Millstone 2 - Status & History

NukeWorker's Operational Excellence Score is a composite ranking that combines 7 publicly observable performance metrics into a single 0-100 number per US nuclear unit.

How it's computed: Each unit's raw value is converted to a percentile (0-100) within the 94 active US units that have at least 2 completed refueling cycles. The 7 percentiles are averaged into 3 sub-scores (Reliability, Efficiency, Discipline), and those are averaged into the final composite.

Active window — 3-year:

• Reliability — Capacity factor (3yr), forced outage days/yr (3yr), B2B avg streak (lifetime)
• Efficiency — in-cycle fuel utilization (3yr), median refuel duration (3yr)
• Discipline — robust mean startup duration, scrams/yr (3yr, annualised)

Cohort rankings: alongside the fleet rank, each unit gets ranked within its reactor type (PWR/BWR), containment design (Ice Condenser, Mark I, etc.), and cycle length (12/18/24-month). Useful for apples-to-apples comparisons.

What's excluded: subjective community ratings, confidential INPO ratings, financial metrics. We only use publicly available NRC + cycle data.

Why three windows? 5yr (default) is the industry standard. 3yr captures recent operational changes and post-pandemic recovery. Lifetime gives newer units a fair comparison and reveals long-term consistency.

Edge cases: units with fewer than 2 cycles (e.g., Vogtle 4) show "Provisional". Decommissioned units excluded.

Refresh cadence: recomputed daily. Each window cached separately.

A data-derived proxy for operational performance; not affiliated with INPO. INPO ratings are confidential and qualitative — ours is public and quantitative.

NukeWorker's Operational Excellence Score is a composite ranking that combines 7 publicly observable performance metrics into a single 0-100 number per US nuclear unit.

How it's computed: Each unit's raw value is converted to a percentile (0-100) within the 94 active US units that have at least 2 completed refueling cycles. The 7 percentiles are averaged into 3 sub-scores (Reliability, Efficiency, Discipline), and those are averaged into the final composite.

Active window — 5-year:

• Reliability — Capacity factor (5yr), forced outage days/yr (5yr), B2B avg streak (lifetime)
• Efficiency — in-cycle fuel utilization (3yr), median refuel duration (5yr)
• Discipline — robust mean startup duration, scrams/yr (5yr, annualised)

Cohort rankings: alongside the fleet rank, each unit gets ranked within its reactor type (PWR/BWR), containment design (Ice Condenser, Mark I, etc.), and cycle length (12/18/24-month). Useful for apples-to-apples comparisons.

What's excluded: subjective community ratings, confidential INPO ratings, financial metrics. We only use publicly available NRC + cycle data.

Why three windows? 5yr (default) is the industry standard. 3yr captures recent operational changes and post-pandemic recovery. Lifetime gives newer units a fair comparison and reveals long-term consistency.

Edge cases: units with fewer than 2 cycles (e.g., Vogtle 4) show "Provisional". Decommissioned units excluded.

Refresh cadence: recomputed daily. Each window cached separately.

A data-derived proxy for operational performance; not affiliated with INPO. INPO ratings are confidential and qualitative — ours is public and quantitative.

NukeWorker's Operational Excellence Score is a composite ranking that combines 7 publicly observable performance metrics into a single 0-100 number per US nuclear unit.

How it's computed: Each unit's raw value is converted to a percentile (0-100) within the 94 active US units that have at least 2 completed refueling cycles. The 7 percentiles are averaged into 3 sub-scores (Reliability, Efficiency, Discipline), and those are averaged into the final composite.

Active window — Lifetime:

• Reliability — Capacity factor (lifetime), forced outage days/yr (lifetime), B2B avg streak (lifetime)
• Efficiency — in-cycle fuel utilization (lifetime), median refuel duration (lifetime)
• Discipline — robust mean startup duration, scrams/yr (lifetime, annualised)

Cohort rankings: alongside the fleet rank, each unit gets ranked within its reactor type (PWR/BWR), containment design (Ice Condenser, Mark I, etc.), and cycle length (12/18/24-month). Useful for apples-to-apples comparisons.

What's excluded: subjective community ratings, confidential INPO ratings, financial metrics. We only use publicly available NRC + cycle data.

Why three windows? 5yr (default) is the industry standard. 3yr captures recent operational changes and post-pandemic recovery. Lifetime gives newer units a fair comparison and reveals long-term consistency.

Edge cases: units with fewer than 2 cycles (e.g., Vogtle 4) show "Provisional". Decommissioned units excluded.

Refresh cadence: recomputed daily. Each window cached separately.

A data-derived proxy for operational performance; not affiliated with INPO. INPO ratings are confidential and qualitative — ours is public and quantitative.