EAM for energy: maintain by condition, not by calendar

In energy, equipment is expensive and downtime is even more so. Substations, transformers, turbines, pumps and power lines are spread over large territories, and some sites sit in mountains or hard-to-reach areas. The familiar approach — maintenance by calendar — is easy to understand, but it either spends effort on healthy equipment or misses a developing fault. Condition-based maintenance solves this, provided the groundwork is in place.

Start with the asset register

No maintenance strategy works without an accurate register. In EAM, the register is not a list of tag numbers but a hierarchy: site, installation, assembly, component. Each item carries its technical data, location, criticality, repair history and costs.

• Define hierarchy levels and coding rules that are the same across all branches.
• Assess the criticality of each asset: what a failure means for safety, output and customers.
• Link equipment to its documentation, spare parts and standard jobs.

This is usually the most labour-intensive part of the project. It can be done in stages, starting with the most critical sites.

The work order at the centre

Every job — planned, emergency, or raised by an inspector on a walk-round — is recorded as a work order. It specifies the equipment, the tasks, the crew, permits, materials and time. A closed work order adds to the asset’s history and shows the real cost of maintaining it.

Once all work runs through work orders, management gets a clear picture: what each substation costs to maintain, which assemblies fail most often, and where deferred maintenance is building up.

Preventive and condition-based maintenance

Preventive maintenance is tied to the calendar or to running hours: every so many months or operating hours. It is easy to plan, and it is mandatory wherever safety rules or the manufacturer require it.

Condition-based maintenance is triggered when equipment readings move outside acceptable limits: vibration or temperature rises, oil composition changes, insulation resistance drops. Equipment life is used more fully, and faults are caught earlier.

In practice you need both. Critical equipment with good monitoring moves to condition-based maintenance; the rest stays on a schedule. The decision is made per asset class, based on criticality and the data available.

Condition-based maintenance starts not with sensors but with an asset register and work-order discipline — without them, monitoring data has nowhere to go.

SCADA and IoT data in work orders

Many sites already have SCADA and sensors. The problem is that the data lives in the control room and never reaches the maintenance team. Connecting ICONICS and IFS Cloud closes that gap.

• ICONICS collects data from controllers and sensors, visualises it and watches threshold values.
• When a reading crosses a limit, the event is passed to IFS Cloud and linked to the specific asset in the register.
• IFS Cloud creates a request or work order according to a predefined rule: which tasks, what priority, which crew.
• Once the work is done, history and costs are saved on the asset record.

Take care not to flood the maintenance team with automatic requests. Start with a small set of proven rules on critical equipment and extend it once the thresholds are calibrated.

Remote sites

Sites without reliable connectivity need mobile tools that work offline. The technician receives the work order in advance, does the job, records readings and photos, and the data syncs when a connection returns. It is just as important to hold a stock of critical spares on site and track their movement, so a crew does not drive hundreds of kilometres for a single part.

What to measure

To know whether the approach works, you need metrics. Typical ones are the share of emergency work in total workload, preventive plan completion, mean time between failures and mean time to repair by equipment class, maintenance cost per asset, and the size of the deferred maintenance backlog. EAM calculates these automatically, provided work orders are closed on time with actual costs. Watch the trend rather than any single value: in the first months, the numbers reflect data quality more than equipment condition.

Where to start

• Pick one branch or one equipment class for a pilot.
• Get the register and criticality assessment in order.
• Move all work onto work orders.
• Connect monitoring data for a few key parameters.
• Review the results and roll the approach out to other sites.

This sequence pays off from the first steps: even without sensors, a clean register and work-order discipline show where maintenance money actually goes.