The Commitment Before the Commitment

Major capital projects are often described as succeeding or failing during execution. That framing is incomplete. A substantial portion of a project’s eventual exposure is shaped before execution, when decision-makers conclude that the project is sufficiently defined to authorize capital, obtain financing, select an EPC contractor, submit or accept a bid, award a contract, issue notice to proceed, or mobilize work.

This paper examines project readiness before capital commitment: the sufficiency of evidence, definition, integration, governance, and uncertainty treatment for a specific project decision. It draws on public research and public records, including the Construction Industry Institute, the U.S. Government Accountability Office, the U.S. National Academies, research on historical nuclear-construction costs, the Muskrat Falls Commission of Inquiry, the Georgia Public Service Commission, and Georgia Power Company.

The central argument is straightforward: major-project exposure commonly arises not from one unforeseen event, but from the interaction of known conditions, foreseeable stresses, concentrated dependencies, and external shocks that exceed the resilience built into the original delivery case.

This is not an argument that uncertainty can be eliminated. It cannot. It is an argument that decision-makers should establish which conditions carry the next commitment, what evidence supports those conditions, what can change them, and what response remains available if they do.

Research question: What evidence should decision-makers require before authorizing a consequential capital, financing, procurement, contractual, or mobilization commitment on a major project?

---

Executive findings

Early decisions constrain later options

The National Academies states that front-end planning defines the project and that decisions made in this phase constrain and support later action. The Construction Industry Institute likewise identifies front-end planning quality as a material influence on capital-project delivery.

Formal approval is not proof of project readiness

Capital authorization, financing approval, EPC selection, contract award, or a notice to proceed may all be valid governance decisions while the conditions necessary to achieve the anticipated outcome remain incomplete, uncertain, or weakly integrated.

A completion date is not an executable delivery path

A schedule date is only credible when the underlying schedule captures the work, logic, resources, durations, interfaces, risk, update rules, turnover activities, and acceptance conditions that govern completion.

Contingency is not resilience

A contingency percentage is not evidence of resilience. Resilience exists when the project can identify the uncertainty being carried, the range it creates, the owner of the response, the fallback capacity available if it materializes, and the decision threshold at which the original delivery case is no longer credible.

The loss of a delivery party is not merely a contractual risk

EPC contractor distress, a major subcontractor failure, supplier disruption, or the loss of a critical specialist can quickly become a project-continuity event. A decision-ready project identifies what must be preserved if that happens: design authority, records, supplier relationships, quality documentation, key personnel, warranties, work-package knowledge, financing capacity, and decision rights.

Headcount is not capacity

Craft availability and productivity are delivery-system conditions. Qualified craft, supervision, released design, usable materials, accessible work fronts, tools, logistics, quality controls, and an executable sequence must be available together. A project that measures labor only by people on site can mistake congestion and paid hours for productive progress.

Turnover and owner readiness belong in the delivery case

Construction completion, mechanical completion, system turnover, commissioning completion, and operational readiness are distinct conditions. A project becomes an operating asset only when the receiving organization can accept, operate, maintain, control, and sustain it.

---

The commitment before the commitment

A major project is committed in stages. Board authorization, financial close, contractor selection, bid submission, contract award, notice to proceed, long-lead procurement, field mobilization, recovery funding, turnover authorization, and commercial operation each create a different form of exposure. Each can reduce some portion of the project’s optionality.

The formal decision is visible: a resolution is passed, a contract is signed, a financing condition is satisfied, or a mobilization instruction is issued. Less visible is the preceding judgment that the project’s definition, estimate, schedule, delivery model, market assumptions, resources, governance, and uncertainty treatment are sufficient for that decision.

That judgment is the commitment before the commitment.

It is not reasonable to require complete certainty before a major project proceeds. Complex projects operate in conditions of technical, commercial, market, regulatory, and execution uncertainty. The relevant question is not whether uncertainty remains. It is whether material uncertainty has been identified, evidence has been weighed, consequences have been made visible, and decision-makers understand which conditions must hold for the commitment to remain credible.

Approval and readiness are different propositions

Formal approval answers a governance question: has the authorized party agreed to proceed under stated terms, authorities, and conditions?

Decision readiness answers an evidentiary question: are the conditions that carry those terms sufficiently established for the decision being taken?

These propositions can align. A project with a clear definition, integrated schedule, credible delivery plan, appropriate contingency treatment, defined interfaces, and explicit decision thresholds may be ready for a particular authorization. They can also diverge. A project may have an approved budget, a selected contractor, or a scheduled start date while material elements of scope, constructability, supply, labor, interface ownership, completion sequence, or governance remain unresolved.

Uncertainty changes form after commitment

Before commitment, uncertainty may appear as an assumption, range, contingency, risk item, information gap, or alternative. After commitment, the same uncertainty may appear as change, delay, rework, claim exposure, financing pressure, productivity loss, constrained turnover, or recovery funding requirement.

The objective is not indefinite study. It is proportionate evidence. A decision with limited irreversibility may reasonably proceed with more uncertainty than one that fixes an EPC relationship, commits large capital, locks in a completion date, begins irreversible field work, or narrows financing and recovery options.

---

What public research says about project readiness

Front-end planning and project definition

The National Academies describes front-end planning as the process that defines the project. Its assessment states that decisions made in this phase constrain and support downstream actions and can determine ultimate success or failure.

The Construction Industry Institute states that capital-project delivery is greatly affected by the quality of front-end planning. Its published work describes front-end planning as a structured process for developing sufficient strategic information to address risk and support resource-commitment decisions.

Project definition is therefore not merely a precursor to “real work.” It is the period in which scope, objectives, delivery approach, estimate basis, schedule premise, governance model, contracting strategy, and major interfaces are made sufficiently coherent to support a commitment.

Schedule credibility

The U.S. Government Accountability Office identifies ten practices for a reliable schedule: capture all work, sequence activities logically, assign resources, estimate durations, ensure horizontal and vertical integration, establish a valid critical path, confirm reasonable total float, conduct schedule-risk analysis, update using actual progress and logic, and maintain a baseline schedule.

A milestone date cannot demonstrate those conditions on its own. A schedule becomes decision-relevant when it represents the conditions that govern the work: resource availability, design release, access, materials, work packages, constraints, interfaces, testing, turnover, start-up, and operating acceptance.

Cost credibility and the delivery conditions carrying cost

A cost estimate is not self-contained. Its credibility depends on scope condition, quantity basis, productivity, escalation, schedule, procurement, labor market, material availability, logistics, site conditions, indirect-cost exposure, contingency logic, and management-reserve philosophy.

Historical research on U.S. nuclear construction found that indirect costs and site- and time-dependent factors materially contributed to observed cost escalation. The wider lesson is not nuclear-specific: a capital decision must examine the conditions that influence cost, not merely the estimate presented at a decision gate.

---

Known conditions, foreseeable stresses, and external shocks

Not every adverse development is a black swan. Using that label too loosely permits organizations to describe foreseeable stress as unforeseeable surprise.

Known conditions

Known conditions include incomplete design, unresolved geotechnical information, a thin craft market, weak contractor liquidity, incomplete turnover records, insufficient supervision, constrained work fronts, or a schedule without credible resource logic.

The required response is direct: establish the condition, assign an accountable owner, price and schedule it realistically, and define what evidence is required before proceeding.

Foreseeable stresses

Foreseeable stresses include productivity degradation, vendor delay, rework, loss of a key subcontractor, extended commissioning, labor attrition, adverse weather, or increased financing need.

The exact timing and magnitude may be uncertain. The response should include scenarios, contingency, schedule float, alternate sourcing, step-in rights, continuity planning, and decision thresholds.

External shocks

External shocks include a pandemic, major geopolitical disruption, natural disaster, abrupt regulatory event, or macroeconomic dislocation. Their specific form may not be predictable.

The response is not to predict every event. It is to test resilience: liquidity, contractual flexibility, supply alternatives, workforce continuity, data access, decision authority, recovery capacity, and governance response.

The critical distinction: a project may not be able to predict the event, but it can test whether the delivery system can absorb its credible consequences.

---

Public-record lessons from Muskrat Falls

What the public record establishes

The Muskrat Falls Project was sanctioned in December 2012. The Commission of Inquiry Respecting the Muskrat Falls Project issued its final report, A Misguided Project, on March 5, 2020.

The Inquiry’s terms of reference included examination of the difference between the estimate at sanction—CAD 6.2 billion plus CAD 1.2 billion in financing costs—and construction costs then stated as CAD 10.1 billion plus CAD 2.6 billion in financing costs.

In its executive summary, the Commissioner concluded that alternatives had not been fully explored; assumptions supporting project economics had not been sufficiently tested; a comprehensive examination of possible outcomes had not been undertaken; and the cost estimate had been understated in several ways. The Commissioner further stated that Nalcor’s DG3 estimate had been influenced by optimism bias, strategic misrepresentation, and political bias.

Those are findings of the Commission of Inquiry, specific to the Muskrat Falls record. They are not Torvane findings and do not establish that the same circumstances exist in every major project.

What conditions could have been interrogated more deeply

The point of a public-record analysis is not hindsight theater. It is to identify questions that could have improved the decision record before irrevocable commitments were made.

The compounding mechanism

Muskrat Falls is not best understood as one unexpected event. Its public record shows why ordinary project uncertainties can become more consequential when they interact:

Incomplete alternatives analysis
+ insufficiently tested assumptions
+ understated estimate conditions
+ unresolved site and delivery conditions
+ inadequate challenge and intervention capacity
= reduced option value after sanction

This is a conceptual model, not a quantitative equation. No single condition needs to determine the outcome for the combination to undermine the credibility of the original delivery case.

---

Public-record lessons from Plant Vogtle Units 3 and 4

Read the financial measures correctly

Plant Vogtle Units 3 and 4 were subject to Georgia Public Service Commission monitoring, public filings, Georgia Power disclosures, and prudency proceedings.

The public record distinguishes between Georgia Power Company’s construction and capital-cost forecast, its financing measure, and the customer-recovery limitation approved through the prudency agreement. These categories are not interchangeable and should not be presented as one undifferentiated all-owner project cost.

Georgia Power’s thirty-first semiannual construction-monitoring report, through June 30, 2024, reported approximately $10.174 billion in construction and capital costs for Georgia Power’s project costs and $13.704 billion in total capital cost and financing. The December 2023 prudency agreement capped construction and capital costs recoverable from customers at $7.562 billion. Unit 4 declared commercial operation on April 29, 2024 after completion of required start-up, pre-operational, and power-ascension testing.

FOAK delivery is a set of conditions, not a label

A first-of-a-kind project does not become predictable because it is described as repeatable, modular, or based on a reference design. A design can be repeatable in theory while the project remains first-of-a-kind in licensing, fabrication, site installation, quality records, workforce proficiency, integrated testing, operating procedures, or commercial structure.

Georgia Power’s public filings and releases cited productivity challenges, additional time for testing and quality assurance, construction-quality deficiencies and associated rework, supplier issues, the Westinghouse bankruptcy, and other conditions. These are Georgia Power’s stated accounts in its public filings and releases; they are not presented here as an independent allocation of root cause.

Questions that build credible confidence in FOAK work

• Which systems, modules, methods, interfaces, test procedures, materials, suppliers, controls, or turnover requirements have not been demonstrated under comparable project conditions?
• Which assumptions are based on a reference project, and how comparable is that reference in regulatory basis, climate, site logistics, labor market, fabrication route, quality regime, and completion model?
• What learning curve is assumed for fabrication, installation, quality documentation, turnover, and start-up?
• Does the schedule include credible allowance for rework, inspection failure, record correction, retesting, and system re-entry?
• Which FOAK element can affect the critical or near-critical path?
• Is there a prototype, mock-up, factory acceptance, pilot installation, or staged release that can reduce uncertainty before full-scale commitment?
• Who owns technical integration where a FOAK component crosses design, supply, construction, testing, and operations boundaries?

How to carry contingency on FOAK work

The question is not “what flat percentage of contingency should a FOAK project carry?” A percentage alone does not establish that the project understands its exposure.

The relevant question is:

What range of cost and schedule outcomes is supported by the uncertainty that remains, and what contingency, management reserve, float, alternate capacity, and staged-commitment structure are required to carry that range?

A serious FOAK delivery case distinguishes:

• Cost contingency for defined scope, productivity, escalation, rework, and supply uncertainty
• Schedule contingency and float for uncertain durations, rework, testing, approvals, and interfaces
• Management reserve governed separately for residual unknown-unknown exposure
• Commercial contingency, including alternate-source arrangements, step-in rights, security, and change mechanisms
• Decision contingency, including staged authorization and hold points before full capital exposure is committed

---

EPC contractor solvency and project-continuity due diligence

EPC contractor due diligence cannot stop at the question, “Is the contractor solvent today?”

The more important question is:

Can the delivery organization absorb the project’s credible downside without reducing the capacity, incentive, or financial ability required to finish the work?

A contractor may be technically qualified and commercially competitive while becoming vulnerable if losses on one or several projects consume working capital, bonding capacity, leadership capacity, supply-chain credit, or the ability to retain key subcontractors.

Questions before EPC contract award

• What are the contractor’s liquidity, working-capital position, leverage, debt maturities, and access to additional capital?
• How much of its backlog is fixed-price, capped, reimbursable, or exposed to escalation and productivity risk?
• What are the contractor’s largest active projects, current forecast positions, claim exposure, and expected cash demands?
• How much loss can the contractor absorb before it must reduce staffing, defer supplier payments, seek commercial relief, breach covenants, or impair delivery capability?
• Does the project rely on parent-company support, a performance bond, letter of credit, escrow, or other security? What is the actual recovery path and limit of each?
• Who owns critical design authority, procurement relationships, quality records, fabrication knowledge, and system-completion data if the EPC contractor exits?
• Can the owner step in, replace the contractor, novate subcontracts, access records, preserve warranties, and maintain the schedule in practice—not merely in theory?
• Which subcontractors are single-point dependencies, and which have credible alternatives?
• What happens if the EPC contractor remains solvent but becomes strategically unwilling or operationally unable to continue under the original commercial terms?

Evidence that should exist

• Financial due-diligence review proportionate to contract value and risk structure
• Audited financial statements, current interim financials, and debt/covenant information where available
• Backlog, concentration, and project-loss exposure analysis
• Bonding and insurance-capability confirmation
• Parent-company-support analysis and legal enforceability review
• Requirements for access to project records, intellectual property, vendor data, quality records, and system-completion information
• Step-in, replacement, novation, assignment, and termination plan
• Critical subcontractor map and alternate-provider assessment
• Contractor-continuity scenario for the first 30, 60, and 90 days after a material distress event

Solvency is not a binary prequalification result. It is a continuing delivery condition.

---

Shelf-ready contractor and subcontractor capacity

A project can have a contract without having a credible replacement market.

When a contractor, fabricator, specialist, or critical subcontractor fails, the theoretical right to replace it is of limited value if there is no qualified, available, financially sound, mobilizable alternative capable of entering the work without destroying cost, schedule, quality, or warranty continuity.

Questions that test replacement capacity

• Which work packages depend on a single contractor, fabricator, supplier, proprietary system, certification holder, or specialist workforce?
• Is there a qualified second source, and has it been prequalified for capability, quality, capacity, location, and timing?
• What is the actual time to replace a critical contractor: procurement, mobilization, record acquisition, work verification, warranty transition, supervision rebuild, supply-chain restoration, and workface recovery?
• Would a replacement contractor accept inherited work, latent quality exposure, incomplete records, disputed changes, and a distressed schedule? At what premium?
• Does the owner have rights to use vendor lists, subcontracts, work-package records, quality records, material status, drawings, models, and turnover documentation?
• Has the market been tested recently, or is replacement capacity assumed from an outdated prequalification list?

Why this affects price

If the market knows a project is distressed, under-resourced, late, or carrying inherited quality and commercial problems, replacement bidders price that risk. The project is no longer buying planned work. It is buying uncertainty, urgency, inherited exposure, and mobilization difficulty.

Shelf-ready capacity is therefore not a procurement convenience. It is a component of project resilience.

---

Craft availability and productivity: the delivery-system risk

Craft availability and productivity must be treated as delivery-system conditions, not human-resources variables.

The Associated General Contractors of America’s 2025 workforce survey reported that 88% of responding firms that directly employed craft workers had openings for those workers; 83% of respondents with craft openings reported that those roles were as hard or harder to fill than a year earlier; and 45% reported that worker shortages had delayed at least one project during the prior year.

CII has also documented skilled-craft shortage as a persistent industry condition with implications for project performance. Georgia Power’s Vogtle construction-monitoring materials identified labor availability, labor-rate inflation, competition from other major projects, training, language, fitness-for-duty requirements, strikes, and walkouts as conditions that could adversely affect cost and schedule. Georgia Power’s 2021 revision cited productivity challenges.

Questions before mobilization

• What is the addressable craft pool by trade, experience level, credential status, location, and shift—not merely the national employment estimate?
• Which competing projects draw from the same labor shed, and what are their ramp-up schedules?
• What productivity basis is being used, and how does it reflect local labor conditions, climate, site access, congestion, workface planning, supervision ratio, training, rework, shift pattern, and turnover?
• What portion of the plan depends on newly trained, relocated, transient, or first-time workers?
• What is the supervision ratio by craft and workfront, and how will it be protected during ramp-up?
• Which workfronts will be constraint-free, design-released, material-ready, accessible, and supervised when labor arrives?
• Is labor planned as average headcount or as a time-phased, trade-specific, location-specific productive-hours model?
• What is the productivity consequence if the project obtains headcount but not the required experience or supervision depth?
• How quickly can the project reduce, redeploy, or retain labor if design release, material, access, or weather conditions change?

Evidence that should exist

• Trade-by-trade labor-market study
• Time-phased craft and supervision histogram tied to work packages and schedule logic
• Productivity basis with explicit sources, local adjustments, and sensitivity ranges
• Workforce accommodation, transport, onboarding, credentialing, training, and retention plan
• Workface-readiness measures: work packages, drawings, permits, materials, tools, access, and supervision available before labor mobilization
• Early-warning indicators: absenteeism, turnover, fill rate, overtime intensity, rework rate, constraint hours, supervision ratio, and productive-versus-paid-hours trend
• Labor contingency plan identifying additional sourcing, schedule resequencing, prefabrication, shift changes, scope packaging, and decision thresholds

Headcount is not capacity. Capacity exists only when qualified craft, supervision, released design, usable materials, accessible work fronts, tools, logistics, quality controls, and a credible sequence are available together.

---

Turnover, system completion, and commissioning readiness

The final 10–20% of physical construction is often not 10–20% of the management effort or schedule risk. Remaining work can be fragmented across systems, work packages, vendors, disciplines, quality records, test boundaries, temporary systems, punch categories, operating procedures, training, spares, and owner-acceptance requirements.

The essential question is:

Can the project prove, at component and system level, that the asset is complete, clean, documented, tested, safe to operate, and transferable to the organization that must run it?

Questions before turnover and start-up

• Is systemization defined early enough that construction work, quality records, testing, preservation, punch listing, and turnover align to actual operating systems?
• Does the project track component-level status, or only area-level or discipline-level percentage complete?
• Can the owner and EPC trace each component from installation through inspection, test, defect closure, record completion, system turnover, and operational acceptance?
• How are punch items categorized by safety, operability, preservation, testing impact, and system boundary—not merely counted?
• What are the historical and forecast punch-item closure rates by category, contractor, system, and age?
• What percentage of systems are physically complete but unable to turn over because of documentation, quality records, access, temporary works, vendor support, cleaning, flushing, preservation, or control-system dependencies?
• Does the integrated schedule contain realistic durations for pre-commissioning, cleaning, flushing, drying, loop checks, energization, functional testing, performance testing, operator training, and handover?
• Are system boundaries stable, owned, and reflected in contracts, schedules, work packages, quality records, and the turnover database?
• Is the owner organization sufficiently staffed and trained to receive, operate, maintain, and accept systems as released?
• Does the EPC contractor have documented procedures, training materials, and qualified personnel to transfer knowledge to the owner’s operating and maintenance organization?

Evidence that should exist

• Systemization and turnover philosophy established before late-stage construction
• Component-level completion and quality-record database
• Defined system boundaries and turnover certificates
• Punch taxonomy, aging report, closure-rate trend, and clear acceptance authority
• Integrated commissioning and start-up schedule linked to the construction schedule
• Mechanical completion, pre-commissioning, commissioning, and operating-readiness criteria clearly distinguished
• Owner staffing, training, procedures, spares, maintenance systems, permits, and operating-documentation readiness plan
• Vendor-support and specialist-availability plan
• System-readiness dashboard showing physical completion, document completion, test completion, punch status, constraint status, and owner-acceptance status by system

Illustrative system-completion scenario: lube-oil systems

A project may allocate four weeks for flushing, cleaning verification, reassembly, and readiness demonstration of a lube-oil system. That duration is credible only if internal cleanliness requirements, inspection hold points, temporary filtration arrangements, flushing velocities, contamination control, laboratory or particle-count acceptance criteria, component preservation, vendor requirements, reassembly controls, and documentation are defined before the activity begins.

If internal in-process cleanliness inspections are absent, contamination may not become visible until late flushing, testing, or equipment-preparation stages. The result may be repeat flushing, component removal, cleaning, reassembly, retesting, vendor re-engagement, and rescheduling of downstream commissioning work. The apparent four-week activity can then affect the completion path materially.

The lesson is not that every lube-oil system requires the same duration. The lesson is that a schedule activity has no credibility unless its acceptance criteria, inspection sequence, constraints, interfaces, rework path, and responsible parties are represented in the delivery case.

---

Owner operational readiness is part of project readiness

A project becomes an operating asset only when the organization receiving it is able to accept, operate, maintain, control, and sustain it. Construction completion, mechanical completion, system turnover, commissioning completion, and operational readiness are distinct conditions.

An owner that has not prepared operating procedures, trained personnel, spares strategy, maintenance systems, permits, control-room readiness, emergency response, warranty management, records acceptance, and system-acceptance authority can become the next constraint in the completion path.

Owner-readiness questions

• Has the owner defined who accepts each system, against what criteria, and with what authority?
• Are operating procedures, maintenance plans, spare parts, chemicals, consumables, and vendor manuals complete and controlled?
• Are operators and maintainers trained against the actual systems, controls, and abnormal conditions they will face?
• Are control-room, cyber, permit, environmental, maintenance-management, and emergency-response processes ready for the intended operating state?
• Does the owner have a system-acceptance process that prevents premature turnover while avoiding unnecessary delay?
• Does the EPC contractor have a documented process and trained team to transfer system knowledge, operating limitations, temporary conditions, vendor requirements, and residual risks?
• Are deficiencies being closed by category and consequence, or simply pushed into a generic punch list?

---

The decision thresholds that preserve option value

A risk register records potential conditions. A decision architecture determines what happens when evidence changes.

For each material condition, a serious decision record should identify:

1. The condition and why it matters
2. The evidence supporting the current view
3. The assumption or uncertainty that remains
4. The possible cost, schedule, completion, operating, or capital consequence
5. The accountable owner
6. The fallback capacity or mitigation available
7. The threshold requiring reauthorization, independent review, revised funding, commercial action, or a change in delivery strategy

The correct response to uncertainty is not a longer risk register. It is a decision architecture that identifies the condition, tests its consequence, assigns the evidence required, establishes fallback capacity, and defines the point at which the original commitment must be reconsidered.

---

The Torvane Decision-Readiness Evidence Architecture

This architecture is not a fixed diagnostic, standardized package, or scoring system. It is a set of evidence domains tailored to the decision under consideration.

Project definition

Decision question: What, precisely, is being authorized, financed, bid, awarded, mobilized, recovered, or handed over?

Evidence may include scope basis and exclusions, definition maturity, open decisions, unresolved interfaces, material assumptions, and completion or operating-acceptance requirements.

Delivery path

Decision question: How does the proposed work become a completed, turnable, startable, or operating asset?

Evidence may include integrated schedule logic, work-package strategy, critical-path conditions, constraints, system-completion sequence, and schedule-risk treatment.

Resource reality

Decision question: Who and what must be available, where and when, for the delivery path to remain credible?

Evidence may include leadership capacity, craft, supervision, contractors, vendors, materials, equipment, temporary facilities, logistics, access, and local-market conditions.

Commercial coherence

Decision question: Are responsibility, authority, information, incentive, and change arrangements coherent with the delivery condition?

Evidence may include contract boundaries, interface ownership, change authority, risk allocation, information obligations, escalation routes, and completion responsibilities.

Capital and uncertainty

Decision question: What does the cost and funding case assume, and what conditions could materially change it?

Evidence may include estimate basis, productivity and escalation assumptions, contingency, management reserve, cost-schedule linkage, and scenario thresholds.

Governance and intervention

Decision question: Who sees emerging evidence, who has authority to decide, and what happens when conditions change?

Evidence may include reporting architecture, decision rights, escalation thresholds, independent-review triggers, issue ownership, and reauthorization procedures.

---

Conclusion

The commitment before the commitment is the judgment that available evidence is sufficient to proceed.

That judgment should not be measured by the confidence of the presentation, completeness of a reporting pack, existence of a cost figure, or presence of a milestone date. It should be measured by whether the project’s material conditions have been defined, evidenced, integrated, challenged, and made visible in a form proportionate to the decision being taken.

Major projects are rarely overwhelmed by only one event. More often, the original delivery case becomes fragile because ordinary uncertainties—scope, site conditions, interfaces, labor, productivity, supply, quality, schedule, turnover, operating readiness, and governance—were insufficiently defined, stress-tested, or connected before commitment. A severe external event then exposes a delivery system with little remaining resilience.

No project proceeds with perfect information. The objective is not to eliminate uncertainty, delay a sound project, or replace management judgment with a universal score. It is to ensure that consequential commitments do not depend on conditions that remain unseen, untested, misclassified, or owned by no one.

Where evidence is incomplete, the appropriate response is not always postponement. It may be further definition, targeted verification, a revised estimate basis, a staged authorization, an alternate commercial allocation, a conditional commitment, independent review, replacement planning, or a clearly defined threshold for reconsideration.

A credible commitment is not a declaration that uncertainty has ended. It is a disciplined decision to proceed with a clear understanding of what the outcome requires, what the evidence supports, what remains uncertain, what capacity exists if conditions change, and when the original delivery case must be reconsidered.

---

About this paper

This publication is an independent discussion of recurring major-project conditions based on public sources. It does not describe a specific Torvane client engagement, project, employer, contractor, owner, vendor, or confidential matter.

It is not legal, technical-design, safety-management, quality-inspection, regulatory, insurance, underwriting, investment, or tax advice. It does not certify a project, determine a party’s contractual obligations, or promise any project outcome.

Torvane conducts principal-led, client-defined Project Assessments for decisions that cannot rely on reported status alone. Scope, evidence requirements, field involvement, deliverables, confidentiality arrangements, timing, and fee are defined for the individual assignment.

---

Sources

1. National Research Council. 2001. “Front-End Planning.” In Progress in Improving Project Management at the Department of Energy: 2001 Assessment, Chapter 3. Washington, DC: National Academies Press. https://www.nationalacademies.org/read/10266/chapter/5

2. Construction Industry Institute. 2012. Adding Value through Front End Planning. Publication SP268-3. Austin, TX: Construction Industry Institute. https://www.construction-institute.org/adding-value-through-front-end-planning

3. U.S. Government Accountability Office. 2015. GAO Schedule Assessment Guide: Best Practices for Developing and Managing Capital Program Costs. GAO-16-89G. Washington, DC: U.S. Government Accountability Office. https://www.gao.gov/assets/gao-16-89g.pdf

4. Lovering, Jessica R., Arthur Yip, and Ted Nordhaus. 2016. “Historical Construction Costs of Global Nuclear Power Reactors.” Energy Policy 91: 371–382. https://doi.org/10.1016/j.enpol.2016.01.011. See also Massachusetts Institute of Technology News. 2020. “Study identifies reasons for soaring nuclear plant cost overruns in the U.S.” November 18. https://news.mit.edu/2020/reasons-nuclear-overruns-1118

5. Commission of Inquiry Respecting the Muskrat Falls Project. 2020. A Misguided Project: Commission of Inquiry Respecting the Muskrat Falls Project. Volume 1: Executive Summary, Key Findings and Recommendations. St. John’s, Newfoundland and Labrador: Queen’s Printer for Newfoundland and Labrador, March 5. https://www.muskratfallsinquiry.ca/files/Volume-1-Executive-Summary-Key-Findings-and-Recommendations-FINAL.pdf

6. Georgia Power. 2021. “Georgia Power Announces Revised Schedule, Cost Forecast for Vogtle Units 3 and 4.” July 29. https://www.georgiapower.com/news-hub/press-releases/georgia-power-announces-revised-schedule-cost-forecast-for-vogtle-units-3-and-4.html

7. Georgia Public Service Commission. 2023. “Commission Approves Vogtle Prudency Agreement.” December 19. https://psc.ga.gov/site/assets/files/7846/media_advisory_12_19_23_vogtle_prudency_agreement.pdf

8. Georgia Power Company. 2024. Vogtle Units 3 and 4: Thirty-First Semi-annual Construction Monitoring Report. Georgia Public Service Commission, Docket No. 29849; reporting period through June 30, 2024. https://services.psc.ga.gov/api/v1/External/Public/Get/Document/DownloadFile/219797/100951

9. Associated General Contractors of America. 2025. 2025 Workforce Survey Analysis. Arlington, VA: AGC. https://www.agc.org/sites/default/files/users/user21902/2025%20Workforce%20Survey%20Analysis%20(3).pdf

10. Construction Industry Institute. 2003. The Shortage of Skilled Craft Workers in the U.S. Research Summary RS182-1. Austin, TX: Construction Industry Institute. https://www.construction-institute.org/the-shortage-of-skilled-craft-workers-in-the-u-s

11. Georgia Power Company. 2017. Seventeenth Semi-Annual Vogtle Construction Monitoring Report. Georgia Public Service Commission, Docket No. 29849. Public record copy: https://cleanenergy.org/wp-content/uploads/17VCM_report_083117.pdf