Nuclear energy: The decade of deliverability

It is not due to safety risk: most nuclear energy plants have been operating without incident for decades and the new generation of advanced reactors are designed with more advanced safety systems. It’s not due to a lack of political will: in the U.S., there has been strong, bipartisan support and investment across multiple administrations. It’s not due to regulatory hurdles: regulators have been racing to align frameworks with the needs of today’s projects. Nor is it even due to cost: a number of large consumers and local communities have expressed the willingness to absorb the cost of early, game-changing power sources. As several state public service commissioners, electric utility leaders, and commercial investors have shared, the main thing thwarting near-term deployment of nuclear energy projects is the risk of cost and schedule overruns. For nuclear energy deployment to be financeable and deployable at scale and at pace, the industry needs to demonstrate that it can consistently build nuclear power plants on time and on budget. There are many actions that can be taken to meaningfully address those risks, which will be shared in this article.

A mixed delivery track record

It is critical for any infrastructure project to address the potential for cost and/or schedule overruns. This is true even for first- and early-of-a-kind (FOAK and EOAK) projects. Nuclear energy projects have historically had a more mixed track record of delivery that includes successful completions, significant challenges, and abandoned projects. Historically, project overruns have tended to be significantly larger (double the estimated cost on average) and more frequent in nuclear, compared with other infrastructure projects, especially in the power sector.

A number of projects have been successfully completed on time and on budget. The Palo Verde nuclear plant (Arizona, 1986 and 1988) was not only delivered as expected, but the Arizona Corporation Commission’s auditor estimated that the project management team saved ratepayers over $300 million through their actions. St. Lucie-2 (Florida, 1983) and Kashiwazaki Kariwa-1 (Japan, 1985) also fall into this category. More recently, Saeul-1 and -2 (South Korea, 2016 and 2019) were completed with relatively minor cost overruns.

Delays and cost overruns of recent nuclear projects.

Palo Verde Nuclear Generating Station. (Photo: Arizona Public Service)

More common is that nuclear projects are completed—but with significant delays and cost overages. Vogtle-3 and -4 (Georgia) were completed in 2023 and 2024, respectively, and are successfully powering over 1 million homes and businesses. That said, the units were completed seven years late and at a cost of more than double the initial estimate—largely the result of starting with an incomplete design and not generating an accurate estimate, as evidenced by the significant productivity gains and cost reductions between completion of Units 3 and 4, as Unit 4 had a clearer starting position.

The significant cost overruns for the most recently constructed U.S. nuclear reactors have been used by some stakeholders to advocate against approving more projects without additional financial protection. France’s Flamanville-3, which reached criticality in 2024 but has not yet entered commercial operation, is also an example in this category, completing construction 12 years late and three times over budget.

The most challenging outcome for stakeholders is having projects abandoned postconstruction. More than 50 projects around the U.S. and dozens more around the globe have met this outcome. South Carolina’s V.C. Summer-2 and -3 encountered similar challenges to the early phases of Vogtle-3 and -4, but the issues were too significant for the utility and project partners to resolve, and so the project was abandoned during construction. Several billion dollars had been spent, but there was no resulting operating project to show for it.

The project’s failure has been attributed in part to insufficient planning, alignment, transparency, and oversight. Other examples of nuclear power plants in this category include Indiana’s Marble Hill—abandoned in 1984—and Bellefonte-1 and -2 in Alabama—abandoned in 1988.

Cost overruns can have a variety of sources, including underestimating project costs and timelines, having to reconstruct parts of the plant to resolve manufacturing challenges, needing to respecify or reorder equipment based on design changes, and having tasks that require additional resources.

Schedule delays can disrupt other tasks in designing and constructing the plant, can impact other coordinated activities outside the plant (e.g., meeting new customer demand and replacing older generation), and can significantly inflate costs, particularly given nuclear projects’ relatively high up-front capital requirements—after all, time is money.

The risk and uncertainty in the magnitude of cost and schedule overruns has been a source of hesitation for the industry, and there has been a range of responses to these concerns. Some stakeholders are choosing to back smaller projects that put less capital at risk, like SMRs, despite larger reactors being more mature and having better economies of scale.

Others are opting for a wait-and-see approach, evaluating the deployment of other projects before they invest. Some stakeholders have requested additional financial protections, even beyond what has already been provided by U.S. energy tax credits. Others are relying on gaining experience through repeated project deployments—in the extreme, not worrying about overruns today and believing that if you just build enough, the industry will eventually reach an efficient point. Though each of these approaches may have merit, they also have significant challenges for implementation and effectiveness.

Not a fait accompli

All that said, nuclear energy projects are not destined to fail. Examining the variance in project outcomes shows distinct differences in how projects have been managed. Though much focus is given to reactor technologies and regulatory frameworks, tremendous value can be created by increasing the focus on how nuclear power plant builds are executed and delivered.

There is a host of proven, readily implementable strategies that are capable of mitigating the underlying factors that lead to cost and schedule overruns. Proactively implementing more modern project management and delivery practices that tackle the root causes of overruns is one of the more impactful and actionable ways to make new nuclear a commercial reality. Successfully delivering projects will increase consumer confidence and financial bankability, unlocking even more deployment opportunities.

There are many approaches for effective project management, so what gives—particularly in such a sector with so much know-how? Some strategies are new and have only emerged over the past few years, especially in the realm of technology. Also, although some strategies are well-known and documented by experts, many are not implemented in practice. That can be partially attributed to other factors that project developers may view as superseding, such as a rushed timeline or other external constraints; at other times, it can be attributed to hubris and to misaligned stakeholder incentives. Some also believe in the vital role of nuclear power that justifies deployment at any cost and ignore the dynamics of the increasingly competitive power markets. By taking a more holistic view of project success and actively implementing the best technological, commercial, and interpersonal tools out there, nuclear project stakeholders can chart a better course—and some already are.

Examples of effective strategies

During the Biden administration, I led a U.S. government working group on nuclear energy project management and delivery. Over the past year, I have had the pleasure of meeting with top leaders and subject-matter experts from nuclear and nuclear-adjacent industries, including project developers, architects and engineers, building contractors, technologists, investors, insurers, regulators, labor unions and workers, national and local policymakers, community leaders, academics, advocates, and more.

Through these conversations, a series of actionable recommendations and best practices for project stakeholders were identified, many of which are discussed in this article. (These have been compiled into a broader report that covers a wider set of recommendations that can be implemented across the full life cycle of a project. The full report, Advancing Modern and Effective Nuclear Energy Project Management and Delivery, will be published online later this year.)

Invest more up front in project planning

In many cases, the battle for successful project delivery is won or lost before construction has even begun. For some types of projects, moving quickly and iterating can be effective, particularly where iteration is cheap and the uncertainties are unresolvable. In nuclear, however, because of the high materials and labor costs in conjunction with tight regulatory standards, iterating during construction tends to be very expensive. Investing more time and effort in early project-development activities and iterating during the planning phases rather than during construction is a much more effective way to avoid overruns down the line.

Design completeness—starting with complete (or at least high-fidelity) designs—can significantly derisk the execution of a project. Because this strategy requires more time and resources at the front end before construction, it could potentially add a little cost and elongate the development process, but it is more likely to pay dividends by preventing unexpected costs during construction, when it is significantly more expensive to make changes.

Though this may seem obvious, this has been a pitfall of several nuclear projects. Take ­Vogtle-­3, for example: For commercial reasons, project construction commenced even though the design was not even close to finalized. As the project progressed, several design changes were made, which resulted in expensive reworking along with construction and licensing delays.

A complement to design completeness is accurate estimation: being able to generate high-­fidelity forecasts of project costs and timing before construction has begun. A number of projects have gone over budget because initial cost estimates were either not detailed enough or were based on insufficient up-front planning. Working from complete and proven designs increases accuracy in estimations.

In projects involving new features, developers may not know the true costs, and issues can arise during construction that are harder to predict. Moreover, it is important to discourage developers from accidentally or deliberately underrepresenting project budgets in order to gain an advantage during selection, only to have them request budgetary increases further down the line. To help with this, having project owners and/or reviewers hire experienced independent engineers and planners is helpful. Adopting milestone-based approaches, where major elements of a project are reviewed extensively at key decision points, also can reduce the likelihood of unpredictable issues further along.

In some instances, project stakeholders may want to proceed without a complete design. There may be a rush to meet exogenous deadlines (such as tax credit requirements), development activities might be parallelized to shorten preconstruction timeline, or it could be a FOAK/EOAK build where there is more uncertainty.

Having additional safeguards in place is even more critical in these instances to avoid significant downstream impacts during other phases like licensing, staffing, and of course construction. Getting as far along as possible (e.g., at least starting with the equivalent of an AACE Class 2 estimate) will still reap dividends and reduce the likelihood of adverse outcomes.

For FOAK/EOAK projects, where there are elements with less certainty and it may be harder to get an accurate cost estimate because of novel elements, creating strong analogies to similar but known parts and using techniques like reference class forecasting can be helpful to bridge the uncertainty gap. Also, verifying the manufacturing capability of key suppliers, performing labor availability risk analyses, and having the building contractors validate constructability should help identify potential pitfalls.

Workforce development

Vogtle Unit 4 steam generator placement. (Photo: Georgia Power)

Nuclear projects are very labor intensive, typically requiring thousands of skilled workers across a wide variety of trades and occupations. At its peak, the Vogtle-3 and -4 project employed 9,000 construction workers. This need becomes even more critical at scale: To meet U.S. deployment goals, the industry will need hundreds of thousands of new construction workers (welders, electricians, pipefitters, laborers, and schedulers). Project delays can mean thousands of workers are left waiting, which incurs downtime costs and could have knock-on delays for other projects.

Labor unavailability due to poor up-front planning, the need for workers with specialized skills, or work stoppages can also cause delays. Having a robust labor strategy is incredibly important and should be proactively optimized for a given project, accounting for the available labor pool, labor requirements, and project schedule. Timely and efficient deployment of an experienced and skilled workforce is key for ensuring project deadlines are met, minimizing quality issues and rework, and spotting potential constructability issues with FOAK/EOAK designs.

Nuclear energy projects are complex and must be built with precision to tight specifications, especially to comply with regulations. A properly trained and certified workforce—for instance, from registered apprenticeship programs and labor unions—is a tremendous asset. On top of safety benefits, building with precision helps to eliminate rework. In addition to reducing time expended and delays, a trained workforce can help avoid large, unplanned capital expenditures (e.g., ordering replacement parts, pouring new concrete).

Furthermore, unions can meaningfully improve workforce productivity and reduce project costs. Many of these workforce organizations also have strong training programs that provide an avenue to pool and transfer knowledge in ways that can benefit multiple projects—that is, they form one of the more important repositories of construction and execution know-how that are important to create a robust learning curve quickly.

Project owners and contractors, especially for FOAK/EOAK projects, should prioritize staffing for uncertainty and longevity. While up-front planning provides significant leverage, developers should expect the unexpected and naturally anticipate that there may be circumstances that require changes. Having a direct and robust contingency plan is key.

Proactively overstaffing—hiring a few extra workers for longer periods of time during high-risk phases of the project—also can be a helpful risk mitigation strategy. The additional cost is outweighed by the benefit of avoiding delays. Furthermore, deploying planning and scheduling experts at multiple levels throughout the organization can yield dividends, as changes can be more quickly implemented and downstream and interdependent tasks can be derisked.

As nuclear energy projects have longer development cycles and are more labor intensive, it is important to manage and monitor workloads throughout the entire life cycle. Labor surges (e.g., extended work weeks, additional shifts) can at times be helpful to accelerate delivery, execute or correct critical path actions, and reduce congestion. However, some studies suggest that excessive use of labor surges leads to worker fatigue, losses in workforce productivity, more rework, and other unintended consequences that may inadvertently prolong construction and increase costs. A strong labor plan needs to carefully factor this in, in tandem with other practices, to ensure project teams are well positioned to minimize this risk.

Project labor agreements (PLAs) are important ways to structure and implement all aspects of a project’s labor strategy into actionable plans and contracts. They are signed early in the project development cycle, typically well before construction starts, and they outline staffing plans, commitments by labor partners to provide staff, and paths for training and prequalification of workers in advance. PLAs are key tools in ensuring workforce availability and worker input and expertise (including on operational efficiencies, health, and safety) throughout the project that can significantly mitigate cost and schedule risk.

Commercial structures to drive greater alignment

Nuclear energy projects can be complex to commercially contract for and underwrite. Larger projects in particular require significant up-front capital for development and construction. Modern contracting and commercial strategies can provide powerful tools to create aligned incentives among project stakeholders and promote greater collaboration. There are several useful approaches, including striking contracts with index-based cost-adjustment mechanisms. Here, we will focus on the importance of aligned multiparty delivery agreements.

Traditionally, most nuclear projects have involved individual tasks that are bilaterally contracted and subcontracted out to various parties (e.g., designers, builders, EPC [engineering-procurement-construction] firms, suppliers, and others) under separate agreements. Though easier to execute, this approach can create tension and incentive misalignment at the project level—especially with FOAK/EOAK projects. For example, if there is unplanned work driving a cost overrun at the project level, it could theoretically be a gain for the contractor. As a result, the contractor may not be as incentivized to proactively address or avoid the need for extra work. Conversely, the contractor could be unfairly blamed for delays and cost overruns resulting from unexpected changes made by project leaders upstream in the design process.

Integrated project delivery (IPD) is a newer, more collaborative project model that brings together key project delivery stakeholders (e.g., owner, builders, suppliers, and others) under a single agreement. It has multiple benefits that help ensure all parties are maximally oriented toward project success. With a no-blame culture and alliance-based decision-making, it provides a powerful way to create the collectivism of a vertically integrated owner and delivery organization with the ability to optimize and bring in the best expert stakeholders for a particular project.

Financial alignment is created through a shared risk-reward pool, and each delivery partner’s compensation is formulaically tied to overall project performance (e.g., actual final cost versus the agreed-on cost target). Though relatively new, there are a few examples of IPD-like structures being employed in the nuclear industry. At the Darlington project in Canada, an IPD is in place with Ontario Power Generation as the owner and utility, AECON as the EPC, GE Vernova as the technology provider, and Atkins Réalis as the design engineer.

Energy Northwest is also implementing a variant of the IPD structure for its project in Washington state, which involves creating a team and compensation structure explicitly tied to the success of the overall project. This creates a shared accountability mechanism among stakeholders. This alignment can better leverage each party’s expertise to see around corners and anticipate issues ahead of time. It can be commercially and contractually trickier to set up, but the reported benefits have greatly outweighed the costs throughout the project.

Active and modern leadership

Strong project management starts and ends with strong project leadership. This is one of the most important drivers for improving project delivery, and it underpins the success of nearly all of the other strategies described. Governance, oversight, and culture are foundational to driving stakeholder alignment, ensuring accountability, making sure the various facets of the project are well-structured and coordinated from the start, and ensuring that the project can adeptly respond to any changes as it proceeds.

The anchor of this strategy is proactive and engaged project owners and operators who help lead an experienced delivery project team that includes architects, constructors, planners, equipment vendors, and others. As is the case with other large infrastructure projects, especially FOAK/EOAK projects, this is critical for creating a strong culture where all stakeholders are oriented to maximize project success.

Even if the project owners are not nuclear experts themselves, it is vital for them to be heavily involved in major decisions, closely track project deliverables, provide guidance, ensure accountability, and be accessible to project team members. This should include creating avenues for stakeholders to quickly and safely escalate critical issues to the leadership teams when they arise to ensure rapid issue remediation and prevent smaller challenges from snowballing into larger ones.

The complement to 21st-century project leadership is leveraging robust, dynamic digital tools to help stakeholders stay connected, informed, and on track. This can include document management systems, which help accurately track project deliverables and changes and can additionally assist regarding compliance with nuclear regulations, which reduces the risk of delays. It also can include plant computer-aided design schematics, site metrology data, work schedules, procurement systems, equipment delivery schedules, and so on. Interlinking these software programs can greatly increase the power of these tools to assist the project. Adding cross-cutting analytics, like with artificial intelligence, can augment the power of deploying these tools.

There is a wide range of more traditional solutions, like Microsoft Project and Oracle Primavera, which provide some of these features, and other offerings that are aiming to bring newer and wider functionality more recently have started to emerge, such as those from Hyundai, InEight, Westinghouse and Google, and The Nuclear Company and Palantir.

These programs can be employed in all phases of the project: helping identify potential issues and analyzing the impact of potential contingency management strategies in the early design phase, training workers, ensuring adequate visibility for the project delivery team, and providing additional troubleshooting and accountability mechanisms.

Making relevant information on the project public can also help maintain community and local governmental support. At the same time, to ensure project team members are not overwhelmed with the potentially high volumes of data, the project team should develop key performance indicators (KPIs) tailored for the various team members. These digital tools can help consistently track those KPIs and summarize the data into relevant and actionable insights for the team members and other stakeholders.

On top of planning and tracking, an underutilized but high-value process involves incorporating project early warning systems. Having ways to detect when an ongoing project is veering off track is critical to minimizing the magnitude of project risk. One such approach is predictive project analytics, which is a suite of mostly quantitative methods and techniques to forecast task and project outcomes, especially on larger-capital projects. This known technique used by some in construction and insurance does not appear to have been widely deployed in nuclear construction thus far.

Project developers can anticipate potential issues under a range of scenarios and develop strategies to address them. Predictive project analytics also can be supercharged by integrating project software tools, leveraging AI, and creating virtual simulation environments and digital twins. Furthermore, whenever project plans do change, these systems enable the potential downstream impacts to be quickly predicted, flagged for stakeholders, quantified, and mitigated.

Creating a better path forward

All stakeholders need to proactively address strategies to mitigate the risk of cost and schedule overruns. Implementing a suite of modern project management and delivery best practices, as described above, can play a significant role in reducing those risks at the source. Though many of these strategies are actionable, adoption has historically been relatively spotty. So, here are a few recommendations for the broader set of stakeholders—especially local and federal policymakers, financial investors, and other groups promoting deployment—to consider when evaluating individual projects or developing deployment programs for the sector:

Urge and incentivize project developers to demonstrate what steps they have taken to mitigate cost and schedule overrun risk. This could include providing direct financial support for early development activities and integrating relevant strategies into funding eligibility criteria.

Encourage the formation of project deployment consortia; cost- and risk-sharing partnerships and joint ventures (even including broader groups of deployment stakeholders who may benefit down the line); and standardized, more productized deployment models.

Promote the formation of multi-project deployment cohorts between states and/or countries (e.g., the National Association of State Energy Officials’ First Mover Initiative) and others to streamline local regulations, standardize design requirements and contracting structures, and encourage information sharing.

Sponsor the formation of centralized, accessible project data repositories that are available to the public and prospective deployment stakeholders to accelerate the ability for industry stakeholders to incorporate learnings into new projects.

Nana Menya Ayensu was the special assistant to the president for climate policy, finance, and innovation in the White House during the Biden administration and has spent his career developing innovative energy infrastructure policies.