NARUKAWA Takafumi

NEWS

• May 13, 2022 This web site has been launched.

Dealing with Uncertainty

Although data-driven research utilizing big data is flourishing thanks to the development of machine learning and other technologies, big data is rarely available for nuclear safety research. Often, safety failures occur where there is little data or knowledge.

It is necessary to correctly assess the risk of events that are rare but have a large impact once they occur, as well as unknown events, for which data and knowledge are scarce, and to use the obtained risk information to formulate effective and efficient safety measures.

How can we assess the risk of events for which we have limited data and knowledge? I believe this is where the true value/evolution of engineering related to nuclear safety is tested.

• Simulate and understand phenomena by analysis and experiments.

• Quantify and reduce uncertainties about the phenomena by maximizing the use of evidence using Bayesian statistical methods.

• Integrate the obtained results as risk information, and make decisions with a bird's-eye view of the risks of the entire system.

The foundation is the scientific principle of decision making in uncertain situations.

In order to ensure nuclear safety, I am working on developments in nuclear safety engineering to dealing with uncertainty.