Most power plant maintenance managers can tell you their current work order backlog and how many jobs closed last week. Far fewer can tell you their EFOR trend over the last four quarters, their MTBF for the fuel gas compressors, or whether PM compliance is actually driving down forced outages or just creating the appearance of activity. The difference between those two groups is rarely effort. It is whether the ten numbers that actually describe plant reliability are measured continuously and visible at the shift level, rather than assembled once a quarter for a management review nobody acts on in time. This list walks through the ten KPIs UK power plant managers are prioritising in 2026, what each one actually measures, and the benchmark range that separates a reliability programme from a maintenance schedule that only looks organised, all calculated automatically inside OxMaint.
2026 Power Plant Analytics · KPI Scorecard
Top 10 Power Plant KPIs Every Plant Manager Tracks in 2026
EFOR, EAF, capacity factor, heat rate, and station consumption tell you how the plant is performing right now. MTBF, MTTR, PM compliance, OEE, and maintenance cost per MWh tell you why performance is trending the way it is, and which lever to pull first to change it.
10
KPIs covering reliability, efficiency, and maintenance programme health in one scorecard
56%
of plants track PM completion at all, leaving less than half with a real view of schedule health
3-6 wks
how far ahead a deteriorating KPI trend typically warns of an outage, when it is actually being watched
6
Reliability engineering in power generation has converged on a small set of numbers that, tracked together, describe the health of a maintenance organisation with more accuracy than any single indicator, including the one most plant managers reach for first: total maintenance spend. A plant can spend heavily on maintenance and still have a rising EFOR if the spend is going to reactive repairs instead of the failure modes actually driving downtime. The ten KPIs below are the ones that show that difference clearly enough to act on it before the next forced outage, not after, and together they cover reliability, efficiency, and the maintenance programme health that drives both.
The Ten KPIs, What Each One Measures, and Where the Benchmark Sits
01
EFOR — Equivalent Forced Outage Rate
(Forced Outage Hours + Equivalent Forced Derated Hours) ÷ (Available Hours + Forced Outage Hours) × 100
The percentage of time a unit was unavailable or derated because of an unplanned event, and the primary metric grid operators and capacity planners use to judge a unit's dependability. Best-in-class gas turbine fleets run EFOR in the 2 to 4 percent range, with a plant above 10 percent signalling a maintenance programme that is reacting to failures rather than preventing them. Because EFOR is what capacity markets and grid operators actually use to price and dispatch a unit, a rising trend here has commercial consequences well beyond the maintenance budget line.
02
EAF — Equivalent Availability Factor
(Period Hours − Unavailable Hours, Weighted for Derates) ÷ Period Hours × 100
A broader availability measure than EFOR, accounting for planned outages, forced outages, and derated running all in one number, giving a plant manager a single figure for how much of the unit's full capacity was actually available to the grid across the reporting period. Falling EAF with a stable EFOR usually points to planned outage scope creeping longer than budgeted, rather than a new reliability problem, which is exactly the distinction a plant manager needs before deciding whether to fund more maintenance or tighten outage planning discipline instead.
03
Capacity Factor
Actual Energy Generated ÷ (Rated Capacity × Period Hours) × 100
How much of a unit's theoretical maximum output it actually produced over a period, shaped as much by dispatch decisions and market demand as by mechanical reliability. A capacity factor trending down while EFOR and EAF stay flat is usually a commercial dispatch signal rather than a maintenance one, which is exactly why the two need to be read side by side, not treated as a single combined reliability score.
04
Heat Rate
Fuel Energy Input (Btu or kJ) ÷ Net Electrical Output (kWh)
How efficiently a unit converts fuel into electricity, with a lower number meaning better efficiency, and the benchmark range varying significantly by technology, from modern combined-cycle gas turbines at the efficient end to older subcritical coal units at the inefficient end. Heat rate creeping upward with no combustion tuning change is often the earliest efficiency-loss signal of fouling, degraded blades, or an instrumentation drift worth investigating, and catching that drift early is usually far cheaper than the fuel cost it accumulates while it goes unnoticed for a full reporting quarter.
05
SEC — Station or Specific Energy Consumption
Auxiliary Power Consumed ÷ Gross Power Generated × 100
The share of a plant's own generation consumed running its own pumps, fans, compressors, and cooling systems before a single megawatt-hour reaches the grid. A rising station consumption ratio with no change in ambient conditions or dispatch pattern usually means an auxiliary system, often a cooling water pump or an induced draft fan, is running less efficiently than its design point, quietly eating into net output long before it shows up as a headline efficiency complaint.
06
MTBF — Mean Time Between Failures
Total Operating Hours ÷ Number of Unplanned Failures
How long assets run between unplanned failures, and the clearest single indicator of whether a PM programme is actually preventing breakdowns rather than just occupying technician hours. Best-in-class critical rotating equipment holds MTBF above 2,000 operating hours, and a flat or falling trend on any single asset class is usually the first visible sign that something in its maintenance plan is being deferred or missed, often weeks before that deferral turns into an actual unplanned trip.
07
MTTR — Mean Time to Repair
Total Repair Time ÷ Number of Repairs
How fast the maintenance team diagnoses and resolves a failure once it happens, measuring execution speed rather than prevention. A rising MTTR alongside a stable failure count usually points to a parts availability problem or a staffing gap rather than a technical one, and separating those two causes is exactly why MTTR needs its own tracked trend rather than being folded into a general downtime number that hides which cause is actually driving it.
08
PM Compliance
Completed On-Time Preventive Work Orders ÷ Total Scheduled Preventive Work Orders × 100
The percentage of scheduled preventive maintenance actually completed on time, and the leading indicator that predicts most of the lagging indicators above weeks before they move. A high PM compliance number with a flat MTBF is worth investigating on its own, since it can mean the PM tasks themselves are not targeting the failure modes that are actually occurring on that asset class, which is a task-design problem rather than a scheduling one.
09
OEE — Overall Equipment Effectiveness
Availability × Performance × Quality
A single combined score for true productive capacity, multiplying how often equipment was available, how close to full speed it ran, and how much output met quality standards. Most power plants run OEE between 60 and 70 percent, with world-class operations reaching 85 percent or higher, and each percentage point of improvement translating directly into measurable additional generation revenue at scale, which is why OEE tends to be the number finance pays closest attention to on the whole scorecard.
10
Maintenance Cost per MWh
Total Maintenance Spend ÷ Total Megawatt-Hours Generated
Maintenance spend normalised against actual output, so cost trends stay meaningful even when generation volume shifts seasonally or with dispatch. A rising cost per MWh while total spend stays flat is an early warning that asset condition is deteriorating faster than the maintenance budget is keeping pace, well before the effect shows up in EFOR or MTBF, giving a plant manager time to act while it is still a budget conversation rather than an outage.
How the Ten Numbers Actually Fit Together
Leading Indicators
PM compliance and maintenance cost per MWh move first, often weeks before anything else on the scorecard shifts. A slipping PM compliance number is the earliest warning that the rest of this list is about to start deteriorating, which is exactly why it deserves a weekly look rather than a quarterly one.
Reliability Response
MTBF and MTTR respond next, reflecting how well the maintenance programme is actually preventing and resolving failures once PM compliance has been sliding for a while. These two together describe the maintenance organisation's execution, separate from anything commercial or dispatch-related happening elsewhere, and are the pair worth reviewing side by side rather than in isolation.
Plant-Level Outcomes
EFOR, EAF, capacity factor, heat rate, station consumption, and OEE are the outcomes everyone outside the maintenance team actually sees, and the numbers that eventually show up in a board report if the leading indicators upstream were never caught in time. Reading all three tiers together, rather than any single number in isolation, is what turns a scorecard into an early-warning system instead of a monthly summary of problems that have already happened.
OxMaint · Power Plant KPI Dashboard
See All Ten KPIs Calculated Live From Your Own Work Order Data
No spreadsheets, no monthly reporting lag, no manual formulas. Every number on this list, calculated continuously as work orders open and close.
Benchmark Tiers — Where Your Plant Likely Sits on Each KPI
| KPI | World-Class | Typical Range | Investigate If |
|---|---|---|---|
| EFOR | 2–4% | 6–9% | Above 10% |
| EAF | 92%+ | 85–90% | Below 80% |
| MTBF (critical rotating equipment) | 2,000+ hrs | 800–1,500 hrs | Below 500 hrs |
| PM Compliance | 90%+ | 70–85% | Below 65% |
| OEE | 85%+ | 60–70% | Below 50% |
Ten KPIs is not a report nobody reads. It is the minimum set that shows whether a maintenance organisation is preventing failures or just documenting them after the fact, and the smallest set where removing any single number would leave a genuine blind spot in the reliability picture.
Every number above, calculated from the same work order and asset data your team is already generating, updated as the shift progresses rather than reconstructed at month end from spreadsheets and shift-log memory.
Frequently Asked Questions
Which of these ten KPIs should a plant manager review weekly versus monthly?
MTTR and PM compliance are operational metrics best reviewed weekly, since they need a fast response to prevent backlog accumulation. MTBF, EFOR, EAF, heat rate, and OEE reflect trends across multiple cycles and are more meaningful reviewed monthly, with cost per MWh usually reserved for a quarterly board-level view. Book a demo to see a review cadence configured for your own reporting structure.
Can MTBF and MTTR be trusted if work orders are still logged manually?
Manual work order logging introduces timestamp errors, missing failure codes, and incomplete repair records that directly corrupt both calculations, since each one depends on accurate start and end timestamps tied to a specific asset and failure mode, not a general note written up after the shift has already moved on.
Why track both EFOR and capacity factor if they seem to measure similar things?
EFOR isolates unplanned mechanical unavailability, while capacity factor reflects both reliability and commercial dispatch decisions such as how often the unit was actually called to run. A capacity factor drop with a flat EFOR usually points to a market or dispatch cause rather than a maintenance one, and separating the two prevents chasing a reliability fix for a commercial problem that maintenance spend cannot solve.
Does heat rate benchmarking differ between coal, gas, and combined-cycle plants?
Yes, significantly. Combined-cycle gas turbines run meaningfully more efficient heat rates than subcritical coal units by design, so heat rate should be benchmarked against the plant's own technology class and historical baseline rather than a single generic industry number that ignores the underlying thermodynamic cycle.
How quickly can a plant start seeing these ten KPIs on a live dashboard?
Once work order and asset data are structured consistently, most of these calculate automatically from day one, since the underlying data is usually already being captured, just not connected into a single scorecard view. Sign in to OxMaint to see how the KPI dashboard pulls directly from existing work order history without a separate reporting project.
A Maintenance Schedule Tells You What Was Planned. These Ten KPIs Tell You What Is Actually Happening.
EFOR, EAF, capacity factor, heat rate, station consumption, MTBF, MTTR, PM compliance, OEE, and cost per MWh, calculated live from the work orders your team already creates, with no separate reporting project or monthly spreadsheet to maintain alongside it.
