Improving Nuclear Plant Construction Processes: How to Build Projects More Efficiently

If you have paid any attention to nuclear power plant construction projects over the years, you know that there is a long history of cost overruns and schedule delays on many of them. In fact, many nuclear power plants that were planned in the 1960s and 1970s were never completed, even after millions (or billions) of dollars were spent on development. As POWER previously reported, by 1983, several factors including project management deficiencies prompted the delay or cancellation of more than 100 nuclear units planned in the U.S.—nearly 45% of total commercial capacity previously ordered.

More recent projects have not fared any better. In July 2017, SCANA Corp. and Santee Cooper abandoned construction of Units 2 and 3 at the V.C. Summer Nuclear Station in South Carolina. The project was estimated to be about 64% complete at the time. Santee Cooper—South Carolina’s largest public power provider—said the decision was based “in large part on a comprehensive analysis of detailed schedule and cost data.” Santee Cooper said it had spent approximately $4.7 billion in construction and interest at that point in time for its 45% share of the nuclear power project. “The analysis shows the project would not be finished until 2024, four years after the most recent completion date provided by Westinghouse, and would end up costing Santee Cooper customers a total of $11.4 billion,” it said, noting that schedule delays increased the projected interest costs by 143% over the original plan.

Projects in Finland (Olkiluoto 3) and France (Flamanville 3) have likewise faced schedule and cost challenges. Construction officially began on Olkiluoto 3 in 2005 with an expected start date in April 2009. Financial difficulties plagued the construction consortium, which consisted of AREVA GmbH (Germany), AREVA NP SAS (France), and Siemens AG (Germany). The problems—at least partly caused by schedule delays and cost overruns—ultimately forced a breakup of AREVA and a restructuring at Siemens. It was not until April 2023 that Olkiluoto 3 was finally brought online to begin regular production. Meanwhile, construction on Flamanville 3, which began in 2007, is still in progress. Current projections suggest costs for that unit will be four times the original estimates.

Yet, at least one construction expert believes nuclear power plants can be built on time and on budget. “To me, nuclear should be far, far more competitive than it is,” Todd Zabelle, a 30-plus-year veteran of the construction industry and author of the book Built to Fail: Why Construction Projects Take So Long, Cost Too Much, and How to Fix It, said as a guest on The POWER Podcast.

Zabelle said the construction industry could learn a lot about designing projects from the automotive industry. “The way they [the automakers] design things is down to levels of detail and tolerance management that the construction industry could just never even fathom,” he said. “They do that because they’ve got to manage tolerances in space and time on how things come together. They can’t say, ‘Well, we’ve got all the large-bore pipe in there, but we don’t have the small-bore.’ That car is designed to the threads on the nuts and bolts. If you’re going to do modularization—industrialization of construction—you’ve got to be thinking at a level of detail that is not prevalent in the construction industry.”

Furthermore, construction managers must also focus on the supply chain. “Because you moved the work off site doesn’t mean the problem goes away,” he noted. And owners have a big role to play in the process. “The owner has to get educated on how to deliver these projects, because the owner gets the value out of any decisions that are made,” Zabelle said. “You cannot just hand it over to a construction management firm and hope for the best, or EPCM [engineering, procurement, construction, and management firm]. It’s just not going to work.”

Past project management philosophies have not served large complex jobs like nuclear plant construction well. Zabelle said many things still done on projects today stem from processes espoused by Frederick Taylor in the early 1900s. Taylor’s goal back then was to get more production out of each worker, which basically meant, make the workers work as hard as they can. He also separated the planning from the doing. Rather than getting input from the guys and gals actually doing the work, the focus was on just having them do what they are told. However, having schedulers sit in a room planning a project without consulting the people who will be tasked with executing the plan is a recipe for disaster.

Another problem area can arise in the competitive bidding process that often occurs on power projects. Owners typically avoid collaborating with contractors to avoid the appearance of collusion. Yet, this results in a real conflict. Contractors that win bids usually present the lowest-cost proposal, but predicting what will happen over the life of a project, which could span many years, is nearly impossible. Winners sometimes find they have not estimated costs appropriately, which can leave them at risk of bankruptcy as the project progresses. With contractors trying to find ways to turn a profit, projects can suffer.

“What it boils down to is a lot of people doing a lot of administrative work—people watching the people doing the technical work or the craft work—and we become an industry of bureaucracy and administration,” said Zabelle. “Everyone’s forgot about ‘How do we actually do the work?’ That has huge implications because of the disconnect between those two.”

According to Zabelle, the problem can be solved by implementing a production operations mentality. “My proposal in all this is: we need way more thinking about operations management, specifically operations science,” he said. “Not that it’s what happens after the asset’s delivered, but it’s actually a field of knowledge that assists with how to take inputs and make their outputs. The construction industry doesn’t understand anything about operations—they don’t understand the fundamentals.”

In Zabelle’s book, he provides a more thorough explanation of the concept. “Operations science is the study of how to improve and optimize processes and systems to achieve the desired objectives. It involves the use of mathematical models and other techniques to analyze and optimize systems,” he wrote. “It is used to improve efficiency and reduce costs, while ensuring that the quality of the output remains high. Operations science is used to improve the effectiveness of operations, while also reducing waste and improving customer satisfaction.”

Near the end of his book, Zabelle noted that the time for business as usual is rapidly closing. “The pain of the status quo in construction is going to increase exponentially as our capacity to develop and execute projects falls short of expectations,” he wrote. “Until we recognize projects as production systems and use operations science to drive project results, we are doomed to failure. We need to free ourselves from the prior eras and instead focus on a new era of project delivery, one in which projects will be highly efficient production systems that utilize the bounty of the technology (AI [artificial intelligence], robotics, data analytics, etc.) we are privileged to have access to.”

Zabelle sounded hopeful about the future of nuclear power construction. “I truly believe—I would actually throw down the gauntlet—we can make the Westinghouse AP1000 financially viable,” he said. “I’m happy to work with anybody on how to make nuclear competitive because I think it should be and could be.”

To hear the full interview with Zabelle, which contains more about construction projects he’s been involved in, lessons he’s learned over the years, the history behind current construction management processes, challenges and benefits of modular construction techniques, and more, listen to The POWER Podcast. Click on the SoundCloud player below to listen in your browser now or use the following links to reach the show page on your favorite podcast platform:

For more power podcasts, visit The POWER Podcast archives.

Aaron Larson is POWER’s executive editor (@AaronL_Power, @POWERmagazine).

SHARE this article