Direct answer
The main types of industrial maintenance are corrective (fix after failure), preventive (act before failure, by schedule or by condition), predictive (forecast failure from monitored trends) and detective (find hidden failures in protection and control). Most plants combine them by asset criticality. Authomathika supports electrical, automation and field service work.
Institutional content: Authomathika
Published:
Every maintenance team already uses more than one strategy, even when nobody named it. Some assets run until they fail, others get inspected on a calendar, and a few are monitored continuously. This guide maps the main types so you can see where each fits and which specific guide to read next.
What are the types of industrial maintenance?
The types of industrial maintenance are corrective, preventive (systematic and condition-based), predictive and detective. They differ in when the work happens and what triggers it: a failure, a calendar or counter, a measured condition, or a test of a hidden function.
| Type | Trigger | Typical electrical or automation example |
|---|---|---|
| Corrective | Failure or abnormal operation | Replacing a burned contactor in a motor control center |
| Preventive, systematic | Time, hours or cycles | Periodic retightening of terminals and cleaning of a panel |
| Preventive, condition-based | Inspection result or measured limit | Replacing a filter when differential pressure or temperature crosses a set limit |
| Predictive | Trend analysis of monitored variables | Thermography and vibration trends on motors and drives |
| Detective | Test of a function that is normally idle | Functional test of protection relays and emergency stops |
The boundaries overlap. Condition-based maintenance is often treated as a bridge between preventive and predictive work, so agree on definitions inside your plant before you build reports around them.
What is corrective maintenance and when does it happen?
Corrective maintenance is the work done after a failure or a degraded condition has occurred, to return the asset to service. It can be planned, when you choose to run to failure on a low-impact item, or unplanned, when the failure interrupts production.
Unplanned corrective work is the expensive version, because it competes with production for time, people and spare parts. Planned corrective work is a legitimate strategy for assets that are cheap, redundant, easy to replace and safe to lose, such as an indicator lamp or a spare analog input channel.
A useful habit is to record the cause of each corrective order. Repeated causes show where preventive or predictive work is missing. The guide Corrective or Preventive Maintenance: What Pays Off in a Plant goes deeper on this trade-off.
What are systematic and condition-based preventive maintenance?
Preventive maintenance is scheduled work meant to reduce the chance of failure. Systematic preventive maintenance follows a fixed interval (time, operating hours or cycles), while condition-based maintenance acts only when an inspection or a measurement shows that action is needed.
Systematic examples include lubrication routes, panel cleaning, calibration of instruments and periodic review of backups for controller programs. Condition-based examples include replacing a component after a visual inspection finds wear, or retorquing connections only where a thermal inspection shows a hot spot.
Systematic work is simple to schedule and audit, but it can replace parts that were still healthy. Condition-based work avoids that waste, but it needs a defined inspection routine and clear limits. To structure either one, read What Is Industrial Preventive Maintenance and What Is It For? and How to Build an Industrial Preventive Maintenance Plan.
What are predictive and detective maintenance?
Predictive maintenance uses measurements and trends to estimate when a failure is likely, so the intervention is scheduled before it happens. Detective maintenance tests functions that stay idle until needed, to reveal hidden failures before a real demand exposes them.
Predictive techniques commonly applied to electrical and rotating assets include thermography, vibration analysis, oil analysis and monitoring of motor current or drive data. The value comes from trends over time, not from a single reading, so you need a baseline and a regular collection routine.
Detective maintenance matters most for safety and protection. A relay that never operates can look healthy for years and still fail on the day of a fault. Typical tasks are functional tests of protection and interlocks, emergency stop checks, alarm and trip tests in the control system, and verification of redundant paths.
For a direct comparison of the two most confused strategies, see Preventive vs Predictive Maintenance: Key Differences.
How do you combine the types in one maintenance plan?
You combine them by asset criticality: the higher the consequence of failure, the more you invest in prevention, monitoring and testing. Low-impact assets can stay on planned corrective work, while critical assets get predictive and detective coverage on top of a preventive base.
A practical sequence:
- List the assets and rank them by effect on safety, production and environment if they fail.
- Record the failure modes and past corrective orders for each critical asset.
- Assign a strategy per failure mode: run to failure, fixed interval, condition-based, predictive or detective.
- Check that you can actually measure what the strategy depends on, and that people are trained to read it.
- Group the tasks into routes, work orders and shutdown windows.
- Review the plan after each failure and each shutdown.
Authomathika works with electrical, automation and field services, and its Services page describes the field work it executes, including planned shutdowns. Scope and budget are defined case by case.
What are common mistakes when choosing a strategy?
The most common mistake is applying one strategy to every asset. Others are buying monitoring tools without a routine to read them, and treating the plan as finished after it is written.
Criteria to check before approving a plan:
- Each strategy is tied to a failure mode, not only to an asset.
- Intervals come from history and condition, not only from habit.
- Predictive measurements have a baseline, limits and a person responsible for acting.
- Protection and interlocks have detective tests on a defined schedule.
- Spare parts for critical, long-lead items are identified in advance.
- Corrective orders feed back into the plan.
Another frequent error is poor records. Without a clean asset list and history, no strategy can be audited. Preventive Maintenance Spreadsheet: Fields and How to Organize It shows the fields worth keeping.
