Office for Nuclear Regulation

This website uses non-intrusive cookies to improve your user experience. You can visit our cookie privacy page for more information.

Risk assessments - Are any comparative studies of SMR types, and large reactor types, regarding the possible impacts of extreme accidents in progress?

Information requested

Risk assessments - Are any comparative studies of SMR types, and large reactor types, regarding the possible impacts of extreme accidents in progress?

Information released

In responding to this enquiry we can tell you that ONR are not currently assessing small modular reactors so we can only address your points on what you are referring to as 'large reactor types'.

ONR's web page on new reactor build ( provides links to our assessment reports on both the AP1000 and EPR.  ONR's web page also provides direct access to websites providing safety cases for the EPR and AP1000.  It also provides links to ONR's Safety Assessment Principles (SAPs) and our technical assessment guides covering the three main types of accident analysis techniques used in nuclear power plant safety cases on: (i) probabilistic safety analysis (which covers what we believe you are referring to as chances), (ii) design basis analysis and (iii) severe accident analysis, the latter two of which we believe links to your impacts.  ONR's assessment reports on the AP1000 and EPR and the safety cases in the linked websites also provide a comprehensive set of references to international standards and major work on the three accident analysis techniques.  We believe that this public domain information addresses your points on chances and impacts.

With this information it is important to note that the terminology you use is not commonly used by the nuclear sector where instead of chances words such as frequency or likelihood are preferred and generally rather than impacts the word consequences is used and this is largely related to effects that a radiological release will have on people and the environment.    It is also important to note that a consequence or impact cannot be evaluated without knowing the accident sequence that leads to the impact.  In many safety cases the sequence frequency is discussed at the same time as the impact analysis.  However an accident sequence frequency is derived by combining techniques such as failure modes analysis, faults tree and event tree analysis.  Whereas the impact is evaluated by a transient analysis where such analysis is based on inherent equations of physics combined with relevant nuclear power plant information.  The transient analysis techniques (what you are referring to as impacts) are thoroughly validated by a wide range of what are known as separate effects tests and integral effects tests which are also referenced in our reports.  Please note that for convenience the impacts (consequences) and chances (frequency) are reported together in a combined frequency/consequence analysis within the PSA as the chance of an impact is an important consideration in the UK's regulatory approach.  For your direct questions:

  1. ONR's SAPs document ( is probably the best single source covering the points you raise. However, rather than prescribe reactor designs, sizes or features that may be ruled in or out, or preferred, the SAPs set standards and expectations in terms of engineering requirements and accident frequency/ consequences (see numerical targets) that must be met, with the burden of proof placed on the licensee or vendor.
  2. We are not aware of any work aimed at producing a technical policy limiting the size of a reactor.

Exemptions applied


PIT (Public Interest Test) if applicable