Power Plant O&M: The Three Numbers, the Definitions, and What the Contract Actually Commits To

An O&M arrangement is judged on three numbers: availability, forced outage rate and heat rate. Each of them can be calculated more than one way, and the contract decides which.

That is not a technicality. Two operators can report different availability figures for the same plant and both be correct, because available hours, planned outages and forced outages can each be counted differently. If your agreement does not name the definition, you have not agreed a target.

One further fact reframes what an O&M scope should cover. Accessories and support systems, not the turbine, typically account for 60 to 80 percent of a plant's unplanned outage events. A scope written around the prime mover is scoped around the minority of your downtime.

This guide covers how the numbers are defined, what the contract should say, and where the money is actually lost.

What O&M Management Is

Operations and maintenance management in power generation is an operational discipline that directly influences availability, safety, efficiency and long-term asset value. It is not routine inspection and corrective repair.

In active power plants, O&M management integrates maintenance execution, outage planning, performance monitoring and asset reliability into a single operating framework. That framework keeps critical equipment such as turbines, generators, transformers and balance-of-plant systems operating within design parameters under real load and environmental conditions.

When aligned correctly, O&M becomes a control mechanism that protects generation output, stabilises operating costs, and supports contractual and grid obligations. Those three outcomes are what the three numbers measure.

The Three Numbers Your Contract Turns On

Availability, forced outage rate and heat rate are the metrics that appear in commercially robust O&M agreements, because each maps to a different part of the owner's revenue.

Availability determines how many hours you can sell. Forced outage rate determines how many of your unavailable hours were unplanned, which is what grid operators and capacity markets penalise. Heat rate determines what each of those saleable hours costs you in fuel.

The most commercially robust O&M contracts tie contractor compensation to all three, aligning the operator's financial interest directly with the asset owner's generation and revenue objectives. A contract paying a fixed fee regardless of outcome makes the operator indifferent to the numbers you live on.

Why heat rate belongs alongside the other two

Heat rate is the fuel energy required per unit of electrical output, in Btu/kWh, and it degrades continuously through fouling, clearance growth and component wear. A plant that is available and inefficient loses money every hour it runs, which is why availability alone is an incomplete target.

Much of that degradation is recoverable through washing and maintenance. See our guide to gas turbine casings, clearance and fouling, and for the separate question of site derate, our guide to gas turbine power stations and ISO rating.

How Availability Is Actually Defined

Availability is the fraction of time a unit is capable of providing service, and the equations that define it are published in a standard almost no O&M contract names.

IEEE Std 762, Definitions for Use in Reporting Electric Generating Unit Reliability, Availability, and Productivity, is the governing reference. It defines the outage states, the equations, the treatment of deratings, and the concept of outside management control.

The outage states

A forced outage is an unplanned removal from service requiring immediate attention. A planned outage is scheduled well in advance for a defined duration, such as a major inspection. A maintenance outage can be deferred beyond a weekend but must be taken before the next planned outage.

NERC GADS uses these three groupings: Forced, Maintenance and Planned. European practice commonly uses two, Planned and Unplanned, and the ORAP system uses Forced and Scheduled, where Scheduled combines Maintenance and Planned.

That difference matters in a contract. Move an event between groupings and the reported availability changes without anything happening at the plant.

Availability factor and equivalent availability factor

Availability factor is the proportion of period hours the unit was available at any output. Equivalent availability factor (EAF) additionally accounts for deratings, meaning hours where the unit was available but could not reach full output.

EAF includes both planned and forced components, which makes it the correct metric for capacity availability reporting to grid operators and for offtake agreement compliance. Availability factor alone overstates what the plant could actually deliver.

The nameplate question

IEEE Std 762 sets nameplate capacity as the maximum "new and clean" capacity the generating unit can produce, which is usually achieved on the coldest day. Every derating is measured against that reference, which means the reference itself is worth confirming before a guarantee is written against it.

Availability is not reliability

Availability measures the fraction of time a unit is capable of service, accounting for outage frequency and duration. Productivity measures total power produced against potential production, accounting for the magnitude of outages as well. A unit with few but long outages and a unit with many short ones can share an availability figure and behave completely differently.

EFOR, EFORd and Grid Reporting

Equivalent forced outage rate is the metric that grid operators, capacity markets and O&M contracts actually use, because it isolates unplanned unavailability from scheduled work.

EFOR = (Forced Outage Hours + Equivalent Forced Derated Hours) ÷ (Available Hours + Forced Outage Hours) × 100.

It expresses the hours of unit failure, meaning unplanned outage hours plus equivalent unplanned derated hours, as a percentage of the total hours the unit was available or forced out.

Best-in-class gas turbine EFOR is 2 to 4 percent. That is the benchmark against which a proposed guarantee should be judged.

EFORd, and why it matters more in a market

Equivalent forced outage rate on demand (EFORd) measures forced unavailability only during periods when the unit was actually needed. In deregulated markets it has taken on greater importance than EFOR, because a forced outage during a low-demand overnight period costs the system far less than one during a summer peak.

NERC GADS applies Method 2 from IEEE Std 762 Annex F for all EFORd calculations.

The reporting obligation

NERC GADS is the main source of power station outage data in North America, and it follows IEEE Std 762 definitions. Generation owners on the NERC Compliance Registry report to it under NERC Rules of Procedure Section 1600.

If reporting lapses, the penalty is severe. PJM Manual 22 sets EFORd to 100 percent for any month of operation in which minimum reporting requirements are not met. Where a unit lacks sufficient GADS history, Class Average Outage Rates are applied instead of the unit's own performance.

That makes data submission an O&M deliverable with a direct capacity revenue consequence, and it belongs explicitly in the contract scope rather than being assumed.

Capacity payments

Where the plant earns a capacity payment, EFORd feeds the calculation. An O&M contractor whose compensation is unconnected to EFORd has no financial exposure to the metric that determines a significant portion of your revenue.

Execution Metrics: What Predicts the Outcome

Availability and EFOR are outcome metrics reported after the fact. Four execution metrics predict them, and they are the ones an owner should see monthly.

Maintenance backlog, the volume of identified but unexecuted work, is the leading indicator underneath all four. A growing backlog with high PM compliance means the PM programme is finding problems faster than the organisation can fix them.

Wrench time, the proportion of a technician's shift actually spent on tools, is the productivity metric almost nobody measures and everybody pays for. Time spent locating parts, waiting for permits or travelling to the work face is time paid for and not applied.

The diagnostic that matters

If PM compliance is high and EFOR is still poor, the PM programme is executing the wrong tasks. The programme is compliant with itself rather than aligned to actual failure modes, which is a strategy problem rather than an execution one.

Very few operators can state their EFOR trend over the last four quarters, their MTBF for a specific system, or whether their PM compliance rate is actually reducing forced outages rather than creating the appearance of activity.

What to require in reporting

Monthly reporting of all four execution metrics plus EFOR and EAF, at unit and system level, drawn from work order data rather than compiled by hand. Reporting produced from the CMMS is auditable; reporting produced in a spreadsheet is a narrative.

For how analytics and machine learning support this, see our analysis of AI in power plant operations.

Where Outages Actually Come From

Accessories and support systems, not the prime mover, typically account for 60 to 80 percent of a plant's unplanned outage events and 50 to 60 percent of unplanned outage time.

That finding, from US Department of Energy-sponsored turbine systems research, is the most useful single fact in O&M scoping and it contradicts how most contracts are written.

Balance of plant covers everything that is not the prime mover and generator: lube oil systems, fuel gas conditioning, cooling water, compressed air, instrument air, fire protection, electrical auxiliaries and controls.

The scoping consequence

An O&M scope, a spares holding and a skills mix built around the turbine addresses the minority of unplanned events. The auxiliary systems that cause most of your downtime are frequently the ones with the thinnest documentation, the fewest spares and the least specialist attention.

Two practical implications. Criticality analysis must cover balance of plant, not just the prime mover. And the O&M contractor's experience with your specific auxiliary systems matters as much as their turbine credentials.

For root cause analysis on the turbine side, see our guide to gas turbine troubleshooting and unplanned downtime. For structuring the planned outage that addresses what you find, see our gas turbine outage planning guide.

Maintenance Strategy: Matching Method to Failure Mode

Preventive maintenance is foundational, and it delivers reliability gains only when tasks are scheduled against operating hours, load profiles and equipment condition rather than fixed intervals alone.

Activities such as equipment inspections, lubrication programmes, filter servicing, calibration checks and vibration monitoring form the core. Consistent execution, accurate documentation and alignment with operational data are what separate a preventive programme from a checklist exercise.

Condition-based maintenance triggers work on measured equipment condition rather than elapsed time. Predictive maintenance uses trending to forecast when condition will reach the intervention threshold.

Applying one strategy across the whole plant is the most common and most expensive O&M error. The right method depends on the failure mode, the criticality of the asset and whether a measurable P-F interval exists between detectable onset and functional failure.

For strategy selection per asset, see our comparison of predictive versus preventive maintenance. For the task-level programme on rotating equipment, see our rotating equipment maintenance field guide.

The O&M Contract

A power plant O&M contract transfers responsibility for operational and maintenance activities across a defined asset or fleet, and what it excludes matters as much as what it covers.

Typical scope

Staffing and shift coverage. Preventive and corrective maintenance. Spare parts management. CMMS administration. Performance reporting. Outage planning and execution. Compliance with grid and regulatory obligations, including GADS submission.

The three commercial structures

Fixed fee gives budget certainty and leaves the contractor indifferent to outcomes unless performance terms are attached. Cost plus passes actual cost through with a margin and leaves the owner carrying cost risk. Performance-linked ties compensation to availability, forced outage rate and heat rate targets.

The most commercially robust agreements are performance-linked, because they align the operator's financial interest with the owner's revenue.

The LTSA boundary, and why owners pay twice

A long-term service agreement (LTSA) with the equipment manufacturer covers major maintenance, parts and technical support on the prime mover. An O&M contract covers plant operation and routine maintenance.

They overlap, and the overlap is where double-charging happens. Map the boundary explicitly: who performs hot gas path inspections, who supplies the parts, who owns the outage schedule, and what happens when an LTSA-covered component fails during an O&M-covered operation.

CMMS and data ownership

Establish that you own the CMMS data, the asset register, the work order history and the maintenance records, in an exportable format, with no restriction on engaging a different operator.

This is the same trap as the control system configuration clause. A contractor holding your maintenance history controls your next tender, because no competitor can price the work without it. It is the most expensive term to omit and the cheapest to include.

Spare parts ownership

Establish who owns the spares inventory, who funds replenishment, what happens to consigned stock at contract end, and whether the contractor may use your spares on another site. For criticality-based holding decisions, see our guide to turbine spare parts management.

What an O&M contract costs

Pricing varies by plant size, configuration, staffing model and whether the site is manned continuously or supported remotely. The structural driver is headcount, so a fixed-fee quotation that looks low against a competitor usually reflects a thinner shift roster rather than better productivity. Compare the staffing schedule, not the fee.

Availability Guarantees and the Exclusions That Matter

An availability guarantee is only as strong as its definition and its exclusion list, and most negotiation effort goes into the number rather than into either.

Three things determine whether a guarantee protects you.

The measurement method. Availability can be calculated several ways. Confirm precisely how available hours, planned outages, forced outages and deratings are counted, and name IEEE Std 762 as the reference so both parties are working from the same equations.

The exclusions. Every guarantee excludes events outside the operator's control. IEEE Std 762 formalises this as outside management control (OMC). Common exclusions include fuel supply failure, grid curtailment, force majeure and owner-directed operation outside design parameters. Make sure the exclusions are reasonable and clearly defined, because a broad exclusion list can remove enough hours to make the guarantee meaningless.

The penalty and bonus structure. A credible guarantee carries real financial consequences for shortfalls. Confirm the penalties are meaningful enough to drive operator behaviour rather than to be absorbed as a cost of doing business.

If the contractor misses the target

The remedy should be defined before it happens: liquidated damages at a stated rate, a cure period, a step-in right, and a termination threshold. A guarantee with no stated consequence is a target, not a guarantee.

Financing and PPA implications

Financiers increasingly require contractual uptime protection before funding power projects, because availability underpins the revenue that services the debt. Where a PPA carries its own availability obligation, the O&M guarantee should be at least as strong and defined the same way, or you carry the gap between two different definitions.

Response time

Where the site is remotely supported, response time directly affects availability, because MTTR includes travel. Specify it, with the measurement point defined.

Self-Perform or Outsource

The decision turns on scale, duty and whether you can recruit and retain the competency, not on cost per hour.

Self-perform suits owners with multiple plants, a stable operating profile and an existing engineering organisation that can carry the specialist skills between sites. The fixed cost is justified by utilisation across the fleet.

Outsource suits single-asset owners, remote sites, plants outside the owner's core business, and situations where specialist competency is needed but cannot be kept busy. It also transfers a recruitment problem, which on hard-to-staff sites is frequently the real reason.

Evaluating a provider

Ask for EFOR performance on a comparable plant, with the definition stated. Ask which balance-of-plant systems their crews have specific experience with, not just which turbines. Ask what reporting the owner receives and at what frequency. Ask what happens to the CMMS data at contract end. And ask for a reference customer on a similar asset who will take a call.

For a fuller treatment of provider selection, see our guide to choosing a power plant maintenance company.

Workforce and Operational Readiness

Reliable plant operation ultimately depends on the expertise and readiness of field personnel, and maintenance strategy that cannot be executed is not a strategy.

Maintenance execution requires trained professionals capable of identifying risks, performing complex repairs, and operating safely within live power generation environments. O&M teams must be familiar with plant-specific equipment, operating procedures and safety protocols, which is why crew continuity matters more than headline qualifications.

Continuous training, competency development and on-site experience ensure maintenance activities are performed accurately and efficiently even under time-sensitive or high-risk conditions.

What to specify

A competency framework with defined roles and assessed competencies, not a training attendance record. A minimum crew composition per shift. A handover process that survives turnover. And safety performance reporting including recordable incident rate and lost time incident rate, which insurers and prequalification processes both ask for.

ISO 45001:2018 is the international standard for occupational health and safety management systems and is the common requirement in prequalification.

If the site is remote

Rotation schedules, accommodation, medical cover and evacuation arrangements become operational requirements rather than HR details, and they drive both cost and achievable response time.

If staff turn over

Documented procedures, a maintained CMMS and a structured handover are what allow a plant to survive crew changes. An operation that depends on individuals knowing things that are not written down has a single point of failure with legs.

For the operations discipline that sits alongside this, including alarm management benchmarks, see our guide to plant operations and maintenance services.

Standards and Asset Management

Four standards frames govern O&M performance measurement, asset management and workforce safety.

Asset management maximises the value of generation assets across their lifespan by optimising resource use, controlling cost and maintaining regulatory compliance. It requires data from maintenance records, performance metrics and financial reports in one place, so that decisions on asset use, maintenance timing and capital investment are made on evidence rather than on the loudest request.

ISO 55001 is the governance standard for that discipline, and a gap assessment against it is a defined engagement where the problem is inconsistent decision-making across sites rather than a specific equipment issue.

Insurance

Machinery breakdown insurers assess maintenance regime, documentation and incident history when pricing industrial risk. A maintained CMMS with complete work order history is underwriting evidence, not just an operational tool.

How O&M Requirements Vary by Plant Type

The metrics and the standards are constant. What changes is where the risk concentrates.

Gas turbine simple cycle. Equivalent operating hours drive maintenance intervals, and starts consume more life than running hours. Fast-start peaking plant reaches intervals on a calendar basis far sooner than base load.

Combined cycle. The steam cycle, HRSG and water treatment add systems, chemistry control and a second set of failure modes. Heat rate degradation is more consequential because the plant runs more hours.

Reciprocating engine plant. More units, more maintenance events, shorter individual durations. Availability is less exposed to any single failure and total maintenance labour is higher.

Industrial captive plant. Availability is measured against the host process, not against a grid. An outage that costs production is more expensive than one that costs export revenue.

Remote and island systems. MTTR dominates, because spares and specialists are days away rather than hours. Spares holding and crew capability substitute for response time.

Renewables and storage. Different failure modes, different availability definitions, and availability guarantee language still maturing relative to thermal plant.

Outside the United States. IEEE Std 762 and ISO standards apply internationally. NERC GADS does not. Other markets operate their own reporting regimes and outage classifications, and European practice commonly uses two outage groupings where GADS uses three, which changes reported availability without changing plant behaviour.

What Prismecs Does

Prismecs operates and maintains power generation assets with resident crews, CMMS administration and parts support, and is OEM-agnostic across gas turbines, reciprocating engines and balance-of-plant systems.

The clearest reference is Duqm, Oman, where four TM2500 units totalling 110 MW are kept grid-ready with resident O&M crews, CMMS and parts support. That is a continuous operating scope rather than a campaign intervention, and it is the model this article describes.

Other delivered project scope includes eight TM2500 dual-fuel units totalling 260 MW at Birr, Switzerland, online in six months as a fast-track reserve plant with a new 220 kV interconnection; an LM2500XPRESS plant at Miaoli, Taiwan delivered in ten months; three LM6000PC units adding 150 MW of fast-start reserve; and an LM6000 fleet decommissioned in Norway, transported and recommissioned at a new site.

Capability spans O&M services for the operating phase, power generation asset services for the equipment, I&C services for controls and instrumentation, supply chain solutions for spares and procurement, EPCM services for project delivery, and technology and consulting for assessment.

OEM-agnostic matters on an O&M contract because the party recommending a maintenance scope is not the party selling the parts.

Apply this article's criteria to any operator, including us. Ask for EFOR on a comparable plant with the definition stated. Ask which balance-of-plant systems the crews have worked on. Ask what you receive monthly and where it comes from. Ask what happens to the CMMS data at contract end.

To discuss an O&M scope, a performance review or a contract structure, send your plant configuration, current availability and EFOR position, existing contract structure and the outcome you need to sales@prismecs.com or call +1 (888) 774-7632.

Frequently Asked Questions

What is O&M in a power plant?

Operations and maintenance is the integrated discipline governing plant operation, equipment reliability, outage planning, fault response and long-term asset health across the generation lifecycle. It directly determines plant availability, heat rate efficiency and return on capital. It covers workforce coordination, maintenance execution, performance monitoring of turbines, generators, transformers and balance-of-plant systems, and alignment between maintenance strategy and real load profiles and grid obligations.

What three metrics should an O&M contract use?

Availability, forced outage rate and heat rate. Availability determines how many hours you can sell, forced outage rate isolates unplanned unavailability that grid operators and capacity markets penalise, and heat rate determines what each saleable hour costs in fuel. The most commercially robust O&M contracts tie contractor compensation to all three, aligning the operator's financial interest with the owner's generation revenue.

How is power plant availability defined?

By IEEE Std 762, Definitions for Use in Reporting Electric Generating Unit Reliability, Availability, and Productivity, which sets the outage states, equations and treatment of deratings. Availability factor is the proportion of period hours a unit was available at any output. Equivalent availability factor additionally accounts for deratings, making it the correct metric for capacity reporting to grid operators and for offtake agreement compliance.

What is EFOR and how is it calculated?

Equivalent forced outage rate expresses unplanned unavailability as a percentage. EFOR equals forced outage hours plus equivalent forced derated hours, divided by available hours plus forced outage hours, multiplied by 100. It isolates unplanned unavailability from scheduled work, which is why grid operators, capacity markets and O&M contracts use it. Best-in-class gas turbine EFOR is 2 to 4 percent.

What is the difference between EFOR and EFORd?

EFOR measures forced unavailability across all hours. EFORd, equivalent forced outage rate on demand, measures it only during periods when the unit was actually needed. In deregulated markets EFORd carries more weight, because a forced outage overnight costs the system far less than one during a summer peak. NERC GADS applies Method 2 from IEEE Std 762 Annex F for all EFORd calculations.

What is NERC GADS and who has to report?

The Generating Availability Data System is the main source of power station outage data in North America, following IEEE Std 762 definitions. Generation owners on the NERC Compliance Registry report under NERC Rules of Procedure Section 1600. The consequence of lapsing is severe: PJM Manual 22 sets EFORd to 100 percent for any month in which minimum reporting requirements are not met.

Why do two operators report different availability for the same plant?

Because availability can be calculated several ways, and the definition is negotiable. Available hours, planned outages, forced outages and deratings can each be counted differently, and NERC GADS uses three outage groupings where European practice commonly uses two. Moving an event between groupings changes reported availability without anything happening at the plant. Name IEEE Std 762 in the contract.

What exclusions should I expect in an availability guarantee?

Every guarantee excludes events outside the operator's control, which IEEE Std 762 formalises as outside management control. Common exclusions include fuel supply failure, grid curtailment, force majeure and owner-directed operation outside design parameters. Make sure exclusions are reasonable and clearly defined, because a broad list can remove enough hours to leave the guarantee meaningless while the headline number looks strong.

Where do most unplanned outages actually come from?

Not the turbine. Accessories and support systems typically account for 60 to 80 percent of a plant's unplanned outage events and 50 to 60 percent of unplanned outage time. Balance of plant covers lube oil, fuel gas conditioning, cooling water, compressed air, fire protection, electrical auxiliaries and controls. An O&M scope, spares holding and skills mix built around the prime mover addresses the minority of downtime.

Which metrics predict forced outage rate?

Four execution metrics. PM compliance, the percentage of scheduled preventive maintenance completed on time, because deferred PM converts into forced outages on a lag. Schedule compliance, showing whether the plan is real. MTBF by system, showing where the next outage will come from. And MTTR, which determines outage duration once failure occurs and depends on spares availability and crew skill.

What if PM compliance is high but EFOR is still poor?

The preventive maintenance programme is executing the wrong tasks. It is compliant with itself rather than aligned to actual failure modes, which is a strategy problem rather than an execution one. The correction is reviewing tasks against failure mode and criticality per asset, not increasing PM frequency, which adds cost and intervention risk without addressing the cause.

What is the difference between an O&M contract and an LTSA?

A long-term service agreement with the equipment manufacturer covers major maintenance, parts and technical support on the prime mover. An O&M contract covers plant operation and routine maintenance. They overlap, and the overlap is where double-charging happens. Map the boundary explicitly: who performs hot gas path inspections, who supplies parts, who owns the outage schedule, and what happens when an LTSA-covered component fails.

Who should own the CMMS data in an O&M contract?

You should, explicitly and in writing. Establish ownership of the CMMS data, asset register, work order history and maintenance records in an exportable format with no restriction on engaging a different operator. A contractor holding your maintenance history controls your next tender, because no competitor can price the work without it. It is the most expensive term to omit.

Should I self-perform O&M or outsource it?

Self-perform suits owners with multiple plants, a stable operating profile and an engineering organisation that can carry specialist skills between sites, where fixed cost is justified by fleet utilisation. Outsource suits single-asset owners, remote sites, plants outside the core business, and cases where specialist competency is needed but cannot be kept busy. Outsourcing also transfers a recruitment problem, which on hard-to-staff sites is often the real driver.

Tags: Power Plant O&M Availability EFOR O&M Contract Plant Reliability