Nuclear Industry Pack

Nuclear Decommissioning Cost & Schedule Forecasting

Decommissioning forecasts fail differently to construction forecasts. The work is discovered rather than designed, the regulator gates progress, and the programme outlives the people who baselined it. The Nuclear pack forecasts on that basis.

How the engine works

What breaks forecasts in nuclear

The maths is the same everywhere. These are the sector-specific conditions that make a generic forecast wrong.

Characterisation uncertainty dominates

You are forecasting the cost of removing material whose extent and activity you do not yet fully know. Early-phase estimates carry uncertainty ranges that would be unacceptable anywhere else, and a single-point EAC hides it entirely.

A point estimate is indefensible — you need a range

Regulatory hold points are not float

ONR permissioning, safety case acceptance and waste consignment approvals sit on the critical path but behave nothing like activities. They have no productivity rate and cannot be accelerated with resource.

CPM treats them as compressible. They are not.

Contingency drawdown over decades

A twenty-year programme holds contingency that must be released on an agreed profile. Forecasting against remaining contingency, not just remaining budget, is what the NDA and Treasury actually scrutinise.

Drawdown profile is a reporting requirement

Waste routes drive cost, not activity duration

Cost is dominated by waste stream, container count and disposal route — ILW versus LLW versus VLLW. Forecasting on labour hours alone misses the majority of the spend.

Cost per m³ by stream, not cost per man-hour

Baselines outlive their authors

Over a multi-decade programme, the assumptions behind the original estimate are lost unless they were recorded as data. Every forecast challenge becomes archaeology.

Undocumented assumptions cannot be defended

What's in the Nuclear pack

The engine is identical across industries. The pack is the content that makes it speak your sector's language from the first time you open it.

WBS skeleton

  • Site characterisation & survey
  • Asset care & maintenance
  • Decontamination
  • Dismantling & size reduction
  • Waste packaging & consignment
  • Final site clearance

Cost codes

  • ILW / LLW / VLLW streams
  • Container types & counts
  • Disposal route & gate fees
  • Transport & shielding
  • PPE and RPE consumables

Risk register

  • Unexpected contamination found
  • Safety case rejection / rework
  • Waste route unavailability
  • Skilled RPS resource shortage
  • Extended care & maintenance

Regulatory milestones

  • ONR permissioning gates
  • Safety case submission & acceptance
  • Waste consignment approvals
  • Environment Agency permits
  • End-state definition sign-off

Worked example

Reactor building decommissioning, year 4 of 11

BAC $88M. To date: EV $27.4M, AC $31.2M, PV $29.1M. Characterisation has found 18% more contaminated concrete than the baseline assumed.

CPI
0.88
EAC (CPI-based)
$100M
P50 forecast
$103M
P80 forecast
$117M
The CPI-based EAC says $100M. The Monte Carlo says you need $117M to be 80% confident. On a programme with this much characterisation uncertainty, the gap between those two numbers is the conversation with your funder — and single-point forecasting hides it completely.

Illustrative figures, shown to demonstrate the method. Your numbers replace them.

Nuclear questions

No. It forecasts outturn from progress you have already earned. Bottom-up estimating stays where it is — this tells you whether that estimate is still credible four years in.
Yes. Extended low-activity periods are modelled as fixed-rate holding costs rather than progress-earning work, so they do not distort CPI.
It forecasts at stream and container level, not item level. For a full radiological inventory you would keep your existing system and feed totals in.
The Assumptions & Audit sheet records every distribution, its source and the date it was set. That record is the defence — the number alone never is.

Other industry packs

Forecast nuclear with numbers you can defend

Forecast Engine $299 · Nuclear pack $99 · all eight packs $499