[3F07-09] AI Application

3:00 PM - 3:15 PM JST(6:00 AM - 6:15 AM UTC)

[3F08] Development of Probabilistic Risk Assessment Methodology Using Artificial Intelligence Technology(5) Application of Automatic Failure Detection Methodology to Secondary Analysis for Reliability Database Construction

*Hiroshi UJITA¹, Tatsuya MORIMOTO¹, Satoshi FUTAGAMI², Hidemasa YAMANO², kenichi KURISAKA² (1. AdvanceSoft Corporation, 2. Japan Atomic Energy Agency)

Keywords:

Generative Artificial Intelligence,Probabilistic Risk Assessment,Reliability Database

To analyze the causes of troubles from the PRA reliability database such as NUCIA, we are trying to utilize ChatGPT considering the groundbreaking performance improvements in generative AI in recent years. A feature of ChatGPT is that it has become possible to illustrate the causal network structure by defining a task ontology that simulates a network including series and parallel causes. Furthermore, in response to the issue of the absence of a subject, which is a characteristic of Japanese, making actors (personnel) unclear, we have gained the prospect of clarifying personnel by defining the organization's management to field workers as a domain ontology.