What a work order is, and how work order management actually works
The work order is the record every maintenance function reads. What it is and is not, the types, the fields it carries and why each matters, the eight statuses and the gate into each, priority by rule, what closing one properly means, what the CMMS has to hold, and the measures that say whether the process is working.
Key takeaways
- A work order is the record of one piece of maintenance work: what is to be done, on which asset, why, by whom, with what, and afterwards what it took and what was found.
- It is not the request. A notification says something was seen. A work order says something will be done, and it exists only after someone has approved the work and given it a priority.
- Work order management is the process that moves every order through the same statuses, with a gate into each, so planning, scheduling and history all read from one record.
- The value of a work order is mostly in how it is closed. Hours, findings and failure codes recorded the same day are the history every later decision reads.
- The CMMS holds the process. Statuses, mandatory fields, priority rules and failure code lists are configuration decisions, and they decide whether the history is worth anything.
Every maintenance system runs on work orders, and most of the arguments about maintenance performance are arguments about them: the backlog is too big, the history is useless, the schedule is full of jobs that were not ready, nobody knows what the crusher cost last year. Each of those is a work order process problem, and the process is the same on every site, whatever the software.
This guide sets out what a work order is and what it is not, the types, the fields it carries and why each matters, the statuses it moves through and the gate into each, how priority is set, what closing one properly means, what the CMMS has to be configured to hold, and the measures that say whether the process is working. It assumes the weekly planning and scheduling cycle, and it is the record that cycle runs on.
What a work order is, and is not
A work order is the record of one piece of maintenance work, from the decision to do it to the history it leaves. It names the asset, says why the work exists and how urgent it is, carries the plan for doing it, is what the schedule loads and the crew is assigned, and when closed holds what the work took and what was found. Planning, scheduling, costing and reliability all read the same record, which is why the fields and the discipline around them matter more than any one of those functions.
It is not the request. A notification or work request is raised by whoever saw something, and it records the observation. A work order exists only once the work has been confirmed as real and necessary and given a priority. It is not the job plan either. The job plan is the standing description of how a job is done, reused every time; the work order is this instance of doing it, and carries the plan. And it is not the maintenance plan, which is the standing schedule in the CMMS that generates preventive work orders on their intervals.
The types
Work orders are typed by where they come from, because that decides whether they can be planned. Preventive and condition-based work can be seen coming. Corrective work from a finding has time to plan. Breakdown work does not, and is the one type that legitimately bypasses planning, with the planning done afterwards for the follow-up. The type is a field on the order and a fixed list, so the shares can be counted.
| Type | Starts from | Planned? |
|---|---|---|
| Preventive | The maintenance plan in the CMMS, generated on its interval | Always, from a standing job plan |
| Corrective | A finding from a preventive task, an inspection or condition monitoring, with time to plan | Yes, through the backlog |
| Breakdown (emergency) | A functional failure that stops production or creates a hazard | No. Done now, planned afterwards for the follow-up |
| Condition based | A condition monitoring alarm or trend crossing its limit | Yes, inside the P-F interval |
| Improvement or project | A modification, a capital job or a reliability action | Yes, often through a change process |
| Standing | Recurring minor work booked against one open order, such as housekeeping | No. Used sparingly, because hours booked here lose their history |
The share of hours on breakdown orders is the plant's reliability, measured in the one place it cannot be argued with.
What it carries
A work order is a form with a purpose for every field. The fields below are the ones that make the record useful afterwards, and the reason beside each is the test for whether a field should be mandatory in the CMMS. A field nobody reads later can be optional. A field the history depends on cannot.
| Field | Why it matters |
|---|---|
| Functional location or equipment | Puts the work and its cost in the asset hierarchy, so history can be read by equipment |
| Type and priority | Says what kind of work it is and how urgent, by the site's rule, so the backlog can be sorted |
| Description and long text | What was observed and what is to be done. The first line is what every list shows |
| Job plan or task list | The steps, trades, durations, parts and permits, so the job is done the same way each time |
| Planned hours by trade | What the schedule loads against capacity, and what actual hours are compared with |
| Parts and materials | Reserved from stock or purchased, with the bill of materials behind them |
| Permits, isolations and access | What has to be in place before the crew can start |
| Actual hours and findings | What it really took and what was seen, which is the history the next decision reads |
| Failure codes | Failure mode, cause and remedy from a fixed list, so failures can be counted across equipment |
| Costs | Labour, materials and services settled against the asset |
The statuses, and the gate into each
Work order management is the movement of every order through the same statuses, in the same order, with a condition that has to be true before it enters each. The names vary by system. The gates do not, and the one that matters most is readiness, because it is the line between the planner's work and the scheduler's. An order that reaches the schedule without passing it is the reason a week starts with jobs waiting on parts.
The gates are enforced in the CMMS where they can be, by a status that cannot be set until its fields are filled, and by discipline where they cannot. A planner who sets Ready before the parts have arrived has defeated the process with one click, and the schedule compliance numbers will say so a week later.
1Requested
Gate
A notification or request is raised against a functional location, with what was seen
Operator, tradesperson, inspector
2Approved
Gate
Reviewed as real, not a duplicate, and given a priority by the rule
Supervisor or planner
3Planned
Gate
Scoped with a job plan: steps, trades, hours, parts, permits, access
Planner
4Ready
Gate
Parts received and kitted, permits identified, access confirmed
Planner
5Scheduled
Gate
Placed on the frozen weekly schedule with a day and a crew
Scheduler
6In progress
Gate
Assigned and started; break-in only through the approval path
Supervisor
7Complete
Gate
Work done, hours booked, findings and failure codes recorded the same day
Tradesperson
8Closed
Gate
Costs settled, history reviewed, follow-up work raised
Planner or supervisor
Planning: the work is made ready
Execution: the work is done and its history kept
| Status | The gate into it | Who moves it |
|---|---|---|
| Requested | A notification or request is raised against a functional location, with what was seen | Operator, tradesperson, inspector |
| Approved | Reviewed as real, not a duplicate, and given a priority by the rule | Supervisor or planner |
| Planned | Scoped with a job plan: steps, trades, hours, parts, permits, access | Planner |
| Ready | Parts received and kitted, permits identified, access confirmed | Planner |
| Scheduled | Placed on the frozen weekly schedule with a day and a crew | Scheduler |
| In progress | Assigned and started; break-in only through the approval path | Supervisor |
| Complete | Work done, hours booked, findings and failure codes recorded the same day | Tradesperson |
| Closed | Costs settled, history reviewed, follow-up work raised | Planner or supervisor |
Priority by rule, not by voice
Priority is set once, at approval, by a rule that reads consequence and urgency, and the rule uses the same scale as the site's criticality assessment so that a high priority means the same thing on every order. The usual shape is a small matrix: the consequence of leaving the failure, from the criticality of the asset and the failure mode, against how long it can be left. The result is a priority with a target completion time, which is what the backlog is sorted by and what the scheduler draws from.
A priority set by who asked loudest produces a backlog in which everything is urgent and nothing can be sorted, and the scheduler ends up choosing by familiarity. The rule is written down, the planner applies it, and a request to raise a priority goes through the same approval as the original.
Closing it properly
A work order's value to the future is decided in the ten minutes after the job. Actual hours by trade, against the planned hours, are what tell the planner the estimate was wrong. Findings in the tradesperson's words are what the next planner reads and what an investigation will want. Failure codes from a fixed list, failure mode, cause and remedy in the pattern ISO 14224 sets out, are what let failures be counted across equipment and ranked into a bad actor list. Follow-up work found during the job is raised as a new request, not written in the long text where nobody will find it.
All of it is recorded the same day, on the device the tradesperson has with them, because by the following week the hours are a guess and the findings are gone. A closed work order with no hours, no findings and the failure code set to other is a cost with no history, and a CMMS full of them is why so many sites believe their data is useless. The data was never captured.
Equipment history is not a report the CMMS produces. It is the sum of how every work order was closed.
What the CMMS has to be set up to hold
The work order process lives in the CMMS or EAM, and the decisions that make it work are configuration decisions made once. Which statuses exist and which fields must be filled before each can be set. The priority rule as a matrix the system applies. Failure code lists per equipment class, short enough to be used and long enough to mean something. Job plans linked to the maintenance plans so a preventive order arrives with its steps and parts. Mobile closure so the record is made at the job. And the asset hierarchy beneath all of it, because an order on the wrong functional location puts its cost and its history in the wrong place.
The temptation is to make every field mandatory and every status strict, which produces a system the crew works around. The fields that are mandatory are the ones the history depends on, and the list above is that list. The rest are optional and most are hidden.
The measures
The process is read through a handful of measures, all of which come from the work order record and none of which mean much alone. The backlog in crew weeks by trade, with the ready share separately. The age of open orders by priority against their target times. The share of hours on breakdown orders, which is reliability. The planned share, which is hours on orders with a job plan. Schedule compliance and the break-in share from the weekly cycle. And the share of closed orders with hours, findings and a failure code that is not other, which is the measure of whether any of the others can be trusted. The maintenance KPIs guide sets out how they are read together.
Where the process fails
- Requests turned into work orders without screening, so the backlog fills with duplicates and wishes.
- Priority by voice, so everything is urgent and the scheduler chooses by familiarity.
- Ready set before the parts arrive, so the schedule is full of jobs that stall.
- Orders raised on the area rather than the equipment, so the history cannot be read by asset.
- Closed with no hours, no findings and the failure code set to other.
- Follow-up work written in the long text instead of raised as a request.
- Standing work orders used for real work, so the hours and the history disappear.
- Every field mandatory, so the crew works around the system instead of in it.
Common questions
What is a work order?
A record in the maintenance system authorising and describing one piece of maintenance work: the asset it is on, the type and priority of the work, what is to be done, the job plan with its steps, trades, hours and parts, and when complete, the actual hours, the findings and the failure codes. It is the unit that planning, scheduling, costing and equipment history all work in.
What is the difference between a work order and a work request?
A work request, called a notification in some systems, is raised by anyone who sees something: an operator, a tradesperson, an inspector. It records what was seen and where. A work order is created from it only after a supervisor or planner has confirmed the work is real, is not already covered, and has given it a priority. Keeping the two separate stops the backlog filling with duplicates and wishes.
What are the types of work order?
Preventive, generated by the maintenance plan on its interval. Corrective, raised from a finding with time to plan. Breakdown or emergency, for a functional failure that has to be fixed now. Condition based, triggered by a monitoring alarm. Improvement or project, for modifications and capital work. And standing work orders, one open order that recurring minor work is booked against, used sparingly because hours booked there lose their history.
What should a work order include?
The functional location or equipment, the type and priority, a description with the observation and the work, the job plan or task list, planned hours by trade, parts and materials, the permits, isolations and access needed, and after the work, the actual hours, the findings, the failure codes and the costs. A work order missing the asset or the failure codes cannot be read back as history.
What is work order management software?
The part of a computerised maintenance management system, or of an enterprise asset management system, that creates work orders, moves them through their statuses, holds the job plans, reserves the parts, feeds the schedule and records the completion. Most sites have one. The difference between sites is not the software but how its statuses, mandatory fields and code lists were configured, and whether the crew closes work orders properly.
Who approves a work order?
Usually the maintenance supervisor for routine work and the superintendent or a cost centre owner above an agreed value, with the planner screening requests for duplicates and completeness before approval. The approval sets the priority by the site's rule and commits the work to the backlog. A second approval, for break-in work added after the weekly schedule is frozen, goes through one named path so the additions are counted.
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)
- ISO 14224:2016 Collection and exchange of reliability and maintenance data for equipment (ISO)
- ISO 55001:2024 Asset management system requirements (ISO)
- Operations and maintenance best practices guide (US DOE Federal Energy Management Program)
- The Asset Management Landscape, third edition (GFMAM)
Related case studies and tools
- Schedule Compliance Calculator (tool)
- Work management and scheduling uplift (case study)
Related reading
Working through something like this?
See how we approach these initiatives, or tell us what you are dealing with.