Maintenance is a planned approach to keeping equipment, buildings, or systems running efficiently for as long as possible. It involves inspecting, repairing, and replacing components before or after they fail, depending on the strategy you choose. There are four primary maintenance strategies: reactive, preventive, predictive, and reliability-centered maintenance. Each takes a fundamentally different approach to managing equipment failures, and choosing the right one (or the right combination) can cut operating costs by 12 to 18 percent or more.
Reactive Maintenance
Reactive maintenance is the simplest strategy: you run equipment until it breaks, then fix it. Sometimes called “run to failure,” this approach requires no upfront planning, no sensors, and no scheduled inspections. For low-cost equipment that’s easy to replace or repair, this can actually be the most cost-effective choice. A cheap desk fan or a light bulb, for example, isn’t worth scheduling preventive inspections for.
The problem is that reactive maintenance gets expensive fast when applied to critical equipment. Emergency repairs cost 25 to 30 percent more than planned ones. Emergency labor runs two to three times the normal rate, after-hours call-outs add a 50 to 100 percent surcharge, and rush-ordering replacement parts tacks on another 25 to 50 percent. Add the cost of unplanned downtime while you wait for parts and personnel, and the bill climbs quickly.
Reactive maintenance breaks into two subcategories. Deferred corrective maintenance is when something fails but the repair can wait. The equipment is down or degraded, but it’s not an emergency, so the work gets scheduled alongside other tasks. Emergency maintenance is the opposite: the failure is urgent enough that everything else gets pushed aside to deal with it immediately. When choosing reactive maintenance for any piece of equipment, the key question is whether that equipment’s failure modes could escalate into emergencies. If they can, a proactive strategy is almost always cheaper in the long run.
Preventive Maintenance
Preventive maintenance means servicing equipment on a schedule, before it fails. This is the most widely used proactive strategy, and it works on a simple principle: components wear out in roughly predictable patterns, so replacing or inspecting them at set intervals prevents most breakdowns. Preventive programs typically deliver around four times the return on investment compared to reactive approaches, through fewer failures, lower energy consumption, and longer equipment life.
Time-Based Preventive Maintenance
Time-based (or calendar-based) maintenance triggers service at fixed intervals regardless of how much the equipment has actually been used. An HVAC system might get inspected every three months. A delivery vehicle might get an oil change every six months. These intervals stay the same whether the equipment runs heavily or barely at all.
The advantage is simplicity. You need minimal data tracking, and scheduling is straightforward. The downside is that fixed intervals often lead to over-servicing equipment that’s been sitting idle, or under-servicing equipment that’s been running hard. For assets with predictable, steady workloads, time-based maintenance works well. For variable workloads, it wastes money.
Usage-Based Preventive Maintenance
Usage-based maintenance ties service intervals to actual operational metrics: running hours, production cycles, mileage, or similar measures. A machine might be serviced after every 1,000 operating hours. A conveyor belt might get inspected every 200,000 cycles. Because maintenance is triggered by real wear rather than calendar dates, this approach more closely matches the equipment’s actual condition.
Usage-based scheduling requires accurate tracking, which adds complexity. You need meters, counters, or software that records how much each asset has been used. But the payoff is lower risk of both over-servicing and under-servicing. For equipment with variable workloads or usage-driven wear patterns, this approach is significantly more efficient than calendar-based scheduling.
Predictive Maintenance
Predictive maintenance uses real-time data to forecast when a failure is likely to occur, so repairs can be scheduled at exactly the right moment. Rather than servicing equipment on a fixed schedule (which might be too early or too late), predictive maintenance monitors the equipment’s actual health and acts only when the data shows deterioration is approaching a critical point.
The monitoring tools vary depending on the equipment. Vibration sensors detect imbalances in rotating machinery. Thermal imaging spots overheating components. Oil analysis reveals microscopic metal particles that signal internal wear. Acoustic sensors pick up unusual sounds from bearings or gears. Modern systems feed all of this data into AI-powered software that can detect subtle patterns and emerging faults, often catching problems weeks before they’d become visible to a technician.
The upfront investment in sensors and software is higher than for basic preventive maintenance, but the precision pays off. You replace components near the end of their useful life rather than on an arbitrary schedule, which means you extract maximum value from every part while still avoiding unexpected failures.
Condition-Based Maintenance
Condition-based maintenance (CBM) is closely related to predictive maintenance, and the two terms are sometimes used interchangeably. The distinction is that CBM focuses specifically on defined thresholds. Sensors continuously monitor parameters like vibration, temperature, or oil quality and compare readings against established baselines. When a reading crosses a preset threshold, a maintenance event is triggered.
A typical CBM system uses two threshold levels. An early warning alert fires when readings reach roughly two standard deviations above the normal baseline. An action threshold fires at three standard deviations, meaning the equipment needs immediate attention. When connected to a computerized maintenance management system, these threshold breaches can automatically generate work orders with the correct priority, assigned technician, and relevant sensor data attached. No manual handoff required.
The core benefit is timing. Maintenance happens neither too early (wasting remaining component life) nor too late (risking a catastrophic failure). For expensive or critical assets, this precision is worth the investment in sensors and monitoring infrastructure.
Reliability-Centered Maintenance
Reliability-centered maintenance (RCM) isn’t a single technique but a decision-making framework. Instead of applying the same strategy to everything, RCM analyzes each piece of equipment individually and assigns the most appropriate maintenance approach based on its failure modes and consequences.
The analysis works through a series of core questions for each system: What does the equipment do, and what are its functions? What functional failures are likely to occur? What are the consequences of those failures? And what can be done to reduce the probability of failure, detect the onset of failure, or reduce its consequences? The answers determine whether a given asset gets reactive maintenance, preventive maintenance, predictive monitoring, or some combination.
RCM is particularly valuable for complex operations with hundreds or thousands of assets. Not everything warrants expensive sensor arrays and AI monitoring. Some equipment is cheap enough to run to failure. Other equipment is so critical that it needs continuous monitoring and redundant backups. RCM gives you a structured way to make those decisions rather than guessing.
Total Productive Maintenance
Total productive maintenance (TPM) takes a broader organizational approach. Rather than leaving all maintenance to a dedicated team, TPM distributes responsibility across the entire workforce. It’s built on eight pillars, each targeting a different aspect of equipment reliability.
- Autonomous maintenance puts routine tasks like cleaning, lubricating, and basic inspection in the hands of the operators who use the equipment daily. They know their machines best and can spot emerging problems earliest.
- Planned maintenance schedules tasks based on predicted or measured failure rates, reducing unplanned downtime and allowing repairs during periods when equipment isn’t needed for production.
- Focused improvement targets specific quality problems with dedicated projects aimed at eliminating root causes of defects.
- Early equipment management feeds real-world maintenance knowledge back into the design of new equipment, so future machines are easier to maintain from the start.
- Quality maintenance builds error detection and prevention directly into production processes.
- Training and education fills knowledge gaps at every level. Operators learn to maintain their own equipment, maintenance staff learn proactive techniques, and managers learn coaching and TPM principles.
The goal is a culture shift: everyone takes ownership of equipment health rather than treating it as someone else’s problem. Organizations that implement TPM well see fewer breakdowns, less waste, and faster new equipment startups because practical experience shapes the entire maintenance lifecycle.
Measuring Maintenance Performance
Two metrics sit at the center of any maintenance program. Mean time between failures (MTBF) measures how long equipment typically runs before breaking down. The formula is simple: total operating time divided by the number of failures. A higher MTBF means more reliable equipment. Mean time to repair (MTTR) measures how quickly you can get a failed asset back online: total downtime from failures divided by the number of failures. A lower MTTR means faster recovery.
There’s no universal “good” number for either metric, because acceptable values depend on the type of equipment, its operating environment, and the maintenance strategy in place. What matters is tracking both over time. If MTBF is trending upward and MTTR is trending downward, your maintenance program is working. If not, something needs to change.
The Cost Difference Between Reactive and Proactive
The financial case for proactive maintenance is well documented. Reactive programs cost 25 to 30 percent more than preventive ones, driven primarily by emergency labor premiums, rush parts costs, and unplanned downtime. On the other side of the ledger, preventive maintenance programs cut operating expenses by 12 to 18 percent, reduce energy consumption by 10 to 20 percent through properly tuned systems, and lower administrative overhead by 40 to 60 percent when paired with automation that eliminates manual triage and paperwork.
There are also indirect savings. Documented preventive maintenance supports regulatory compliance and can lower insurance premiums by 5 to 15 percent. Energy-efficient equipment running on a proper maintenance schedule consumes 10 to 20 percent less power than neglected equipment. These savings compound over the life of an asset, which is why most organizations aim for a maintenance mix that’s heavily weighted toward proactive strategies, with reactive maintenance reserved only for low-consequence, low-cost equipment.