Why maintenance types matter
Every piece of equipment in your operation will eventually need attention. The question is not whether you will spend money on maintenance, but how and when. Choose the wrong approach and you either bleed cash on unnecessary servicing or suffer the chaos and cost of unplanned breakdowns. Choose the right approach, matched to each asset's criticality, and you control costs while keeping your operation running.
Understanding the different types of maintenance is not an academic exercise. It is the foundation of every decision your maintenance team makes: what gets serviced today, what gets monitored, what gets left alone. Operations that apply a single strategy to every asset, typically reactive maintenance by default, consistently overspend and underperform compared to those that deliberately match strategy to asset.
This guide covers five maintenance types from the simplest to the most sophisticated. Each has a legitimate place. The goal is not to adopt one exclusively, but to understand where each fits so you can build a blended maintenance programme that works for your operation.
Research from the Australian Maintenance Engineering Society of Australia (MESA) indicates that organisations with a structured, blended maintenance approach achieve 25 to 40 per cent lower maintenance costs than those relying primarily on reactive methods. The savings come from fewer emergency repairs, longer asset life, reduced spare parts inventory and, critically, less unplanned downtime.
Corrective (reactive) maintenance
Corrective maintenance, also called reactive or run-to-failure maintenance, is the simplest approach. You do not touch the asset until it breaks. No scheduled servicing, no inspections, no monitoring. When it stops working, you repair or replace it.
This sounds irresponsible, but for certain assets it is the most rational choice. If the asset is cheap to replace, carries no safety risk when it fails, does not affect production, and degrades in a way that is not detectable before failure (think light globes or basic hand tools), the cost of preventive servicing may exceed the cost of periodic replacement.
When corrective maintenance makes sense
- Low replacement cost. The asset costs less to replace than it would to inspect and maintain on a schedule.
- No safety impact. Failure does not create a hazard for personnel or the environment.
- No production impact. The asset is not in the critical path of your operation. If it fails, work continues.
- Redundancy exists. A backup is available, so failure of one unit does not cause a stoppage.
When it becomes a problem
The danger is when corrective maintenance becomes the default for assets where it should not be. A critical pump that feeds an entire processing circuit, a crane on a construction site, a refrigerated transport unit, these are assets where unplanned failure carries serious consequences: lost production, safety incidents, contract penalties, spoiled inventory.
The hidden cost of over-reliance on reactive maintenance is operational chaos. When breakdowns are unplanned, technicians are pulled from scheduled work. Parts need to be sourced urgently, often at premium cost. Overtime hours spike. The planned work that gets deferred creates a backlog that triggers more breakdowns, feeding a cycle that consumes the entire team's capacity. If this sounds familiar, the guide on moving from reactive to preventive maintenance covers how to break out of it.
Preventive maintenance
Preventive maintenance (PM) is scheduled servicing at fixed intervals. Change the oil every 250 hours. Inspect the belts every month. Replace the bearings every 12 months. The schedule is based on time, usage (such as kilometres or engine hours) or manufacturer recommendations, and it happens regardless of the asset's current condition.
This is the workhorse of most maintenance programmes, and for good reason. PM is straightforward to plan, easy to resource and does not require sensors or advanced analytics. You know what needs doing and when. That predictability lets you order parts in advance, schedule labour efficiently and minimise disruption to operations.
Advantages of preventive maintenance
- Reduces unplanned breakdowns by 30 to 50 per cent compared to reactive-only approaches.
- Extends asset lifespan through timely servicing and component replacement.
- Provides a structured framework for compliance with AS/NZS standards and WHS regulations.
- Creates audit-ready service records when managed through a CMMS platform.
The trade-off: over-servicing
The main drawback is over-servicing. A bearing that could run for 18 months gets replaced at 12 because the schedule says so. Oil gets changed at 250 hours even though analysis would show it is still within specification at 350. Over-servicing adds up: unnecessary parts consumption, unnecessary labour hours and unnecessary downtime for servicing.
Despite that, preventive maintenance is the right strategy for the majority of assets in most organisations. If you are currently running mostly reactive, start here. A solid preventive programme covers 70 to 80 per cent of what most operations need. Our step-by-step guide to building a PM programme walks through implementation from scratch.
Predictive maintenance
Predictive maintenance (PdM) uses data to forecast when an asset will fail, then schedules intervention just before that point. Instead of servicing at fixed intervals, you service when the data tells you the asset needs it.
The data comes from various sources: vibration sensors on rotating equipment, oil analysis for engines and gearboxes, thermography for electrical systems, ultrasonic testing for structural components. Each technique measures a specific indicator of degradation. When the indicator crosses a defined threshold, a work order is triggered.
Key predictive technologies
- Vibration analysis: detects imbalance, misalignment and bearing wear in rotating equipment. Common in pumps, motors and conveyor drives.
- Oil analysis: measures particle contamination, viscosity breakdown and chemical composition. Essential for engines, gearboxes and hydraulic systems.
- Thermography: infrared imaging identifies hotspots in electrical panels, bearings and insulated systems before visible damage occurs.
- Ultrasonic testing: detects leaks, bearing defects and structural cracks through high-frequency sound analysis.
Studies consistently show that predictive maintenance reduces maintenance costs by 25 to 30 per cent compared to preventive-only programmes, and cuts unplanned downtime by 30 to 50 per cent. The barrier is investment: sensors, data infrastructure, analytical software and people with the skills to interpret the data. For a deeper look at implementation, our predictive maintenance starter guide covers technology selection and ROI calculation.
Where predictive maintenance fits
Predictive maintenance is most valuable for high-cost, high-criticality assets where unplanned failure is expensive and where failure modes are detectable through monitoring. A $500,000 haul truck engine justifies the investment. A $2,000 workshop pump probably does not. PdM complements preventive maintenance; it does not replace it entirely.
Condition-based maintenance
Condition-based maintenance (CBM) is closely related to predictive maintenance, but the distinction matters. Where predictive maintenance forecasts future failure using algorithms and trend analysis, condition-based maintenance responds to current condition. When a measured parameter (temperature, vibration, pressure, fluid level) moves outside its normal range, maintenance is triggered.
Think of it as a traffic light system. Green means the asset is operating within normal parameters, no action needed. Amber means a parameter has shifted and should be investigated at the next opportunity. Red means intervention is required now.
Practical CBM examples
- Monitoring hydraulic fluid contamination levels in excavators and servicing when contamination exceeds a threshold.
- Tracking bearing temperatures on conveyor systems and scheduling replacement when temperatures trend upward.
- Performing regular oil sampling on generator sets and replacing oil based on actual degradation rather than a calendar.
- Using digital pre-start inspections to capture operator-reported condition data that triggers work orders automatically.
CBM does not require the forecasting algorithms that predictive maintenance uses. It requires monitoring (sensors, inspections or regular testing) and defined thresholds for action. This makes it more accessible for organisations that are not ready for full predictive analytics but want to move beyond calendar-based preventive schedules.
Reliability-centred maintenance
Reliability-centred maintenance (RCM) is not a maintenance technique. It is a structured decision framework for deciding which technique to apply to each asset. Originally developed for commercial aviation in the 1970s, RCM analyses every asset and its failure modes to determine the most cost-effective maintenance strategy.
The RCM process asks a series of questions for each failure mode: what is the function of this component? How can it fail? What happens when it fails? What is the consequence (safety, environmental, operational, economic)? Is the failure detectable before it occurs? What maintenance task can prevent or mitigate the failure?
RCM outputs
The output is a tailored maintenance programme where every task has a documented justification. Some assets get preventive schedules. Some get condition monitoring. Some get a redesign to eliminate the failure mode entirely. And some, where the consequence of failure is negligible, are deliberately run to failure.
RCM is resource-intensive to implement. A full analysis of a complex plant can take months and requires cross-functional input from maintenance, operations and engineering. For large mining operations, processing plants and fleet-heavy logistics businesses, the investment pays off. For a small construction business with 50 assets, it is likely overkill.
Even if you do not conduct a formal RCM study, the underlying principle is worth adopting: match the maintenance strategy to the asset, based on criticality and failure consequences. Not everything needs the same level of attention. Australian Standard AS IEC 60300.3.11 provides the formal methodology for those who want to implement RCM rigorously.
Comparison table
The following table summarises the five maintenance types by their key characteristics. Use it as a quick reference when evaluating which strategy to apply to a given asset or asset group.
| Type | Trigger | Best for | Relative cost | Complexity |
|---|---|---|---|---|
| Corrective | Asset failure | Low-cost, non-critical assets | Lowest upfront, highest over time | Very low |
| Preventive | Time, usage or calendar | Most assets (foundation strategy) | Moderate | Low |
| Predictive | Sensor data and forecasts | High-value, critical assets | Higher setup, lowest over time | High |
| Condition-based | Parameter threshold breach | Assets with measurable degradation | Moderate to high | Medium |
| Reliability-centred | Risk and criticality analysis | Large, complex asset portfolios | High setup, optimised over time | Very high |
Choosing the right blend
No operation of any complexity runs a single strategy across every asset. The most effective approach is a deliberate blend that matches strategy to criticality. Here is a practical framework for deciding.
Step 1: Classify your assets by criticality
Group assets into three tiers. Critical assets are those where failure stops production, creates a safety hazard or triggers regulatory non-compliance. Important assets affect efficiency but do not halt operations. Non-critical assets have minimal impact when they fail.
Step 2: Match strategy to tier
- Critical assets: Predictive or condition-based maintenance, backed by a preventive schedule as a safety net.
- Important assets: Preventive maintenance with defined intervals based on manufacturer guidance adjusted for your operating conditions.
- Non-critical assets: Corrective maintenance, with a conscious decision documented for audit purposes.
Step 3: Implement with the right tools
A blended strategy requires a system that supports multiple trigger types: time-based, meter-based, condition-based and manual. MapTrack's maintenance module handles all four, so you can manage preventive schedules alongside condition triggers in a single platform. Work orders flow from whatever trigger fires first.
Step 4: Measure and refine
Track maintenance KPIs to verify your blend is working. Key indicators include mean time between failures, planned versus unplanned maintenance ratio, and maintenance cost per asset. If unplanned work is still dominating, either the preventive intervals are too wide or assets that should be on condition-based monitoring are still being run to failure.
Australian regulatory context
In Australia, WHS regulations require duty holders to maintain plant and equipment in a safe condition. The Work Health and Safety Regulation 2017 specifically addresses maintenance obligations for plant. Choosing the right maintenance type is not just an economic decision; it is a compliance requirement. For heavy vehicles, the National Heavy Vehicle Regulator (NHVR) mandates regular roadworthiness inspections that form part of your preventive schedule. Meeting AS/NZS 4024.1 (safety of machinery) and AS 3788 (pressure equipment inspection) may require specific inspection intervals regardless of your chosen strategy.
The right maintenance blend evolves as your operation matures. Start with a strong preventive foundation, add condition-based monitoring for your most critical assets, and use reporting and analytics to identify where the current strategy is falling short. If you are still operating primarily on reactive maintenance, book a demo to see how MapTrack helps teams build structured maintenance programmes that actually get followed.
