Understanding Human Error and Aiding Human Diagnostic Behaviour in Nuclear Power Plants

Author(s):  
Thomas B. Sheridan
Author(s):  
Yoonik Kim ◽  
Kwang-Won Ahn ◽  
Chang-Hyun Chung ◽  
Kil Yoo Kim ◽  
Joon-Eon Yang

Organization can make influences on all the systems. Especially in case of nuclear power plants in which safety is established to be one of the most important operating goals, there have been a lot of research efforts for the hardware advancement. However in recent years, it has been widely recognized that organizational factors in nuclear power plants have an important influence on the safety attitudes and the safe behavior of individuals. Until now, any means to include assessments of organizational structure in probabilistic risk assessments have not been universally accepted. The objective of this work is to develop a method to assess organizational influences on component maintenance. Influence diagrams are introduced in this method as a decision making tool and fuzzy theory is used to reflect the vagueness in considering relevance of human activities in maintenance tasks. Introducing fuzzy theory to assess the organizational factors is deemed to a somewhat new trial, which makes it possible to convert linguistic vague descriptions into mathematical ones. Fuzzy linguistic descriptions offer an alternative and often complementary language to conventional, i.e., analytic approaches to modeling systems. Among the existing methodologies to assess organizational factors, the concept of the ω-factor model is utilized and the mechanism that organizational factors have influences on component maintenance is evaluated through composing influence diagrams. These influences go to failure rates and eventually affect component unavailability. Further study will make it possible that the influences of organizational factors on human error probabilities are incorporated into human reliability analysis and furthermore probabilistic safety assessment.


Author(s):  
Curtis Smith ◽  
David Schwieder ◽  
Trond Bjornard

As commonly practiced, the use of probabilistic risk assessment (PRA) in nuclear power plants only considers accident initiators such as natural hazards, equipment failures, and human error. Malevolent initiators are ignored in PRA, but are considered the domain of physical security, which uses vulnerability assessment based on an officially specified threat (design basis threat). This work explores the implications of augmenting and extending existing PRA models by considering new and modified scenarios resulting from malevolent initiators. Teaming the augmented PRA models with conventional vulnerability assessments can cost-effectively enhance security of a nuclear power plant. This methodology is useful for operating plants, as well as in the design of new plants. For the methodology, we have proposed an approach that builds on and extends the practice of PRA for nuclear power plants for security-related issues. Rather than only considering “random” failures, we demonstrated a framework that is able to represent and model malevolent initiating events and associated plant impacts.


2011 ◽  
Vol 121-126 ◽  
pp. 4156-4160
Author(s):  
Shou Yu Cheng ◽  
Xin Kai Liu ◽  
Min Jun Peng

The Human Machine Interface (HMI) of Monitor& Control System (MCS) is an important part of main control room in Nuclear Power Plants (NPP). The MCS is integrated by networks, digital computers and computer software. The paper discusses a new design mode for the HMI of MCS in Nuclear Power Plants. How to make the MCS information available is the focus of the HMI research. The paper discusses the design and research about the main layout and functions of the HMI. In order to verify and validate the design of HMI, the HMI of MCS is developed by a configuration software of JADE. HMI debugged with the NPP simulator. The Test illustrates that the HMI can help the operators to get very important information from MCS, which help the operators to operate the nuclear plant safely and reduces the human error.


Sign in / Sign up

Export Citation Format

Share Document