SPAR-H方法中行为形成因子间的相关性识别

刘建桥; 张力; 邹衍华; 孙倩琳; 刘雪阳; 陈帅

doi:10.13832/j.jnpe.2021.04.0144

SPAR-H方法中行为形成因子间的相关性识别

doi: 10.13832/j.jnpe.2021.04.0144

刘建桥^1,,
张力^{1, 2, ,},
邹衍华²,
孙倩琳¹,
刘雪阳²,
陈帅¹

1.
南华大学核科学技术学院，湖南衡阳，421001
2.
湖南工学院人因与安全工程研究院，湖南衡阳，421002

基金项目: 国家自然科学基金（71771084，71501068）；大型复杂人机系统人误预警技术湖南省工程实验室资助（湘发改高技[2015]1084号）；湖南省教育厅科研项目（18A424）；湖南省自然科学基金（2020JJ4016）；湖南省级应用特色学科安全与科学工程开放基金（KFB20017）；湖南省研究生科研创新项目（CX20200908）

详细信息

作者简介:
刘建桥（1992—），男，博士研究生，现从事人因可靠性方面的研究工作，E-mail: 1076218029@qq.com

通讯作者:
张　力，E-mail: 13807340602@139.com

中图分类号: TL364
计量
- 文章访问数: 528
- HTML全文浏览量: 190
- PDF下载量: 51
- 被引次数: 0
出版历程
- 收稿日期: 2020-06-18
- 修回日期: 2020-08-03
- 刊出日期: 2021-08-15

Identification of Correlation among Performance Shaping Factors of SPAR-H Method

1.
School of Nuclear Science and Technology, University of South China, Hengyang, Hunan, 421001, China
2.
Institute of Human Factors & Safety Engineering, Hunan Institute of Technology, Hengyang, Hunan, 421002, China

摘要

摘要: 标准化核电厂风险分析-人因可靠性分析方法（SPAR-H）是目前国际上认可和接受的人因可靠性分析方法，但其8个行为形成因子（PSFs）间存在交叉部分，导致人因失误概率重复计算或高估。为了改进SPAR-H的PSFs体系，通过统计2007年到2017年219份国内核电厂运行事件报告，筛选出与主控室操纵员运行有关的89份人因事件/事故报告进行PSFs相关性的研究，运用数据挖掘技术（关联规则分析、探索性因子分析、皮尔森相关性分析）对统计结果进行分析。结果表明：①复杂度、压力、职责适宜以及可用时间4个PSFs之间存在相关性。其中，复杂度分别与压力和职责适宜相关，职责适宜与压力、压力与可用时间相关；②工作过程、规程、人因工程/人机界面和经验/培训之间存在关联。在涉及经验/培训、人因工程/人机界面和规程的事件中，很大概率还涉及到工作过程。这些结论可以给改进SPAR-H的PSFs体系提供参考，为定量研究PSFs间的因果关系建立基础。
- 行为形成因子 /
- 相关性 /
- 标准化核电厂风险分析-人因可靠性分析方法（SPAR-H） /
- 数据挖掘
Abstract: Standardized Plant Analysis of Risk-Human reliability analysis (SPAR-H) is an internationally known and accepted human reliability analysis (HRA) method in nuclear power plants (NPPs). However, the eight performance shaping factors (PSFs) overlap, resulting in the double count or over estimation of the human error probabilities (HEPs). To improve its PSFs system, 89 human error event reports related to the operation of operators in the main control room were collected from 219 operating event reports of Chinese NPPs from 2007 to 2017. The correlation among the PSFs was then studied. Therein, three kinds of data mining methods, i.e., association rule analysis, exploratory factor analysis and Pearson correlation analysis, were used. Results show that: a. there is a significant correlation among complexity, stress/stressor, fitness for duty and available time, in which the c the complexity correlates with stress/stressor and fitness for duty, the fitness for duty correlates with stress/stressor, and the stress/stressor correlates with available time; (2) there also exists correlation among work process, procedure, ergonomics/HMI and experience/training. In the event involving the procedure, ergonomics/HMI or experience/training, work process is involved with a high probability. These findings can be used as the reference in the improvement of the PSFs system of SPAR-H and as the basis in the quantitative study of the cause-and–effect dependences among the PSFs.
- Performance shaping factors /
- Correlation /
- Standardized plant analysis of risk-human reliability analysis /
- Data mining

HTML全文

图 1 研究过程

Figure 1. General Process of Research Approach

下载: 全尺寸图片幻灯片

图 2 SPAR-H方法中PSFs出现在报告中的百分比

Figure 2. Percentage of Contributions to Events by PSFs of SPAR-H

下载: 全尺寸图片幻灯片

图 3 8个PSFs的关联图

Figure 3. Association Diagram of Eight PSFs

下载: 全尺寸图片幻灯片

表 1 关联规则建模结果

Table 1. Results of Association Rule Modeler

后项	前项	规则标识	支持度/ %	置信度/ %	提升度
工作过程	经验/培训	1	46.067	80.488	1.405
	人因工程/人机界面、经验/培训	2	8.989	87.5	1.527
	规程、经验/培训	3	11.236	70.0	1.222

下载: 导出CSV

表 2 8个PSFs探索性因子分析的结果

Table 2. Results of Exploratory Factor Analysis of Eight PSFs　　　　　

PSFs	因子1	因子2
复杂度	0.762	—
规程	—	−0.693
人因工程/人机界面	—	−0.714
经验/培训	—	0.526
可用时间	0.608	—
压力	0.839	—
工作过程	—	0.690
职责适宜	0.720	—
特征值	2.551	1.882

下载: 导出CSV

表 3 8个PSFs的皮尔森相关分析结果

Table 3. Results of Pearson Correlation Analysis of Eight PSFs

PSFs	复杂度	规程	人因工程/人机界面	经验/培训	可用时间	压力	工作过程	职责适宜
复杂度	1	—	—	—	—	—	—	—
规程	−0.204^①	1	—	—	—	—	—	—
人因工程/人机界面	−0.247^①	0.312^②	1	—	—	—	—	—
经验/培训	−0.167	−0.162	−0.176^①	1	—	—	—	—
可用时间	0.232^①	0.081	−0.136	−0.200^①	1	—	—	—
压力	0.681^②	−0.091	−0.168	−0.159	0.374^②	1	—	—
工作过程	−0.191^①	−0.321^②	−0.320^②	0.433^②	−0.251^②	−0.130	1	—
职责适宜	0.384^②	0.056	−0.095	−0.140	0.333^②	0.564^②	−0.176	1
注：①—在0.05级别（双尾），相关性显著；②—在0.01级别（双尾），相关性显著；"—"—无内容

下载: 导出CSV

参考文献(28)

[1]	PARK J, JUNG W, KIM J. Inter-relationships between performance shaping factors for human reliability analysis of nuclear power plants[J]. Nuclear Engineering and Technology, 2020, 52(1): 87-100. doi: 10.1016/j.net.2019.07.004
[2]	KANG S, SEONG P H. Performance shaping factor taxonomy for human reliability analysis on mitigating nuclear power plant accidents caused by extreme external hazards[J]. Annals of Nuclear Energy, 2020(145): 107533. doi: 10.1016/j.anucene.2020.107533
[3]	GROTH K M, MOSLEH A. A data-informed PIF hierarchy for model-based human reliability analysis[J]. Reliability Engineering & System Safety, 2012(108): 154-174.
[4]	FORESTER J, DANG V N, BYE A, et al. The international HRA empirical study: lessons learned from comparing HRA methods predictions to HAMMLAB simulator data: NUREG-2127[R]. Washington, DC: U.S. Nuclear Regulatory Commission, 2014.
[5]	LIU P, LYU X, QIU Y P, et al. Identifying key performance shaping factors in digital main control rooms of nuclear power plants: a risk-based approach[J]. Reliability Engineering & System Safety, 2017(167): 264-275.
[6]	FORESTER J, LIAO H F, DANG V N, et al. The U.S. HRA empirical study-assessment of HRA method predictions against operating crew performance on a U.S. nuclear power plant simulator: NUREG-2156[R]. Washington, DC: U.S. Nuclear Regulatory Commission, 2016.
[7]	GERTMAN D I, BLACKMAN H S, MARBLE J L, et al. The SPAR-H human reliability analysis method: NUREG/CR-6883[R]. Washington, DC: U.S. Nuclear Regulatory Commission, 2005.
[8]	GROTH K M, SWILER L P. Bridging the gap between HRA research and HRA practice: a Bayesian network version of SPAR-H[J]. Reliability Engineering & System Safety, 2013(115): 33-42.
[9]	何旭洪, 黄祥瑞. 工业系统中人的可靠性分析: 原理、方法与应用[M]. 北京: 清华大学出版社, 2007: 159-164.
[10]	张力,邹衍华,黄卫刚. 核电站运行事件人误因素交互作用分析[J]. 核动力工程,2010, 31(6): 41-46.
[11]	王国平, 郭伟宸, 汪若君. IBM SPSS Modeler数据与文本挖掘实战[M]. 北京: 清华大学出版社, 2014: 182-193.
[12]	DOELL C, HELD P, MOURA R, et al. Analysis of a major-accident dataset by Association Rule Mining to minimise unsafe interfaces[C]//Proceedings of the 13th International Probabilistic Workshop. Liverpool, UK: Research Publishing, 2015.
[13]	ZOU Y H, XIAO Z, ZHANG L, et al. A data mining framework within the Chinese NPPs operating experience feedback system for identifying intrinsic correlations among human factors[J]. Annals of Nuclear Energy, 2018(116): 163-170. doi: 10.1016/j.anucene.2018.02.038
[14]	薛薇. SPSS统计分析方法及应用[M]. 北京: 电子工业出版社, 2004: 327-338.
[15]	郭志刚. 社会统计分析方法: SPSS软件应用[M]. 北京: 中国人民大学出版社, 1999: 87-114.
[16]	吴明隆. 问卷统计分析实务[M]. 重庆: 重庆大学出版社, 2010: 476-477.
[17]	CRAMÉR H. Mathematical methods of statistics (PMS-9)[M]. Princeton: Princeton University Press, 1946: 500.
[18]	GUILFORD J P. Psychometric methods[J]. Journal of the American Statistical Association, 1956(51): 413-414. doi: 10.2307/2281384
[19]	CHANG Y H J, MOSLEH A. Cognitive modeling and dynamic probabilistic simulation of operating crew response to complex system accidents. Part 2: IDAC performance influencing factors model[J]. Reliability Engineering & System Safety, 2007, 92(8): 1014-1040.
[20]	GROTH K M. A data-informed model of performance shaping factors for use in human reliability analysis[D]. Washington Metropolitan: University of Maryland, College Park, 2009.
[21]	HOLLNAGEL E. Cognitive reliability and error analysis method (CREAM)[M]. Amsterdam: Elsevier, 1998: 107-117
[22]	MOHAGHEGH-AHMADABADI Z. On the theoretical foundations and principles of organizational safety risk analysis[D]. Washington Metropolitan: University of Maryland, College Park, 2007.
[23]	ANGELOPOULOU A, MYKONIATIS K, BOYAPATI N R. Industry 4.0: the use of simulation for human reliability assessment[J]. Procedia Manufacturing, 2020(42): 296-301. doi: 10.1016/j.promfg.2020.02.094
[24]	LAUMANN K, RASMUSSEN M. Suggested improvements to the definitions of Standardized Plant Analysis of Risk-Human Reliability Analysis (SPAR-H) performance shaping factors, their levels and multipliers and the nominal tasks[J]. Reliability Engineering & System Safety, 2016(145): 287-300.
[25]	WANG W, DI MAIO F, ZIO E. Considering the human operator cognitive process for the interpretation of diagnostic outcomes related to component failures and cyber security attacks[J]. Reliability Engineering & System Safety, 2020(202): 107007.
[26]	FORESTER J, KOLACZKOWSKI A, LOIS E, et al. Evaluation of human reliability analysis methods against good practices: NUREG-1842[R]. Washington, DC: U.S. Nuclear Regulatory Commission, 2006.
[27]	WHALEY A M, KELLY D L, BORING R L, et al. SPAR-H step-by-step guidance: INL/CON-12-24693[R]. Washington, DC: Idaho National Laboratory, 2012
[28]	张力,刘建桥,邹衍华,等. 核电厂数字化主控室操纵员界面管理任务特征的研究[J]. 核动力工程,2019, 40(4): 91-95.