Industrial preventive maintenance is the set of interventions scheduled at predetermined intervals (time, operating hours or cycles) to reduce the probability of equipment failure across a plant. This guide covers preventive maintenance in industry: how to build the plan, set the interval by equipment family and measure whether it is working, not the maintenance of air conditioners or vehicles.
Almost every plant has a preventive maintenance plan. A far smaller share has a plan that works: with intervals that reflect real wear, tasks that attack actual failure modes and indicators that show whether the effort is buying reliability or just consuming labor hours. This guide walks the full path: definition, types of preventive maintenance, building the plan in six steps, the trap of the plan copied from the manual and the indicators that separate preventive maintenance that protects from preventive maintenance that only spends.
What is preventive maintenance
The definition from the ABNT NBR 5462 standard: preventive maintenance is maintenance "carried out at predetermined intervals, or according to prescribed criteria, intended to reduce the probability of failure or the degradation of an item's function". The trigger for the intervention is not the failure (as in corrective maintenance) nor the measured condition (as in predictive maintenance): it is the interval defined in the plan.
In industrial practice, preventive maintenance is the package of lubrication, inspection, technical cleaning, retightening, calibration, adjustment and scheduled replacement of components with a known service life: filters, belts, seals, wear elements. ABRAMAN, in its Documento Nacional survey, classifies it as maintenance "by time": carried out according to a pre-established schedule, including systematic routines, lubrication, inspections, major repairs, major shutdowns and general overhaul.
The technical premise of preventive maintenance is that the failure mode is age or usage dependent: the more the component runs, the higher the probability of failure. That is true for wear, fatigue and corrosion; it is not true for random failures (electrical surge, operating error, manufacturing defect), which do not follow a calendar. That is why preventive maintenance is indispensable and, at the same time, insufficient on its own: it protects against predictable wear, and the other strategies cover the rest, as a comparison of the different types of maintenance makes clear.
A note on scope: "preventive maintenance" is also a term used for air conditioning, vehicles and building maintenance, and much of the content available online deals with those contexts. The principle is the same (intervene at scheduled intervals), but the scale is different: here we are talking about industrial plants, with hundreds or thousands of assets, their own legal requirements, contested production windows and a plan running inside an ERP.
Advantages of preventive maintenance
Applied well, preventive maintenance delivers four direct gains:
Fewer unplanned shutdowns: predictable wear is handled before it turns into a breakdown, and production gains stability.
Predictable cost: material bought on the normal lead time, service in the scheduled window, a maintenance budget that finance can actually plan.
Longer asset service life: lubrication and adjustment kept up to date slow the degradation of the whole assembly, postponing capital replacements.
Safety and compliance: mandatory inspections (NR-13, NR-12, overhead cranes) fulfilled and documented, with traceability for audit.
The limit is equally clear: preventive maintenance does not see random failure, and preventive maintenance in excess destroys value: each unnecessary teardown costs a part with service life to spare, labor hours and the real risk of introducing a defect during reassembly. The goal is never "more preventive maintenance": it is the right preventive maintenance, in the right dose.
Types of preventive maintenance
By time (calendar): the intervention happens on fixed dates: monthly inspection, semiannual overhaul, annual shutdown. Simple to schedule, but it ignores how much the asset actually ran.
By usage (counter): the trigger is accumulated operation: operating hours, cycles, kilometers, tons produced. Fairer to real wear; it requires a reliable counter integrated with the system.
By opportunity: it takes advantage of shutdowns that are already going to happen (product changeover, general shutdown, breakdown of another asset on the line) to bring forward scheduled services with no extra window cost.
A mature plant mixes the three criteria, asset by asset: the compressor by operating hours, the overhead crane by calendar (a legal inspection requirement), the heat exchanger by opportunity shutdown.
How to build a preventive maintenance plan in 6 steps
1. Inventory and prioritize the assets
A complete register, with TAG, installation-location hierarchy and technical data. Next, criticality: the impact of each asset on safety, production, quality and cost. The plan is born from class A and B assets; low-criticality ones may, by documented decision, stay in planned corrective maintenance.
2. Identify the failure modes that matter
For each equipment family, the question is not "what does the manual tell me to do?", but "how does this asset fail in this operation?". Does the bearing dry out from contamination? Does the belt slip from misalignment? Does the panel overheat from a loose connection? Every preventive task in the plan should exist to block a specific failure mode. A task with no associated failure mode is a candidate for removal.
3. Set the interval by equipment family
The right interval comes from three sources, in this order of refinement:
| Source of the interval | When to use it | Limit |
|---|---|---|
| Manufacturer's manual | Starting point at implementation and for new assets, and a mandatory floor where there is a warranty or legal requirement | Assumes standard operating conditions, which are rarely yours |
| The plant's own failure history | Refine the interval after months of records: shorten it where it failed before the deadline, lengthen it where the part comes out intact | Requires disciplined failure recording to exist |
| Standard and legal requirement (NRs, insurer, environment) | Pressure vessels, boilers, overhead cranes, protection systems: the interval is mandatory | It is a floor, not a ceiling: real wear may demand more |
The honest rule of thumb: start from the manual, correct with the history. If the inspection always finds the belt intact for three cycles in a row, the interval is probably too short (unnecessary cost). If the failure arrives before the preventive service, the interval is too long (insufficient protection) or the task does not attack the right failure mode.
4. Standardize the task
Each plan needs to say what to do, with which tool, with which acceptance criterion and what to record. "Inspect pump" is not a task: it is an intention. "Measure the drive-end bearing temperature, record the value, alert above X degrees as defined in the equipment standard" is a task. The quality of the plan is measured by whether a new technician can carry it out without guessing.
5. Schedule with real capacity
A good plan in a system with unrealistic capacity turns into backlog: the orders are generated, do not fit into the week and grow old. Scheduling preventive maintenance requires looking at the team's available capacity, the production windows and the week's priorities, and deciding what goes in, what waits and what is rescheduled with judgment. Preventive maintenance chronically behind schedule is the first sign that the plan was sized for a team that does not exist: in that case, either capacity grows, or the plan is re-prioritized by criticality, consciously, before the delay decides on its own.
6. Close the loop with field reporting
The plan only learns if execution returns data: what was found, what was measured, what was replaced, how long it took. It is this feedback that feeds step 3 (refine the interval) and step 2 (discover new failure modes). Execution with no reporting is a plan running blind.
The minimum that each closed preventive order should record:
Confirmation of each task carried out (not the whole package in one block);
Measurements taken, with the value, not just "ok";
Condition of the component as found (intact, early wear, defect) to calibrate the interval;
Materials actually consumed;
Actual execution time, for the next scheduling round;
Abnormalities found outside the scope, converted into a notification for triage.
It is this level of recording that turns preventive maintenance into a source of data, and it is exactly what is lost when execution runs on a clipboard and the system receives a summary days later.
Examples of preventive maintenance by equipment family
The table below shows the format of reasoning by family: which preventive tasks typically make up the plan and where the interval for each one comes from. The intervals themselves should come from the manufacturer's manual and from your plant's history, never from a generic table on the internet (including this one):
| Equipment family | Typical preventive tasks | Basis for the interval |
|---|---|---|
| Electric motors | Bearing lubrication, insulation measurement, ventilation cleaning, retightening of connections | Operating hours (lubrication) and calendar (electrical) |
| Centrifugal pumps | Seal and packing inspection, alignment, noise and temperature analysis, replacement of wear elements | Operating hours + inspection findings |
| Compressors | Filter and oil changes, valve checks, safety testing, condensate draining | Operating hours (the asset's own counter) |
| Gearboxes | Oil analysis and replacement, coupling inspection, checking clearances and leaks | Operating hours + lubricant condition |
| Overhead cranes and hoists | Inspection of ropes, brakes, hooks and limit switches; mandatory periodic tests | Mandatory calendar (standard and legal requirement) |
| Electrical panels | Retightening, cleaning, checking for heating and protection devices | Calendar + available de-energization window |
| Belts and conveyors | Tensioning, alignment, inspection of splices and scrapers, roller lubrication | Short calendar (inspection) + wear history |
Notice the pattern: equipment with a reliable counter tends toward usage-based intervals; safety and lifting items are mandatorily by calendar; and fast-wearing components call for frequent inspection that converts a finding into planned corrective maintenance.
The trap of the copied plan (and how to get out of it)
The most common vice of industrial preventive maintenance is the inherited plan: copied from the manual, from another plant or from the previous version of the system, and never reviewed again. The symptoms are recognizable:
Generic tasks that cite neither a failure mode nor an acceptance criterion;
Intervals that are too round (everything monthly or semiannual, regardless of the equipment);
Preventive orders closed in minutes, with no measurement recorded (the famous "done, ok");
Failures repeating on assets whose preventive maintenance is rigorously up to date.
The last symptom is the most serious: preventive maintenance fulfilled with recurring failure means the plan is busy with the wrong tasks. The way out is not to do more preventive maintenance, it is to do the right preventive maintenance: cross the failure history with the plan's tasks, cut what blocks no failure mode and reinforce where the failure keeps getting through. Fewer, better tasks protect more than many generic tasks.
A practical way to start the review without a giant project: pick the ten assets with the most corrective maintenance in the past year and, for each one, answer three questions with the history in hand. Which failure modes actually occurred? Which plan task should have blocked them? Was that task carried out and reported? The answers separate the three possible diagnoses (wrong plan, right plan poorly executed, or failure beyond the reach of preventive maintenance) and each diagnosis has a different remedy. Reviewing ten assets per quarter already puts the plan into continuous improvement.
How to measure whether preventive maintenance is working
Preventive maintenance consumes labor hours every month; the management question is what it gives back. Four indicators answer:
Plan compliance: the percentage of preventive orders carried out on time. Below a healthy threshold, the plan is fiction and the other indicators do not even make sense.
Share of emergency corrective maintenance: if preventive maintenance protects, emergencies fall over the months. High preventive maintenance with high emergencies is the alarm of a wrong plan.
MTBF of critical assets: the mean time between failures should rise where the plan acts; it is the technical proof that the tasks block the right failure modes.
Findings per inspection: how many defects the inspections find and convert into planned corrective maintenance. An inspection that finds nothing for years is looking at the wrong item or at the wrong interval.
It is worth closing the account in money too: add up the labor hours and material consumed by preventive maintenance in the quarter and compare it with the evolution of emergency cost (express freight, overtime, downtime) over the same period. Healthy preventive maintenance is the kind that makes the second curve fall more than the first rises; when the two only grow together, the plan needs review, not reinforcement.
On critical assets where recording is already reliable, the next step is to evolve part of the plan toward predictive maintenance, trading fixed intervals for measured condition; and the reduction of emergency corrective maintenance is the result that justifies the entire program.
Preventive maintenance in SAP: from plan to execution
In a plant running SAP PM, the maintenance plan lives in plans and task lists that generate orders automatically through the defined strategy. The weak link is usually not order generation, but the return trip: the reporting of what was carried out and found in the field, which is what keeps the plan alive. When execution runs on paper and reporting is typed in days later (or never), SAP becomes an order generator that nobody feeds back. How to structure this cycle inside SAP, from strategy to reporting, is the subject of our guide to industrial preventive maintenance in SAP.
Frequently asked questions about industrial preventive maintenance
What is industrial preventive maintenance?
It is the set of interventions scheduled at predetermined intervals of time, operating hours or cycles (lubrication, inspection, calibration, scheduled replacements) to reduce the probability of equipment failure across a plant. NBR 5462 defines it as maintenance carried out at predetermined intervals or prescribed criteria to reduce the probability of failure or degradation.
What is the difference between preventive and corrective maintenance?
Preventive maintenance acts before the failure, at scheduled intervals, to keep it from occurring. Corrective maintenance acts after the failure, to restore the equipment's function. Preventive maintenance has a predictable, recurring cost; emergency corrective maintenance has an unpredictable and almost always higher cost, because it adds production downtime and urgency to the repair.
How do you set the interval for preventive maintenance?
Start from the manufacturer's manual and the legal requirements, which are the floor. Then refine with the plant's own history: shorten the interval where the failure arrives before the preventive service and lengthen it where the component is found intact cycle after cycle. On assets with a reliable counter, prefer usage-based intervals (hours, cycles) over pure calendar.
When does preventive maintenance become waste?
When the task blocks no real failure mode: replacing a part with service life to spare, an inspection that never finds anything, a generic task with no acceptance criterion. The clearest sign is a recurring failure on an asset whose preventive maintenance is up to date: the effort exists, but it is pointed at the wrong place. The answer is to review the plan against the history, not to add more tasks.
A maintenance plan only protects while it is alive: carried out on time, reported in the field and reviewed against the history. Get to know PM Run and turn your preventive plan into real execution in SAP, with mobile reporting and data that feeds the strategy back.
