What a reliability engineer does, and what the role is not
The role between what the plant records and who changes it: what a reliability engineer reads, what they decide, who gets each decision, where the role stops against the planner, scheduler and supervisor, the rhythm it runs on, the skills and certifications behind it, and the ways the role is lost.
Key takeaways
- A reliability engineer works out why equipment fails and what to change so it fails less. The role works in months and years, from data, and hands its decisions to the people who change the plant.
- It reads failure history, condition data, downtime, cost and criticality, and it produces the maintenance strategy, the investigations, the spares decisions, the condition monitoring design and the reading of the measures.
- It is not the planner, who decides how a job is done, nor the scheduler, who decides when, nor the supervisor, who runs the day. A reliability engineer pulled into the day stops being one.
- The role runs on a rhythm: the week's failures, the month's bad actors, the quarter's program review, the year's criticality and strategy refresh, and the investigations as they are triggered.
- The role is only as good as the failure data it reads. Half of the job in the first year is getting work orders closed with hours, findings and failure codes.
Every site has someone whose job title says reliability. On some sites that person owns the maintenance strategy, runs the investigations, decides the spares, and can say which three items of equipment will cost the most next year and why. On others the same title belongs to whoever picks up the engineering nobody else has time for, and the strategy is whatever the vendor manual said when the plant was built.
This guide sets out what the role is when it is done properly: what it reads, what it decides, who it hands its decisions to, where its boundaries with the planner, the scheduler and the supervisor sit, the rhythm it runs on, the skills and certifications behind it, and the ways the role is lost. It is written for the engineer stepping into the title, and for the manager deciding what to ask of them.
The role in one picture
The role sits between the plant's records and the people who change the plant. It reads what is recorded, failure history, condition data, downtime and cost, criticality and findings, and it produces decisions: the strategy, the investigations and their actions, the spares, the condition monitoring design and the reading of the measures. It then hands each decision to the function that acts on it, the planner, engineering, the store, the maintenance manager, and watches the data to see whether the change worked.
That shape is why the role needs distance from the day. The reliability engineer who is also fixing today's breakdown reads no data and changes no strategy, and next year's breakdowns are this year's.
Reads
What the plant records
- Failure history. Work orders with failure codes, by equipment
- Condition data. Vibration, oil, thermography, process trends
- Downtime and cost. Hours lost by cause, maintenance cost by asset
- Criticality. Which equipment matters most, and why
- Findings. What the tradespeople saw and the planners could not fix
↓→
Decides
Why it fails, and what to change
- Which failure modes are managed, by what kind of task, how often
- Which failures earn an investigation, and what the cause was
- Which spares are held, and which risks are accepted
- What the measures mean, read as a set
In months and years, from data
↓→
Hands to
Who changes the plant
- The maintenance strategy to the planner, who loads it as maintenance plans
- Investigations and actions to engineering, operations, the planner
- Spares decisions to the store, the asset owner
- Condition monitoring design to the condition monitoring provider, the control engineer
- The measures, read to the maintenance manager, the Monday meeting
Never the planner, never the fitter
| Reads | What it is | From |
|---|---|---|
| Failure history | Work orders with failure codes, by equipment | The CMMS, closed properly |
| Condition data | Vibration, oil, thermography, process trends | The condition monitoring program and the historian |
| Downtime and cost | Hours lost by cause, maintenance cost by asset | Downtime tracking, the ledger |
| Criticality | Which equipment matters most, and why | The criticality assessment |
| Findings | What the tradespeople saw and the planners could not fix | Checklists, job feedback, the backlog review |
What it decides, and who gets it
The outputs are decisions with owners on the other end. The maintenance strategy, which failure modes are managed by what kind of task and at what interval, with the basis recorded, goes to the planner to load as maintenance plans. The investigations on the bad actors go to whoever owns the latent cause, which is usually a document, a design or a process. The spares decisions go to the store and the asset owner. The condition monitoring design goes to the technicians and the control engineer. And the measures, read as a set, go to the maintenance manager with what they mean and what to do.
| Produces | What it is | Handed to |
|---|---|---|
| The maintenance strategy | Which failure modes are managed, by what kind of task, at what interval, with the basis recorded | The planner, who loads it as maintenance plans |
| Investigations and actions | Root cause analyses on the bad actors, with actions that change a document, a design or a process | Engineering, operations, the planner |
| Spares decisions | Which spares are held, in what quantity, and why | The store, the asset owner |
| Condition monitoring design | Which measurements, on which equipment, at what interval, with what alarm limits | The condition monitoring provider, the control engineer |
| The measures, read | What the KPIs say about the process, and what to do about it | The maintenance manager, the Monday meeting |
A reliability engineer does not fix the pump. They work out why it keeps needing fixing, and change that.
Where the role stops
The role is defined as much by what it is not. The planner decides how a job will be done and makes it ready; the scheduler decides when; the supervisor runs the day. The condition monitoring technician takes the readings; the maintenance manager runs the function. Each of them is the reliability engineer's source or customer, and none of them is the reliability engineer's job. The table sets the boundaries out, because the commonest way the role is lost is by absorbing the one beside it.
| Role | Decides | Against the reliability engineer |
|---|---|---|
| Reliability engineer | Why equipment fails and what to change so it fails less. Works in months and years, from data | Owns the strategy and the failure data. Does not plan jobs or run the week |
| Planner | How a job will be done: scope, parts, permits, the job pack. Works a week or more ahead | Loads the strategy as plans. Hands findings back to the reliability engineer |
| Scheduler | When a job is done and by whom, against capacity. Works in the week before | Builds the week from ready work. Not a reliability role |
| Supervisor | The day: assignments, break-ins, the crew. Works in hours | Executes. The reliability engineer's best source of what actually happens |
| Condition monitoring technician | Takes the readings and reads the trends | Reports to the reliability engineer's program design; does not set the strategy |
| Maintenance manager | The function: people, budget, the process | Reads the measures the reliability engineer interprets, and decides |
The rhythm
The role runs on a calendar, because none of its work is urgent and all of it is important. The week's failures and findings are read while they are fresh. The month's bad actor list decides which investigations open. Each quarter a slice of the preventive program is reviewed task by task against the signals, and the spares register against usage. Each year the criticality assessment is revisited for what changed and the strategy refreshed on the equipment that cost the most. The investigations and the management of change reviews arrive as they are triggered.
| When | What |
|---|---|
| Weekly | Read the week's failures and findings; walk the worst actors with the supervisor; attend the Monday review for the reliability reasons behind the misses |
| Monthly | The bad actor list from the failure codes and the downtime; open or close investigations; review the condition monitoring exceptions; read the KPIs as a set |
| Quarterly | Review a slice of the preventive program task by task against the signals; review the critical spares register against usage and failures; present the reliability plan's progress |
| Yearly | Revisit the criticality assessment for what changed; refresh the strategy on the equipment that cost the most; audit the failure coding and the data the year rests on |
| On event | A failure that meets the investigation trigger; a modification under management of change that touches a strategy; a new asset's strategy before it is commissioned |
The data is the job
Every decision in the role rests on failure history, and failure history is the sum of how every work order was closed. A CMMS in which the failure code is always other, the hours are a guess and the findings are blank gives the reliability engineer nothing to read, and the first year in the role on most sites is spent fixing that: the failure code lists per equipment class, the mandatory fields, the mobile closure, and the walk to the crib room to explain why it matters. The ISO 14224 structure is the reference, and the planner and the supervisor are the allies.
The second source is condition data, from the monitoring program and the historian, and the third is cost, by asset, from the ledger through the CMMS. A reliability engineer who can join the three, failures, condition and cost, by functional location, can say which equipment to work on next and prove it. One who cannot is working from opinion, however good.
Skills, and what certifies them
The base is an engineering degree and years in maintenance or operations, so the data is read with the plant in mind. The toolkit is FMECA and reliability centred maintenance for the strategy, root cause analysis for the investigations, Weibull and the failure patterns for reading what the data says, criticality for the priorities, and a working knowledge of the CMMS and the historian as sources. The Certified Maintenance and Reliability Professional from SMRP and the Certified Reliability Engineer from ASQ are the recognised certifications, and both test the toolkit above.
The skills nobody certifies are the ones the role turns on: writing a task so a tradesperson will do it the same way twice, explaining to a supervisor why the failure code matters, and holding a strategy decision against the pressure to add a task after every incident.
How the role is lost
- Absorbed into the day: the reliability engineer is the maintenance engineer, the planner's backup and the project's spare resource.
- No data to read, and no mandate to fix the closure discipline that would produce it.
- Measured monthly, on numbers that take a year to move.
- Strategy owned by nobody, so the vendor manual is the program and every incident adds a task.
- Investigations that stop at the person, because the engineer has no standing to change the system.
- Reporting to the supervisor, so the urgent always wins.
- The title given without the role: no criticality assessment, no bad actor list, no review rhythm.
Common questions
What does a reliability engineer do?
Works out why a plant's equipment fails and what to change so it fails less, and then changes it through the people who own the plant. In practice: owns the maintenance strategy and the basis for every task and interval, runs the bad actor list and the root cause analyses on it, makes the critical spares decisions, designs the condition monitoring program, and reads the maintenance measures as a set and says what they mean. The role works in months and years, from data, as distinct from the planner and the supervisor who work in the week and the day.
What is the difference between a reliability engineer and a maintenance engineer?
Titles vary by site, and the useful distinction is by what each decides. A reliability engineer decides what the maintenance strategy is and why, from failure modes and data. A maintenance engineer, where the title is used separately, usually supports the execution: engineering the repairs, the modifications and the job plans. At a smaller site one person does both, and the risk is that the second consumes the first, because repairs are urgent and strategy never is.
What qualifications does a reliability engineer need?
An engineering degree in a relevant discipline is the usual base, mechanical or electrical most often, and several years in maintenance or operations so that the data is read with the plant in mind. The recognised certifications are the Certified Maintenance and Reliability Professional from SMRP and the Certified Reliability Engineer from ASQ. Training in FMECA and reliability centred maintenance, root cause analysis and Weibull analysis is the toolkit, and the ability to get a work order closed properly is the skill nobody certifies.
What measures a reliability engineer's performance?
The plant's reliability over years, read through the measures the role itself reads: breakdown hours by equipment and the share of hours on breakdown work, MTBF on the critical equipment, the corrective backlog trend, the bad actor list getting shorter and its members changing, the share of preventive tasks with a failure mode behind them, and the share of closed work orders with a failure code that is not other. Over a year or two the trend is the measure. Over a month, the role has not had time to change anything.
How many reliability engineers does a site need?
Fewer than planners and more than most sites have. A common shape is one reliability engineer for a few hundred critical items of equipment, with condition monitoring technicians reporting into the program they design, and the planners loading the strategy they produce. A site with none has a strategy nobody owns, which is usually the vendor manual. A site with one who is also the maintenance engineer, the planner's backup and the project's spare resource has the title without the role.
What tools does a reliability engineer use?
The CMMS, read rather than administered, for failure history and costs. The historian and the condition monitoring system for condition and process data. FMECA and reliability centred maintenance for the strategy, root cause analysis for the investigations, Weibull analysis for the patterns, criticality assessment for the priorities, and a reliability block diagram for the plant's availability. The spreadsheets and calculators are the easy part; the data in the CMMS is the hard one.
Key terms
Plain-language definitions from our glossary for the concepts this article leans on.
Standards and further reading
- Best practices, metrics and guidelines for maintenance and reliability (SMRP)
- Certified Maintenance and Reliability Professional (CMRP) (SMRP)
- Certified Reliability Engineer (CRE) (ASQ)
- SAE JA1011 Evaluation criteria for reliability-centered maintenance processes (SAE International)
- ISO 14224:2016 Collection and exchange of reliability and maintenance data for equipment (ISO)
- ISO 55001:2024 Asset management system requirements (ISO)
Related case studies and tools
- Reliability engineering and asset strategy program (case study)
Related reading
Working through something like this?
See how we approach these initiatives, or tell us what you are dealing with.