Scheme showing the technical scope of NEA ICDE project activities, NEA 

Common-cause failure (CCF) events can significantly affect the availability of safety systems of nuclear power plants. In recognition of this, CCF data is systematically collected and analysed in several countries. Due to the low probability of occurrence of such events, it is not possible to derive a comprehensive evaluation of all relevant CCF-phenomena only from the operating experience from one individual country. Therefore, it is necessary to make use of the international operating experience from other countries using similar technology. The usage of international NPP operating experience with CCF requires a common understanding of what CCFs are and how to collect data about them. To develop a common understanding, an international common-cause failure working group was founded in 1994. This working group has elaborated the project International Common-Cause Failure Data Exchange (ICDE) which the NEA has formally operated since 1998.

The ICDE project collects qualitative and quantitative information about CCFs in NPP, analyses the collected data and distributes the gained insights about CCFs and methods to prevent CCFs, as reports to the concerned professional audience. The objectives of the ICDE project are to:

• provide a framework for multinational cooperation;
• collect and analyse CCF events over the long term to better understand such events, their causes, and their prevention;
• generate qualitative insights into the root causes of CCF events which can then be used to derive approaches or mechanisms for their prevention or for mitigating their consequences;
• establish a mechanism for the efficient feedback of experience gained in connection with CCF phenomena, including the development of defences against their occurrence, such as indicators for risk-based inspections;
• generate quantitative insights and record event attributes to facilitate quantification of CCF frequencies in member countries; and
• use the ICDE data to estimate CCF parameters.

The ICDE project operates with a clear separation between data collection and analysis. The analysis results mostly in qualitative CCF information. This information may be used for the assessment of 1) the effectiveness of defences against CCF events and 2) the importance of CCF events in the probabilistic safety assessment (PSA) framework. The qualitative insights on CCF events generated by the analysis are being made public as CSNI reports.

The data collection and qualitative analysis result in a quality assured database with consistency verification performed with the project. ICDE activity defines the format for collection of CCF events to achieve a consistent database. This task includes the development and revision of a set of coding guidelines describing the classification, methods, and documentation requirement necessary for the ICDE database.

The general coding guidelines have been developed during the ICDE project and are continually revised. General Coding Guidelines, Issue 5 (April 2026)

They describe the methods and documentation standards necessary for the development of the ICDE database and reports. The format for data collection is described in the general coding guidelines and in the component-specific guidelines. Component-specific guidelines are developed for all analysed component types as the ICDE project evolves. So far, the ICDE Project is the only international effort where large amounts of data from different countries are collected and analysed to draw conclusions about CCF.

ICDE data collection also includes potential CCF events, or ICDE Events, which include impairment of two or more components (with respect to performing a specific function), which exists over a relevant time interval and is the direct results of a shared cause.

The ICDE Project shares the engineering insights of its analyses in public reports published by the NEA Committee on Safety of Nuclear Installations (CSNI).

The ICDE project is based on the willingness of the participants to share their operating experience; to encourage that, participating organisations get access to the database in accordance with their own contributions to the data collection. The project has successfully completed the previous eight phases from 1994 to 2022 and is currently achieving its Phase 9 until December 2026. A phase 10, running from January 2027 to December 2030 is under preparation.

Two reports have been published summarizing project phase VII and phase VIII.

• Summary of Phase VII (June 2019)
• Summary of Phase VIII (June 2025)

Specific component type reports have been published on collection and analysis of CCF events of:

• Centrifugal pumps
• Emergency diesel generators
• Motor-operated valves
• Safety and relief valves
• Check valves
• Batteries 
• Level measurements 
• Control rod drive and assemblies
• Switching devices and circuit breakers
• Heat exchangers

Currently, data analysis and exchange cover the following component types in addition to the list of component types above:

• Main steam and isolation valves
• Fans
• Digital I&C
• Inverters
• Cross-component CCFs

Topical reports have been published on a number of different topics that examine specific CCF themes and recurring patterns across the collected operating experience:

• Diesel events affecting the entire exposed population  (143 events)
  Provides insights into failure mechanisms that can affect all redundant emergency diesel generators and identifies possible improvements in design, maintenance, testing and management control.
• CCFs due to plant modifications (53 events)
  Highlights how modifications, replacements and change of settings can introduce CCFs, and underlines the need for systematic impact assessments and post-modification testing.
• Provision against CCFs by improving testing  (59 events)
  Identifies cases where existing tests failed to reveal component impairments and provides lessons for improving test scope, quality assurance and verification of operability.
• Multi-unit CCF events (87 multi-unit events)
  Provides insights into shared dependencies between units at the same site, including design, organisational, environmental, and also fleet-related factors.
• Inter-system dependencies (25 intersystem events)
  Shows how a single failure mechanism can affect multiple systems or safety functions, and why such dependencies may need to be considered.
• External environmental factors (64 events)
  Provides operating experience on CCFs triggered or influenced by external and environmental conditions, supporting both external event modelling and the identification of preventive measures.
• Pre initiator human failure events (51 complete CCF events)
  Provides insights into latent human-induced dependencies, such as misalignments, miscalibrations and maintenance errors.

In addition, two topical reports—on safety culture, and on the ICDE quantification procedure and data interpretation in component-specific quantitative applications—are under preparation. The project is also currently exploring topics related to CCF events associated with ageing and I&C. 

ICDE members area (password protected | reminder)