What is maintenance management? The cycle, the roles and how to run it

Maintenance management means organising and controlling all maintenance on your installations: you set objectives, budget and responsibilities, and you make sure the work is planned, prepared, carried out, recorded, analysed and improved. The aim is for assets to perform their function reliably and safely at an acceptable cost. So it is less about the repair itself and more about the organisation around it.

What is maintenance management?

The European terminology standard EN 13306 describes maintenance management as all the management activities that set the objectives, strategy and responsibilities for maintenance and put them into effect through planning, control, supervision and improving methods, including the economic side. Since 2024 the discipline has had its own standard: EN 17948, Maintenance management and functions. It helps maintenance managers set up their process: policy, objectives, strategy, what you do in-house and what you outsource, functions and responsibilities, budget, supervision, communication and continuous improvement.

That makes maintenance management broader than handling work orders and narrower than asset management, which covers an asset's entire life cycle from investment to disposal. Set up well, maintenance management helps reduce the likelihood of unplanned downtime, supports safe working and makes costs explainable.

The maintenance management cycle: from planning to improvement

Maintenance work runs in a cycle of six steps, and what you learn in the last steps changes the first. The diagram below shows that cycle, with the notifications that feed into it and the urgent work that goes straight to execution.

1. 1Plan. Decide what work is needed in the coming period: preventive tasks from the maintenance plan, inspections, shutdowns and assessed notifications. Set capacity and budget against that.
2. 2Prepare. Make every job ready to execute: scope, steps, parts, tools and permit. Only ready work goes into the weekly schedule.
3. 3Execute. Technicians or contractors do the work according to the work order, with a last-minute risk assessment, lockout and a permit to work where needed. Safety belongs in this step, not beside it: according to EU-OSHA, around 10 to 15 per cent of fatal workplace accidents in several European countries in 2006 were related to maintenance work.
4. 4Record. When closing out the job, record what was found and done, how many hours it took, which parts were used and what the cause was. This is the raw material for everything that follows.
5. 5Analyse. Look at the history regularly: which assets cost the most, which failures come back and which tasks never find anything? A recurring failure is a candidate for a root cause analysis.
6. 6Improve. Adjust intervals, tasks and stock, or opt for a modification. That changes your maintenance concept, and with it the plan for the next round.

Notifications and failures come in from outside. What can wait goes to planning after assessment. What cannot wait goes straight to execution as urgent work, and then carries on through recording and analysis as normal. This is exactly where the cycle tends to break. The first three steps happen by themselves because the work has to be done, the last three only if someone gets time for them. Without that time, you keep fixing the same failures.

If you want to test your own process, EN 17007 is a useful tool. It describes the maintenance process as sub-processes with their inputs and outputs, so you can spot missing actions and responsibilities without an owner.

Roles and duties in a maintenance organisation

A cycle only works if each step has a clear owner. Many organisations have these six roles:

Operators sit outside this list, but they often notice problems first and raise many of the notifications. Stores often gets forgotten, even though a job without its part is not a job; spare parts management explains how to link the two. The European standard EN 15628 describes the competences for four roles (maintenance technician specialist, supervisor, maintenance engineer and maintenance manager) and is useful for job profiles and training plans.

In a small maintenance team these are hats, not job titles: the team leader schedules, prepares and orders. That works as long as it is clear who does what. Just make sure urgent work does not eat everything else, because someone dealing with today's failure is not preparing next week.

What does a maintenance manager do? The main duties

• Objectives and strategy: agreeing with senior management and production what maintenance should deliver and which assets get what attention.
• Budget: drawing up and monitoring the maintenance budget, including major shutdowns and replacements.
• Organisation: arranging roles, staffing and competences, and deciding what you do in-house and what you outsource.
• Safety and legislation: making sure safe working and statutory inspections are properly in place.
• Coordination: with production on downtime windows and with purchasing on parts and contracts.
• Performance: following KPIs, reporting on them and adjusting course.
• Improvement: freeing up time for analysis and improvement, even when there are failures to deal with.

Managing reliability, availability and maintenance costs

These three are often mentioned in the same breath, but they measure different things. Reliability says whether an asset does what it should over a given period, usually expressed as MTBF. Availability says whether the asset can be used at the moment you need it. According to the IEC vocabulary, it depends on reliability, maintainability and supportability, and on the maintenance and support provided. Maintenance costs are what you spend on hours, parts and contractors, but the bill is only complete with the cost of downtime.

This fictional example shows why you need all three. Two dosing pumps in a chemical plant are needed all year (8,760 hours) and are both down for 24 hours, so operating time is 8,736 hours. Downtime costs €2,000 an hour plus €1,500 per incident for restarting and rejects.

The availability is the same, but the problem is not. Pump A fails too often, which calls for a root cause analysis or a different maintenance type. Pump B takes too long to repair, which calls for a spare on the shelf or a better repair procedure. And because every failure also carries a fixed cost, pump A costs €16,500 a year more, which availability does not show.

In practice, that means:

• Agree what counts per asset class. For critical assets, safety and availability come first; for assets whose failure has little consequence, the cheapest approach can win.
• Look at total cost, not just the budget. Less maintenance looks good for a year and comes back later as downtime and consequential damage. Set maintenance costs against the replacement value of your assets, so you can compare years and sites.
• Keep the set small. Availability, MTBF, MTTR, share of urgent work and cost per asset are enough for many teams; maintenance KPIs explains how to choose.
• Link each figure to an owner and a meeting. A metric nobody acts on changes nothing.

A fixed rhythm makes that concrete:

How mature is your maintenance organisation?

These four levels help you work out where you stand. It is a working model for choosing your next step, not a certificate, and not every asset has to be at the top level.

The difference also shows up in the numbers. In a survey by the US institute NIST of discrete manufacturers, an average of 45.7 per cent of machinery maintenance was reactive. In the quarter of companies with the highest share of reactive maintenance, 13.0 per cent of planned production time was downtime, against 4.0 per cent in the quarter with the lowest share. That is an association for 2016 and not proof of cause, but it is a good reason to track your share of urgent work.

Three questions help you see where you stand. What share of hours went to urgent work last month? Can you show the failure history of your critical assets for the past year? When did you last drop a task or change an interval based on findings?

This is how to move up, one step at a time:

1. 1Run all work through a work order, including the small jobs. Otherwise your history is wrong and you do not know where the hours go.
2. 2Get your assets and criticality in order. You do not have to do everything: an accurate list of the critical assets is enough to start.
3. 3Create a preventive plan for those critical assets rather than half a plan for everything.
4. 4Introduce a fixed weekly meeting with production and keep a reserve for failures, based on your own figures.
5. 5Review the biggest sources of failures every month and tackle one properly.
6. 6Then expand with measurements on critical assets and a maintenance concept per asset group.

Where maintenance software helps

A spreadsheet or whiteboard can work for a small crew with a limited number of assets. It gets difficult once several people work on the same jobs at once, the history grows and you want figures per asset. A maintenance management system, or CMMS, then supports every step of the cycle:

• Plan: rounds on the calendar or on running hours that turn into work orders themselves, with crew capacity in view.
• Prepare: tasks, parts and documents linked to the work order.
• Execute and record: hours, materials and findings go straight onto the work order, so closing out is not an afterthought.
• Analyse and improve: history per asset and KPIs from the same data, without merging lists first.

Software records what you have agreed, but it does not make the agreements. Without a division of roles and rules for priorities, you get the same chaos, only digital. You can see what it looks like in practice, from notification to completed work order, under maintenance software.

Common pitfalls in maintenance management

• Looking only at the budget. Deferred maintenance looks like a saving for a year. The bill comes later, often bigger and at a bad moment.
• Scheduling without preparation. Scheduling work that is not ready means scheduling waiting time.
• Recording that nobody uses. Technicians stop filling things in if nothing ever comes of it. Show what a notification or a cause code has led to.
• No time for analysis and improvement. If all the hours go on day-to-day problems, the cycle never closes.
• Everything with one person. If the knowledge of an installation or the schedule sits in one head, things grind to a halt when that person is away.
• Leaving production out. Without agreements on downtime windows, planned work gets called off at the last minute.