Cutting the cost of nuclear

1 April 2000

Declining fossil fuel prices and the introduction of the combined cycle gas turbine have eroded the competitive edge of nuclear power plants over the last 15 years. An OECD report discusses how the nuclear industry could build cheaper plants and make a comeback.

If countries are serious in their commitment to mitigate climate change and sustain future development, nuclear power will reappear as a viable alternative to fossil-fuel generation, but it can only do this by reducing its costs. As capital costs make up around 60% of the total cost of nuclear generation, this is where the cutting should begin.

Based on a synthesis of experience, a group of experts selected from NEA member countries have agreed a number of capital cost reduction measures and compiled them into a new report*. The measures are:

•Increased plant size.

•Improved construction methods.

•Reduced construction schedule.

•Design improvement.

•Improved procurement, organisation and contractual aspects.

•Standardisation and construction in series.

•Multiple unit construction.

•Regulatory and policy reform.

Is bigger better?

Many investigations have varied the economic parameters of nuclear power plants with plant size and capacity, producing much uncertainty. In the 1970s and 1980s, experts studied the implications of economies of scale when construction time schedules and costs had increased steeply, and the spread of performance between excellent and poor had become vast. The scarcity of new orders in recent years has done little to make comparisons easier. The specific costs of large nuclear plants are quoted within such a large band that the derivation of scaling factors is very difficult. Savings arising from increasing reactor unit and plant sizes are obscured by other factors – improved construction techniques, shorter construction schedules, the construction of multiple-unit plants, etc.

Generally speaking, bigger is better. Economies of scale have reduced the cost of power production more than any other factor. In Canada for example, the specific cost of building a single 881MW plant instead of a 670MW plant (a 31% increase in unit size) will result in a 12% saving in specific capital cost (US$/kW). Similarly in France, the cost of building a single 1350 MW plant instead of a 1000 MW plant (a 35% increase in unit size) will lead to a 13% saving in specific capital costs.

But uncertain load growth, cash constraints and the longer time it takes to build larger plants are defining a new planning regime for some utilities. For them it may be too risky to commit scarce capital to build a large plant that must be committed many years in advance of the anticipated need. A mismatch between installed capacity and demand in any direction carries substantial costs. Economies of scale may be limited by the dimensions of some systems or components.

Standardisation, and building units in series may contain the greatest potential for capital costs reduction. Standardising the design, manufacturing, construction, licensing and operation approaches developed for a first project consolidates plant safety and avoids first-of-a-kind costs such as functional studies, power block layout, detailed equipment design, safety studies and drawing up testing and commissioning procedures. EdF and Framatone have conducted a study that shows that the average cost of one unit in a many-unit programme decreases by 20-40% compared to the cost of a single unit, depending on the pairs of units at a site. The effects of standardisation in Korea resulted in cost savings of between 15% and 20%. Similar effects by replication and series construction are reported from Canada and the US.

Further savings can be realised when more than one unit of the same design is built on one site. In France, the specific capital cost of building two plants on the same site will lead to a 15% saving in specific capital cost, as compared to building only one unit. Data from Canada show the same level of savings.

In addition to the obvious sharing of site acquisition and land costs, multiple unit plants share site-licensing costs. Significant cost savings are created by using common facilities and sharing the costs of services such as: environmental impact and hazard studies, administrative procedures, water intakes and outfalls, roads and utility networks, on-site fire protection and general platform earthworks.


The first building techniques for nuclear power plants were adopted from fossil power plant experience.

Until recently, development and improvements in this area were driven by the need to respond to regulatory requirements and quality assurance concepts, not to reduce costs. But there are efficient and cost effective means of improving plant design to reduce costs and construction schedules. A good example is open top access.

Open top construction allows access to the inside of the reactor building from outside the perimeter walls. Large heavy lift cranes move equipment through the top of the reactor building rather than using the traditional horizontal technique. This means that a steam generator can be installed in two days instead of two weeks. Full use of the open top technique will reduce construction time, but the consequential cost reduction will depend on the cost of hiring heavy lift equipment and the prefabrication of modular components.

Open top access has led to savings of 2.4% of total plant cost in Canada, and to a 15% reduction in construction schedule.

Modular construction is important to cutting capital costs. Modules can be prefabricated in a factory for structural assemblies, piping, tubing, cable trays, reinforcing bar mats, instrumentation, electrical panels, support and access platforms, etc. A factory setting provides a better working environment than the plant site, as activities have a higher quality assurance, with the potential for greater automation and higher productivity. Modular construction reduces plant congestion and shortens the construction schedule. It also reduces site labour and supervision, although these reductions may be offset by increased transport and factory equipment costs.

Modularisation can save as much as 4% of the total construction cost but, although promising, its applications are not yet significant, mainly because of the requirement to complete the total design of the plant and buy all the materials and components before module fabrication.

Other construction methods that can reduce costs are: slip-forming, parallel construction, improved instrumentation and control cabling, formed pipe elbows (to eliminate welding and welding inspections) and contractor sequencing.


The design engineering cost of a nuclear power plant accounts for around 15% of the total capital costs. Advances in computer technology and electronics have lowered construction costs by leading to consistent documentation and more accurate material lists.

Improved design can be achieved in a number of areas:

• Plant arrangements.

• Accessibility.

• Simplification of design (multi functional supports, control process information display, computer control etc).

• Application of computer technology and modelling.

Computer-aided design (CAD) and computer-aided engineering (CAE) have become invaluable in developing the optimum construction management and schedule. The ability to visualise and animate various construction scenarios and evaluate conflicts, risks and critical path activities improves material delivery and ensures the most efficient use of manpower and resources.


Owing to the large capital investment involved, nuclear projects are more sensitive to delays in project completion than conventional power projects. The timing of a specific expense and its effect on interest during construction can add more than 25% to the capital cost.

Apart from accumulating interest, construction delays can escalate equipment, material and labour cost. Technology can become outdated, licences can expire and public opposition can grow to a state of unrest.

A shorter construction schedule reduces many variable costs, and leads to early profits. It can be attained by introducing a number of measures into construction, such as: advanced engineering methods; simplified reactor base construction; modularisation techniques; prefabrication of the reactor liner and primary and secondary shield walls; using heavy lift cranes; reducing embedments; up-front engineering and licensing; effective change control, project planning, monitoring and control; and improving manpower development and training.

Multiple shift work will maximise working hours for a rapid completion, as will computerised project management scheduling and contingent procurement. But the cost savings that result from a shorter construction schedule will be compromised by increased risk from more work being carried out at the same time, and a large preconstruction investment in access logistics and site preparation.

The NEA’s expert group looked at how construction times could be improved for different reactor types. One example shows the schedule for a 1200 MW PWR successfully reduced from 63 to 55 months, through modularisation, better access logistics and improvements to the design process.


An important element in project management is the contracting strategy applied to procuring equipment and materials.

As electricity markets have opened up, procurement has become competitive. For nuclear generation, this has happened quite slowly because of the regulatory and safety aspects involved. But, as the cost of building a nuclear plant represents 60% of the cost of the power it produces, compared to 20% for a gas-fired plant, the deregulation of markets will be far more significant in the nuclear sector.

At one end of the procurement spectrum is the turnkey approach. A single vendor supplies all the equipment and co-ordinates the overall construction work. He bears the bulk of the cost and risk, and interfaces between owner and contractor are minimised. This is an expensive approach.

At the other end of the spectrum are multiple package contracts, wherein the owner conducts a multiple-contract procedure for the supply and installation on-site of several hundred items of equipment. The owner has to spend more on co-ordination and interface control, and he will have to supervise operations at the site.

Between the turnkey approach and multiple package contracts, there are a number of options that reduce technical co-ordination and certain interfaces. The optimal balance will differ according to the country’s nuclear infrastructure, the engineering resources of the owner and the number of plants that will be built in series. The fact that there are several solutions means that they can compete with each other. This in itself should lead to capital cost reductions.


Experience in the US illustrates the influence that licensing activities can have on power plant costs. In the late 1960s and early 1970s, around 50 US utilities embarked on separate and independent nuclear power development programmes. In response, at least six nuclear vendors, 20 architects/engineers and 26 construction contractors entered the new market to supply equipment, materials and services. The end result is 54 utilities managing 110 plants, most having unique design and operating characteristics.

Each of these plants has had to contend with a growing number of regulatory requirements, particularly after the Three Mile Island accident. NRC-related changes during the 1970s increased capital modification costs to the tune of around 60%.

The US nuclear industry has complained that existing regulations are overly prescriptive (a legacy from early days when knowledge of nuclear safety was limited). It also claims that the NRC constantly changes its interpretation of requirements, and that its use of generic communications lacks rigour.

Now that the industry is mature with over 2000 reactor-years of safe operation behind it, there is broad agreement between the industry and the NRC to move toward risk-informed performance-based regulation. In this approach, the regulator would establish basic requirements and set overall performance goals. Plant management would then decide how to meet these goals. If the management is not happy with performance or fails to meet the operating criteria, it can fine-tune the programme as needed without having to go through a protracted regulatory process for approval.

Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.