous maintenance transfers cleaning, basic lubrication, tightening and sensory inspection to the operator, and frees the technician for work that needs their specialty.
Cleaning here is not cosmetic: it is inspection. An operator who cleans their equipment finds the leak, the loose fastener and the new vibration before anyone else. The indicator that organises the programme is overall equipment effectiveness, which combines availability, performance and quality into a single number and shows where capacity is really being lost.
The indicators worth tracking
- MTBF, mean time between failures: measures reliability. It rises when root causes are corrected.
- MTTR, mean time to repair: measures response capability. It falls with available spares, procedures and training.
- Availability: the combined result of the two above.
- Ratio of planned to unplanned work: the most honest indicator of where the programme really stands.
- Weekly schedule compliance: what percentage of planned work was executed in its window.
- Backlog: identified pending work hours. A zero backlog is not good news; it means nobody is inspecting.
Spares: the invisible half of the plan
A maintenance plan without a spares policy turns into a list of jobs that cannot be executed. The rule is to align inventory with criticality and lead time: critical items with long lead times are stocked, critical items with short lead times are agreed with the supplier, and non-critical items are bought when needed.
That agreement with the supplier is a technical component of the plan, not a purchasing arrangement: it means the distributor knows the equipment, keeps equivalent references identified and can respond while the machine is down.
Implementation route
- 01
Build the technical inventory
Equipment, components, part references and operating data. Without an asset hierarchy there is no history, and without history no analysis is possible.
- 02
Classify by criticality
With production, maintenance and safety involved. A classification made by maintenance alone does not hold once the urgency arrives.
- 03
Define a strategy per class
Condition monitoring for critical assets, preventive for intermediate ones, deliberate corrective for low-consequence items.
- 04
Start with a pilot machine
One critical machine, with a full plan and measurement. The pilot result is what convinces the rest of the organisation.
- 05
Measure and publish
The indicators must be visible to production, not only to maintenance. The conversation about priorities changes when both sides see the same number.
- 06
Extend down the criticality list
Widen the scope as the share of planned work rises. Trying to cover the whole plant in the first month is the most common way to abandon the programme.
Rodacaribe supports these programmes from the technical supply side: SKF bearings and seals, Q8 Oils lubricants, WIX filtration and power transmission components, with equivalent references identified and availability for each plant's critical equipment.
Frequently asked questions
- How long does it take a plant to move from reactive to planned?
- It depends on the starting point and on consistency, not on the method chosen. What can be said is that the change is gradual and measurable: the ratio of planned to unplanned work rises steadily when the programme works, and stalls when the plan keeps being displaced by urgencies.
- Do I need maintenance software to start?
- Not to start. You can begin with well-structured spreadsheets and a clear asset hierarchy. Software becomes essential when the volume of work orders and history exceeds what a sheet can handle, but deploying it on top of a process that does not exist only digitises the disorder.
- Do RCM and TPM compete with each other?
- No. RCM decides which task to do and how often, based on failure mode and consequence. TPM defines who does it and how the operator is involved. They complement each other: RCM provides the technical content of the plan and TPM the capacity to execute it day to day.
- Is it acceptable to run equipment to failure?
- Yes, if it is a deliberate, documented decision: low consequence, no safety risk, no secondary damage and a spare available. What is not acceptable is for it to happen by omission, without anyone having assessed it.
Related brands
Related industries
Need to solve this in your plant?
Our technical team reviews your case against real operating data and proposes the component and the procedure that fit.
About this contentThese guides collect industrial maintenance good practice and widely used standard references. They do not replace the technical documentation of the equipment or component manufacturer, which always prevails.