The six strategies at a glance
"Types of maintenance" gets presented as a menu you pick one item from. In a real plant it is nothing of the sort — a mature operation runs several strategies at once, matched to each asset's criticality and cost of failure. The strategies form a rough ladder from reactive to proactive, and the skill is knowing where on that ladder each machine belongs. Here is the map, before we take each rung in turn.
| Strategy | Trigger | Best for |
|---|---|---|
| Reactive | After the asset fails | Cheap, low-criticality, quick-to-replace assets |
| Preventive | Calendar or usage interval | Most critical machines — the workhorse strategy |
| Condition-based | A measured condition crosses a limit | Assets with a measurable wear signal |
| Predictive | Analytics forecast the failure | High-value, high-criticality, well-instrumented assets |
| RCM | A method, not a trigger | Deciding the right strategy per asset |
| TPM | A culture, not a trigger | Whole-plant equipment effectiveness |
1. Reactive maintenance
Reactive maintenance means acting after a failure. It has two honest sub-types. Run-to-failure is a deliberate choice: you decide, in advance, that an asset is cheap and non-critical enough that letting it break is the cheapest sensible policy. Breakdown maintenance is the unplanned emergency version — the machine you did not plan to lose has stopped, and you scramble. The difference between them is intent, and it is the whole game: run-to-failure is a strategy; a breakdown on a critical asset is a strategy that failed.
Reactive work is not villainous. On the right assets it is optimal. It becomes a problem when it is the default for everything because no one has sorted assets by criticality or set up any preventive schedules. For the full trade-off, see preventive vs breakdown maintenance.
2. Preventive maintenance
Preventive maintenance (PM) is scheduled work done before failure, on a fixed trigger. That trigger is either a calendar interval (every 30 days, every quarter) or a usage count (every 500 running hours, every 100,000 cycles or strokes, every 10,000 km). A checklist defines the tasks, a safety work permit governs the hazardous ones, and the job is done in a chosen window rather than under emergency pressure.
For the overwhelming majority of assets in the overwhelming majority of plants, preventive maintenance done consistently is where the largest, fastest reliability gains live. It does not require instrumentation or data science — it requires a clean asset register, real schedules, and the discipline to do the work on time. That last part is exactly what a CMMS enforces, through PM schedules with due-date alerts. This is the workhorse strategy, and getting it right is the prerequisite for everything above it on the ladder.
3. Condition-based maintenance
Condition-based maintenance (CBM) triggers work not by a calendar but by a measured condition. You monitor a signal that tracks wear — vibration, bearing temperature, motor current, oil particle count, pressure differential across a filter — and you act when it crosses a threshold. The advantage over fixed-interval PM is that you neither change a healthy part too early nor miss a part degrading faster than the schedule assumed; you act when the asset actually needs it.
CBM can be as low-tech as a technician taking a monthly vibration pen reading and recording it against a checklist item, or as high-tech as permanently mounted sensors. The low-tech version is very achievable for an SME — the reading is just another value captured on a PM checklist, and a not-OK reading spawns a corrective job. That makes condition-based work a natural, affordable extension of a solid preventive programme rather than a separate technology project.
4. Predictive maintenance
Predictive maintenance (PdM) takes condition data a step further: it uses trend analysis and analytics to forecast when an asset will fail, so the repair can be planned for just before the failure would have happened. Done well, it captures the best of both worlds — minimal premature part changes and almost no surprise failures. It is the most sophisticated rung on the ladder.
It is also the most demanding. Predictive maintenance needs instrumentation to gather condition data continuously, a data pipeline to store and clean it, and the analytical capability — statistical or AI-based — to turn trends into reliable forecasts. When those are present on a high-value, high-criticality asset whose surprise failure is genuinely catastrophic, predictive maintenance pays for itself. When they are not, it is an expensive answer to a question preventive maintenance already answers well enough. This is where honesty matters, which brings us to the India take below.
Not sure which strategy each of your machines needs?
Start with a clean asset register and real PM schedules. We can show both on your own assets in 30 minutes — cloud or on-premise.
5. Reliability-centred maintenance (RCM)
Reliability-centred maintenance (RCM) is the odd one out — it is not a trigger type at all, but a method for choosing the right strategy for each asset. RCM works systematically through an asset's functions, the ways it can fail (its failure modes), and the consequences of each failure — on safety, on production, on cost — and then selects the most cost-effective response for each mode: preventive, condition-based, predictive, run-to-failure, or even a redesign so the failure cannot recur.
RCM is the intellectual reason a mature plant runs several strategies simultaneously. Rather than imposing one philosophy on everything, it asks, asset by asset and failure mode by failure mode, "what is the smartest, cheapest way to manage this risk?" You do not need a full formal RCM study to borrow its logic. The core habit — rate each asset's criticality, understand how it tends to fail, and match the strategy to the consequence — is available to any plant with a decent asset register and a bit of breakdown history to learn from.
6. Total Productive Maintenance (TPM)
Total Productive Maintenance (TPM) operates on a different axis again. It is a maintenance culture, built around maximising Overall Equipment Effectiveness (OEE), whose signature idea is autonomous maintenance — training machine operators to do basic cleaning, inspection and lubrication so small problems are caught by the people closest to the machine, freeing skilled technicians for the harder work. TPM is organised around a set of pillars (autonomous maintenance, planned maintenance, quality maintenance, early equipment management, training, and more) and it is deeply embedded in Indian automotive and engineering plants.
TPM and the strategies above are not alternatives — TPM is the organisational frame that gets preventive and condition-based work actually done, by making maintenance everyone's job rather than one department's. Its evidence needs — PM compliance, downtime and OEE data, planned-vs-done tracking — map directly onto what a CMMS produces, which is why Fast Maintenance's PM-plus-KPI model is described as TPM and ISO 55000 aligned.
The honest India take
Global maintenance content pushes predictive maintenance and IIoT sensors as the destination every plant should sprint toward. For most Indian SME and MSME plants, that advice is premature and expensive. The reality on the shop floor is that the biggest gains are still sitting in the basics: many plants do not yet have a clean asset register, real preventive schedules, or timestamped breakdown data. Bolting sensors onto that is building a roof before the walls.
The honest sequence is: get preventive right first. Register and tag every asset, put calendar and usage-based PM schedules on the machines that warrant them, stock spares to reorder levels, and capture every breakdown with downtime so MTTR and MTBF become real numbers. Done consistently, that alone moves most plants further than any sensor. Then, on the small handful of high-value, high-criticality assets where a surprise failure is genuinely catastrophic, add condition monitoring or predictive analytics selectively, where the arithmetic justifies it. Calendar and usage-based PM done well beats sensor hype — and it is what a CMMS makes achievable today.
How a CMMS supports every strategy
Fast Maintenance Software is the system that lets you run a mix of strategies and prove which each asset warrants — on the shared Fast Suite platform, cloud or on-premise:
Frequently asked questions
What are the main types of maintenance strategies?
The main types of maintenance strategy are: reactive (run-to-failure and breakdown repair, done after a failure); preventive (time- or usage-based scheduled work done before failure); condition-based (work triggered by a measured condition such as vibration, temperature or oil analysis); predictive (using condition trends and analytics to forecast failure); reliability-centred maintenance or RCM (a method for choosing the right strategy per asset based on failure modes); and Total Productive Maintenance or TPM (an operator-led culture that maximises overall equipment effectiveness). Most plants run a mix, matched to each asset's criticality.
What is the difference between preventive and predictive maintenance?
Preventive maintenance is scheduled by a fixed trigger — a calendar interval or a usage count such as running hours or cycles — regardless of the asset's actual condition. Predictive maintenance instead watches the asset's real condition, using sensors and trend analytics (vibration, temperature, current, oil debris) to forecast when failure will occur and act just in time. Predictive can avoid both premature part changes and surprise failures, but it needs instrumentation, data and analysis skills. Preventive is simpler, cheaper to start, and for most assets in most plants is where the biggest, quickest gains come from.
Is predictive maintenance worth it for Indian SME plants?
For most Indian SME and MSME plants, predictive maintenance with IIoT sensors is not the first place to invest. It suits a small number of high-value, high-criticality assets where a surprise failure is very costly. For the majority of machines, calendar- and usage-based preventive maintenance done consistently — with a clean asset register, real PM schedules, and disciplined spares — beats sensor hype and delivers most of the benefit at a fraction of the cost. The honest sequence is: get preventive right first, then add condition monitoring selectively where the numbers justify it.
What is reliability-centred maintenance (RCM)?
Reliability-centred maintenance (RCM) is not a single technique but a structured method for deciding which maintenance strategy to apply to each asset. It works through the asset's functions, the ways it can fail (failure modes), and the consequences of each failure, then selects the most cost-effective response — preventive, condition-based, predictive, run-to-failure, or a design change. RCM is why a mature plant runs several strategies at once: it applies the right one to each asset instead of forcing one philosophy across everything.
How does a CMMS support different maintenance strategies?
A CMMS is the system that lets you apply different strategies per asset and prove they work. It holds an asset register with criticality, drives calendar and usage-based preventive schedules, captures breakdowns for run-to-failure assets, can record condition readings against a checklist to support condition-based work, and reports MTTR, MTBF, availability and PM compliance so you can see which strategy each asset actually warrants. Fast Maintenance also folds gauge calibration in as scheduled work, supporting TPM and ISO 55000 practice.