How to Build a Reliability-Centred Maintenance Strategy

A maintenance strategy based on Reliability-Centred Maintenance (RCM) gives an organisation a disciplined way to decide what equipment needs, when it needs attention and how much maintenance is justified. Rather than treating every asset alike, RCM connects maintenance tasks to an asset’s purpose, likely failure modes, business consequences and acceptable risk.

For Australian operators, this approach is particularly useful across processing plants, energy facilities, utilities, mining support operations, manufacturing sites and renewable energy projects. It can reduce unplanned downtime while supporting obligations under workplace safety laws, environmental controls and engineering standards. The strongest results come when planners combine reliable data with the practical knowledge of operators, tradespeople and supervisors.

Define What Each Asset Must Do

RCM starts with functions, not equipment lists. A pump may be described as “a 45 kW centrifugal pump”, but its operational function is more useful: transfer 80 cubic metres of cooling water per hour at a specified pressure, within a defined temperature range, for a particular production system. This description establishes the performance standard that maintenance must protect.

Write each function with measurable requirements wherever possible. Include capacity, quality, speed, containment, availability, control response and safety performance. A relief valve, for example, must open at its set pressure, discharge safely and reseat without creating a continuing leak. A conveyor may need to move a specified volume while protecting personnel from entanglement and preventing material spillage.

Functional failures occur when an asset cannot meet one or more of those requirements. They may be total, such as a failed motor, or partial, such as a pump that still runs but cannot maintain pressure. Defining partial failures matters because degraded performance can increase energy consumption, cause quality problems or place extra load on connected equipment long before a complete breakdown occurs.

Map Failure Modes And Consequences

The next step is to identify how each functional failure can happen. Typical failure modes include bearing wear, seal degradation, corrosion, blocked strainers, electrical insulation breakdown, instrument drift, software faults, incorrect alignment and human error. The analysis should be specific enough to support a maintenance decision rather than simply stating “equipment failed”.

For each failure mode, record the likely cause, local effect, wider system effect and evidence that would reveal the problem. A leaking process line might result from external corrosion under insulation, an unsuitable gasket, vibration-induced fatigue or an incorrect installation. Each cause has a different inspection method, interval and corrective action. For assets handling chemicals or contaminated waste, material selection and containment deserve particular attention; guidance on chemical drainage pipes can help inform discussions about compatible pipework and installation practices.

Consequences determine priority. Categorise them under headings such as safety, environmental impact, production, product quality, legal compliance and maintenance cost. A failed standby fire pump has a different consequence from a failed non-critical workshop exhaust fan, even if both are electrically driven. This consequence-based view prevents scarce maintenance resources being spent evenly on assets with very different levels of risk.

Select The Right Maintenance Task

RCM does not assume that preventive maintenance is always the answer. For each significant failure mode, ask whether a technically feasible task can detect or prevent the failure at a useful point in time. If so, determine whether the task is worth doing based on risk, cost, access requirements and the consequences of missing the failure.

Condition-based maintenance is suitable when deterioration can be measured before functional failure. Vibration analysis can reveal developing bearing or coupling faults, thermography can identify abnormal electrical resistance or overloaded connections, and oil analysis can show contamination or wear particles. Ultrasonic testing may help detect compressed-air leaks, steam-trap faults or early bearing problems. These techniques work best when readings are trended against a known baseline rather than treated as isolated pass-or-fail results.

Scheduled restoration or replacement can be appropriate when a component has a predictable useful life and failure consequences are high. Examples include replacing a safety-critical battery at a verified service interval or overhauling a valve design known to wear in a particular duty. Run-to-failure is reasonable for some low-consequence, inexpensive and easily replaceable items, provided the failure creates no unacceptable safety, environmental or production risk.

Build A Practical Asset Hierarchy

A usable RCM programme depends on a clean asset hierarchy. Structure equipment from site to area, system, functional location, asset and maintainable component. Link each item to manufacturer details, drawings, spare parts, operating instructions, inspection points and relevant hazards. Without this structure, work orders become difficult to search and failure data cannot be compared reliably.

Begin with critical systems rather than attempting to analyse every asset at once. Prioritise emergency shutdown systems, pressure equipment, electrical distribution, hazardous-area equipment, rotating machinery that limits production, environmental control systems and assets with long replacement lead times. In Australia, the hierarchy should also make it easy to identify equipment covered by site procedures, statutory inspection requirements and state or territory work health and safety obligations.

Maintenance planners should agree on naming conventions, criticality scores and failure codes before loading large amounts of information into a computerised maintenance management system (CMMS). A consistent system might record the asset, failure mode, cause, observed condition, work completed, parts used, labour hours and post-maintenance test. This creates a feedback loop between the strategy and real operating experience.

Integrate People, Skills And Work Control

RCM is a team activity. A workshop involving an operator, mechanical fitter, electrician, instrument technician, reliability engineer, planner and safety representative will usually identify more credible failure mechanisms than an office-based review. Operators understand changes in noise, temperature, pressure, product quality and process behaviour that may not appear in a formal equipment history.

The strategy must also reflect how work is actually performed. A technically excellent inspection is of little value if access requires a lengthy shutdown, isolation is unreliable or the required instrument is unavailable. Each task should state the skill level, tools, permits, isolation requirements, expected duration, acceptance criteria and follow-up action. This is where hands-on training in a realistic live-process-plant environment can strengthen planning. HCF CATCH supports practical development across process, energy, engineering, electrical and health and safety disciplines, helping teams connect written procedures with field conditions.

Australian sites should account for competency and consultation expectations under applicable Work Health and Safety legislation, including the model WHS framework adopted with variations by states and territories. Electrical work, high-risk work, confined spaces, hazardous chemicals and work at height may require specific licences, authorisations or controls. A maintenance strategy that ignores competency requirements can create a safety exposure even when its technical recommendations are sound.

Set Intervals, Measures And Review Rules

Maintenance intervals should be based on evidence rather than habit. Use manufacturer information as a starting point, then test it against operating context, duty cycle, climate, contamination, loading, failure history and consequences. Equipment operating in the heat and dust of inland Australia may need a different inspection frequency from identical equipment in a controlled indoor environment near Melbourne or Brisbane. Coastal salt exposure around Perth, Newcastle or Darwin can also accelerate corrosion on structures, electrical enclosures and outdoor plant.

For condition-monitoring tasks, define the measurement method and alarm limits clearly. A vibration route should specify sensor location, machine state, speed and data quality. An infrared inspection should state the load conditions and temperature difference that trigger investigation. For intrusive inspections, specify what constitutes acceptable wear, cracking, corrosion, insulation condition or seal leakage. Vague instructions produce inconsistent decisions and weak trend data.

Measure whether the strategy is delivering value. Useful indicators include planned maintenance percentage, schedule compliance, emergency work, mean time between functional failures, mean time to repair, repeat failures, maintenance backlog and availability of critical systems. Track leading indicators such as overdue inspections and bad-actor equipment as well as lagging indicators such as lost production. Review the strategy after a major failure, process change, design modification, near miss or repeated deferral.

Roll Out The Programme In Controlled Stages

A staged implementation is more effective than launching hundreds of new tasks at once. Select a representative pilot area, complete the criticality assessment, run the RCM review and load the approved tasks into the CMMS. Test whether technicians can find the asset, understand the job plan, obtain the necessary parts and close the work order with useful information.

Remove low-value activities as well as adding new ones. Duplicate lubrication routes, calendar-based inspections with no failure-prevention benefit and repeated visual checks that never change a decision can consume capacity without improving reliability. The objective is a balanced maintenance mix: condition monitoring where deterioration is detectable, scheduled intervention where life is predictable, functional testing for hidden failures and corrective work where run-to-failure is acceptable.

Plan the transition around shutdowns, seasonal demand and workforce availability. In the Australian energy and resources market, specialist contractors and critical spares may have long lead times, while regional sites can face travel and accommodation constraints. Keep a risk register for unavailable parts, single-person dependencies, deferred work and temporary repairs. A practical strategy recognises these realities and provides contingency actions rather than assuming ideal supply and staffing conditions.

Compare Strategy Options And Keep Improving

No single maintenance method suits every asset. The decision should reflect failure behaviour, detectability, consequence, cost and operational context. A safety instrument may need regular proof testing even when it rarely operates, while a basic workshop fan may be safely repaired only after failure. The table below provides a concise starting point for selecting the appropriate approach.

Maintenance approach	Best suited to	Typical activities	Main limitation
Condition-based maintenance	Detectable deterioration with warning time	Vibration, thermography, oil analysis, ultrasonic checks	Requires competent analysis and reliable trends
Scheduled preventive maintenance	Predictable wear or mandated intervention	Replacement, overhaul, calibration, lubrication	Can cause unnecessary work if life is not predictable
Failure-finding maintenance	Hidden protective or standby functions	Proof tests, trip checks, alarm tests, emergency equipment tests	Requires clear acceptance criteria and controlled testing
Run-to-failure	Low-consequence, low-cost and accessible items	Corrective replacement after breakdown	Unsuitable where failure affects safety, environment or critical production
Design-out maintenance	Repeated or systemic failure modes	Material change, guard redesign, filtration upgrade, control modification	May require capital approval and engineering change control

Reliability improvement should follow the evidence produced by the programme. If the same coupling fails repeatedly, investigate alignment, baseplate stiffness, soft foot, lubrication and installation quality rather than simply shortening the replacement interval. If a filter blocks sooner than expected, assess contamination sources, process changes and filter selection. Root cause analysis is most valuable when it leads to a design, operating or maintenance change that prevents recurrence.

Treat the RCM strategy as a living management system. Revalidate criticality after production changes, review failure codes for accuracy, audit completed tasks in the field and involve technicians in updating job plans. Training providers such as HCF CATCH can support employers and apprentices with practical engineering, process safety and electrical capability, while facility tours and realistic plant-based learning can help teams discuss reliability decisions in a setting that resembles operational work.

A well-designed reliability-centred programme turns maintenance from a calendar of recurring jobs into a risk-informed system for protecting function. Australian employers can begin with one critical process, establish clear functions and consequences, involve the people who operate and maintain the equipment, and then expand using measured results. Contact HCF CATCH to explore practical training, apprenticeship pathways or an industry-focused learning session that can help develop the skills behind a stronger maintenance strategy.