But turning one-offs into a regular production line is trickier than it might first seem. The Synthos Green Energy (SGE) development platform has applied, through the government’s Advanced Nuclear Framework, to build and deploy 14 GE Vernova Hitachi BWRX-300 reactors across three UK sites. The proposed £35bn programme would comprise six 300MW units at its first site and four at each of the other two, providing 4.2GW new capacity in total, with a goal of first power in 2034.
Government-backed programmes are also joining the fray, including at Wylfa, where Great British Energy – Nuclear (GBE-N) signed a contract with Rolls-Royce SMR in April for three reactors providing at least 1.4GW capacity. GBE-N is also procuring a delivery partner, under a contract worth around £1bn, to provide programme management, engineering and risk expertise across the roll-out.
While the projects are good news, they will require a shift in how we think about nuclear power, according to experts. “Nuclear, to date, has been a series of almost standalone projects,” says Andrew Bowfield, a civil nuclear expert at the Manufacturing Technology Centre in Coventry. “You have to reinvent the wheel every single time.” The UK should build fleets instead, he says.
Bowfield estimates that 70-80% of a nuclear power plant involves thermal generation tasks such as moving heat and fluids, turning turbines and exporting electricity, all of which are not nuclear-specific. There is no reason, then, why the pumps, valves and switches could not be borrowed from approaches used elsewhere.
There are other ways to automate the system. Jeremy Stain, nuclear business leader for the UK, India, Middle East and Africa at Arup, says the nuclear island – including the reactor pressure vessel, primary cooling systems, refuelling equipment and spent-fuel pools – could largely be standardised independently of the site.
Not everything can be, however. The size and type of cooling systems is dictated by the available water source and grid connections on existing transmission infrastructure, while ground conditions and environmental permits also play a role. But even cooling infrastructure could be assembled from repeatable mechanical-draught cooling tower modules, Stain argues.
If the UK is to standardise its SMRs, then fleet-wide configuration management will be critical. Changes made on one unit cannot simply be copied to the remaining units without understanding the consequences on safety and manufacturing. At the same time, allowing each site team to optimise its own reactor would end up creating the same ‘bespoke megaproject’ model the UK has used for years.
Finding that Goldilocks point between modularity and standardisation could be aided by looking at other industries. Aircraft manufacturers require suppliers to maintain a stable architecture, use agreed materials and target specific prices, but allow firms room to innovate in how components are produced. Shipbuilding takes the approach of dividing large, safety-critical structures into modules manufactured separately and assembled later.
Both those industries have stable order books, which encourage investment. Nuclear does not have that yet, but the UK’s emerging pipeline could help encourage standardisation. Alongside the SGE and Rolls-Royce fleets, Holtec International, EDF and Tritax have proposed an £11bn nuclear-powered data-centre campus at the former Cottam coal station in Nottinghamshire. The East Midlands Combined County Authority has also submitted an AI Growth Zone proposal covering Cottam and High Marnham, linking nuclear deployment to growing demand for computing power.
There is a fine line to draw, Bowfield says, between improving cost and speed by standardising production without weakening quality. Having multiple vendors and reactor types is an advantage because they do not all rely on the same supply chain.
Regulation is nonetheless moving towards the idea of fleets of standardised facilities, rather than one-and-dones. The BWRX-300 completed Step 2 of the UK’s Generic Design Assessment in December 2025, while British and US regulators have agreed to reduce duplicated work on advanced reactor assessments. But it would still require further detailed assessment before safety-significant construction could begin at a British site.
The clearest sign of industrialisation, Stain argues, will be vendors and suppliers committing money to centralised factories that repeatedly meet nuclear quality and throughput targets for several projects.
The UK will need to move quickly to secure that cash, with demand growing around the world. “There’s an international ship currently sailing right past the UK,” Bowfield says. “I want it to stop here and take on loads of expertise and capability, which we absolutely have.”
Want to learn more about SMR technologies and use cases? Next running in Manchester on 6 October, IMechE's Small Modular Reactors (SMRs): Concepts, Technology, Challenges and Opportunities training course will explore the key factors involved in SMR development and commercialisation in the UK and globally. Learn more and check for future dates on the training course page.
Want the best engineering stories delivered straight to your inbox? The Professional Engineering newsletter gives you vital updates on the most cutting-edge engineering and exciting new job opportunities. To sign up, click here.
Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers.