What is preventive maintenance?
Preventive maintenance is the practice of servicing assets on a fixed schedule before failures occur. The schedule is based on time intervals, usage thresholds, or both. Oil changes every 10,000 kilometres. Bearing lubrication every 500 hours. Fire system inspections every six months. The asset is serviced whether it shows signs of wear or not.
The logic is straightforward. Components wear out over time and use. If you replace them at predictable intervals based on manufacturer recommendations and operating experience, you avoid the majority of in-service failures. The trade-off is that you will sometimes service components that still have useful life remaining, which means some maintenance spend is "wasted" on parts that did not yet need replacing.
Despite that trade-off, preventive maintenance is dramatically better than reactive maintenance. Organisations that shift from reactive to preventive programmes typically reduce unplanned downtime by 25 to 50 per cent and lower total maintenance costs by 20 to 40 per cent within the first year. The savings come from fewer emergency repairs, fewer catastrophic failures that damage related components, and the ability to schedule work during planned windows rather than in the middle of production.
Preventive maintenance is the backbone of most operational maintenance programmes. It works for every asset type, from vehicles to HVAC systems to production machinery, and does not require sophisticated monitoring technology. A CMMS or scheduling platform that tracks intervals and generates work orders automatically is all you need to run an effective preventive programme.
What is predictive maintenance?
Predictive maintenance takes a different approach. Instead of servicing on a fixed schedule, it uses data from the asset itself to determine when maintenance is actually needed. The premise is that most failures give warning signs before they occur: increasing vibration, rising temperature, degraded oil quality, unusual noise, or declining performance. If you can detect those signs early enough, you can intervene before the failure happens.
The data comes from various sources depending on the asset and the failure mode you are trying to catch.
- Vibration analysis detects bearing wear, shaft misalignment, and imbalance in rotating equipment like motors, pumps, and compressors.
- Thermal imaging identifies hot spots in electrical systems, bearings, and mechanical components that indicate excessive friction or poor connections.
- Oil analysis measures contamination, metal particles, and degradation in lubricants, revealing internal wear before it becomes visible externally.
- Ultrasonic testing detects leaks, bearing defects, and electrical discharge that are inaudible to the human ear.
- Performance monitoring tracks output metrics like flow rate, pressure, and temperature against baseline values to identify gradual degradation.
The goal of predictive maintenance is to maintain assets at exactly the right time, not too early (wasting parts and labour) and not too late (risking failure). When implemented well, it delivers the most efficient maintenance spend possible because every intervention is justified by data, not by a calendar.
The trade-off is complexity and cost. Predictive maintenance requires monitoring equipment, data collection systems, and the skills to interpret the results. For some assets, the investment is clearly worth it. For others, the cost of monitoring exceeds the savings it would generate.
Key differences compared
Understanding the practical differences between preventive and predictive maintenance helps you decide where to apply each approach.
Trigger mechanism. Preventive maintenance is triggered by time or usage, regardless of condition. Change the oil every 10,000 km whether it needs it or not. Predictive maintenance is triggered by condition data. Change the oil when analysis shows it has degraded beyond acceptable levels. The predictive approach is more precise but requires the data to make the decision.
Setup cost. Preventive maintenance is straightforward to set up. Define intervals, configure a scheduling system, and generate work orders. Predictive maintenance requires an upfront investment in sensors, monitoring systems, and potentially specialist technicians or analysts. The setup cost for a predictive programme on a single machine can range from a few hundred dollars for basic vibration monitoring to tens of thousands for a full condition monitoring system.
Maintenance precision. Preventive maintenance sometimes results in over-maintenance (replacing parts that still have life) or under-maintenance (intervals too long for actual operating conditions). Predictive maintenance targets interventions more precisely, servicing the asset when data indicates the need. This can extend component life by 20 to 40 per cent compared to fixed-interval schedules.
Skill requirements. Preventive maintenance requires standard trade skills. Technicians follow a checklist and perform prescribed tasks. Predictive maintenance often requires additional skills: vibration analysts, thermographers, oil analysis interpretation, and data analysis capabilities. For small teams, this skill gap can be a barrier.
Asset coverage. Preventive maintenance applies to virtually every asset. Predictive maintenance is most cost-effective for critical, high-value assets where the cost of failure is high and the cost of monitoring is justified. Most organisations apply predictive techniques to 10 to 20 per cent of their asset base, the critical few, and use preventive schedules for the rest.
When preventive maintenance is the right choice
Preventive maintenance is the right default for most organisations and most assets. It works well in the following situations.
Consistent, predictable wear patterns. Components that degrade at a known rate, like oil, filters, belts, brake pads, and tyres, respond well to fixed-interval replacement. The manufacturer has already determined the expected life, and your operating experience refines it. There is no need to monitor the condition of an oil filter when you know it should be changed every 500 hours.
Regulatory and compliance requirements. Many inspections and services are mandated by regulation regardless of asset condition. Fire system testing, vehicle roadworthiness inspections, test and tag compliance, and equipment certifications must happen on a fixed schedule. These are inherently preventive and cannot be replaced by condition monitoring.
Large fleets of similar assets. When you manage 50 identical vehicles or 200 similar machines, applying the same preventive schedule across the fleet is efficient and manageable. The statistical law of large numbers works in your favour: while some individual assets may be over-serviced, the fleet average delivers a reliable outcome.
Teams with limited monitoring capability. If your maintenance team does not have access to vibration analysers, thermal cameras, or oil analysis labs, predictive maintenance is not practical. Preventive maintenance works with the skills and tools your team already has. A maintenance management platform that automates scheduling and tracks completion is the primary tool required.
Low-value assets. The cost of monitoring a $2,000 pump with a $5,000 condition monitoring system does not make financial sense. For low and mid-value assets, a preventive schedule based on manufacturer recommendations is the most efficient approach.
When predictive maintenance makes sense
Predictive maintenance delivers the most value in specific situations where the cost of monitoring is outweighed by the savings from precisely timed interventions.
Critical assets where failure is extremely costly. A main production line motor, a mine site crusher, a hospital HVAC system, or a data centre cooling unit. When failure of a single asset can halt operations, cause safety incidents, or generate losses measured in tens of thousands of dollars per hour, the investment in condition monitoring pays for itself quickly.
Assets with expensive components. If a bearing replacement costs $500 but a catastrophic bearing failure that damages the shaft, housing, and seals costs $15,000, detecting the bearing degradation early through vibration analysis is a clear win. Predictive maintenance is most valuable when early detection prevents secondary damage.
Long-lead-time parts. If a critical component takes six weeks to source, you need early warning of its degradation so you can order the part before the failure occurs. Predictive monitoring gives you that lead time. Preventive schedules would require keeping the part in stock, tying up capital in spares inventory.
Assets with variable usage patterns. Equipment that operates in widely varying conditions, such as a generator that runs 20 hours one week and 200 hours the next, does not respond well to fixed-interval schedules. Condition monitoring accounts for actual operating stress and triggers maintenance when the asset genuinely needs it, not on an arbitrary calendar.
Mature maintenance organisations. Predictive maintenance builds on a foundation of good preventive maintenance. If your organisation already has reliable asset data, consistent work order completion, and disciplined maintenance scheduling, adding predictive techniques to your critical assets is a natural evolution. If you are still struggling with basic scheduling and work order compliance, start by getting preventive maintenance right first.
Building a hybrid approach
The most effective maintenance programmes use both preventive and predictive strategies, applied to different assets based on criticality, value, and monitoring feasibility. Here is how to build a hybrid approach.
Classify your assets by criticality. Rank every asset based on the operational impact of its failure. Critical assets, those whose failure would halt production, create safety risks, or generate significant financial losses, are candidates for predictive monitoring. Non-critical assets stay on preventive schedules.
Start predictive on a small scale. Choose two to three critical assets and implement basic condition monitoring. Vibration analysis for rotating equipment is often the easiest starting point because the technology is mature, relatively affordable, and does not require invasive installation. Run the monitoring alongside your existing preventive schedule for six months to build baseline data and validate the approach.
Use inspection data as a predictive input. You do not always need sensors for predictive insights. Structured digital inspections with condition-rating fields (good, fair, poor) and photo documentation create a dataset that reveals degradation trends over time. If an inspection consistently rates a component as "fair" over three consecutive checks, that is an early warning that scheduled intervention is needed.
Keep preventive as the baseline. Even with predictive monitoring on critical assets, preventive maintenance remains the backbone for routine servicing (oil, filters, belts), compliance inspections, and the 80 per cent of assets that do not justify monitoring investment. A robust CMMS that automates preventive scheduling is essential regardless of how far you take predictive techniques.
Review and adjust. After six to twelve months, review the results. Did predictive monitoring catch failures early? Did it reduce over-maintenance on the monitored assets? Use the data to decide which additional assets warrant monitoring and which preventive intervals should be adjusted based on condition data. A hybrid programme is never finished; it evolves as you collect more data and learn more about your assets.
How MapTrack supports both strategies
MapTrack provides the scheduling and tracking foundation that both preventive and predictive programmes require.
Preventive scheduling. Define maintenance plans by time, distance, or meter readings. The platform generates work orders automatically when thresholds are crossed and escalates overdue items. Every completed service is recorded against the asset, building the service history that supports compliance and repair-versus-replace decisions.
Condition data from inspections. Structured inspection checklists capture condition data in the field: condition ratings, photos, defect notes, and meter readings. Over time, this data reveals trends that inform predictive decisions. A component rated "fair" on three consecutive inspections can trigger a maintenance task before it reaches "poor".
Meter-based triggers. For assets equipped with GPS or telematics, MapTrack reads engine hours and odometer data in real time. This supports meter-based maintenance scheduling that responds to actual usage, a step toward condition-based maintenance without requiring dedicated sensors.
Complete asset visibility. Whether you are running a purely preventive programme or a hybrid that includes predictive elements, MapTrack gives you a single view of every asset's maintenance status, upcoming services, condition history, and location. That visibility is the foundation for any maintenance strategy.
If your team is evaluating its maintenance approach and wants a platform that supports both preventive scheduling and condition-based workflows, book a demo to see how MapTrack fits your operation.
