Study on operator’s SA reliability in digital NPPs. Part 3: A quantitative assessment method

Highlights

• The Bayesian network model of operators’ SA reliability analysis is built by data analysis and experts’ opinions. It considers the causal relationships of PSFs.

• A quantitative assessment method of SA reliability is established to calculate operator’s SA reliability on the basis of simulator experiment data. It provides more reliable data and improves the quality of HRA.

Abstract

Situational awareness (SA) is a key element impacting operators’ and decision-making and performance in nuclear power plants. SA reliability is an important part of human reliability analysis. In order to quantitatively assess SA reliability, the Bayesian network model of operators’ SA reliability analysis in digital main control rooms is built by data analysis. Furthermore, a quantitative assessment procedure of SA reliability is established, and the conditional probability distribution of influencing factors are more objectively determined on the basis of the full-size simulator experiment data of nuclear power plants and Bayesian theory. It is used to quantitatively calculate SA reliability, and a case is used to illustrate specific application of the established model. The results show that the model can provide more reliable data and standardized analysis procedures to support operators’ SA reliability evaluation in digital nuclear power plants.

Introduction

Situation awareness (SA), which is used within human factor research to explain to what extent operators of safety–critical and complex real systems know what is going on with the system and the environment. It is considered a prerequisite factor for effective decision making and performance (Lee et al., 2012). Due to the loss of Situation Awareness (LSA), failure to complete the complicated follow-up activities will possibly result in disastrous consequences. For example, the operator’s failure to keep the right understanding of the status of primary circuit in the nuclear incident in Three Mile Island (He and Huang, 2007). The pilots’ loss of the right understanding of the flying status in various air accidents (Woodhouse and Woodhouse, 1995). The research by Endsley points out that among various commercial air accidents resulted from human error, 88% of the accident reason is related to LSA (Endsley, 1994). The analysis report of incidents caused by air traffic control compiled by Jones and Endsley (1995) says that 69% incidents are somehow connected to the failure of information collection in air traffic controllers’ SA.

Faced with the continuous change of the system status in the operation of nuclear power plants (NPPs), operators in main control rooms need to deal with lots of information in a dynamic environment. The system’s current situation is accurately understood and the right decision is made to ensure the safety of NPPs. Therefore, it is of utmost importance to maintain sound SA in ensuring the safety of NPPs. With the upgrade of information technology and automatic level, the Human-Machine Interface (HMI) in NPPs has evolved from traditional analog system interface to digital system one. The digital Human-System Interface (HSI) has changed the context. For example, information’s presence (the limited presence of huge amount of information) (Zhang et al., 2010), procedure (computerized procedure) (Huang and Hwang, 2009), control (soft control) (Lee et al., 2011), task (interface management task) (O’Hara et al., 2002), team’s structure, communication and cooperation (O’Hara et al., 2000a) etc. Under the changed context, new problems caused by human factor might occur, especially the operators’ SA issue (O’Hara et al., 2009). For example, the information display of control panels in traditional main control rooms is direct and fixed in location, which is helpful for operators to understand the status of the whole power plant. However, the information’s locations on digital HSI are not fixed with information being discontinuous, more abstract, physically limited, with less displayed window, and more dynamically hidden. To obtain the status information of power plants, operators must complete it through complicated interface management task such as navigation, configuration screen and etc. The above-mentioned features will increase the cognitive load of operators, consume attention resources, produce Keyhole Effect (Seong, 2009) etc., which affect operators’ SA, and keep operators out of the loop (Kaber et al., 2006). Hence, operators’ SA issue is more outstanding in digital main control room compared to traditional main control room in NPPs. Especially in a high risk system, operators’ SA has become the hotspot for research. Endsley, 1995, Bendy and Meister, 1999, Adams et al. (1995) and others have all developed SA assessment model, which describes the internal process of SA’s occurrence of operators in MCR. These models basically describe the basic principle and general feature of operators’ dealing with the interaction of information and environment in order to acquire status perception, and contribute to the elaboration of cognitive mechanism of SA and main factors affecting SA. But it doesn’t consider the SA feature in digital control system, neither conduct quantitative analysis on the reliability of SA except qualitative analysis. Miao et al. (1997), Kim and Seong (2006) and Dai et al. (2012) have developed the assessment method on quantitative analysis of reliability of operators’ SA, and there are also quantitative calculation on SA in CREAM (Hollnagel, 1998) and HCR’s (Hannaman et al., 1984) methods for human reliability. However, almost all data are based on hypothesis. Operators’ influence by contextual environment has not be fully taken into consideration. The causal relationships of contextual factors has not be taken into consideration. It might cause the possibility of repeated calculation of its influence and lead to the wrong estimation of the failure probability of SA. Therefore, to reflect the contribution of SA reliability to human factor risk in a more objective manner, this article provides a procedure to quantitatively assess operator’s SA reliability on the basis of data-driven SA causality model established in Paper 2 (Li et al., under review) to serve HRA of NPPs.