Abstract: Operating condition risk assessment method (LEC) is a semi-quantitative evaluation method for risk assessment of safety hazards in a potentially hazardous environment, and has become one of the most commonly used safety risk assessment methods due to its simple calculation principle and easy operation. The LEC method uses the risk value (D) to measure the degree of danger of safety risks, and its assessment criteria mainly include the possibility of hazardous events (L), the frequency of exposure to hazardous environments (E) and the possible consequences of hazardous events (C). For past years, although the LEC method has been widely used in the fields of building construction, mining and water conservancy construction, there are still certain limitations. In order to further improve the accuracy of safety risk assessment, this paper intends to improve the traditional LEC method from the following four aspects: (1)The shortage of evaluation information omission due to the inability of precise values to accurately portray the hesitancy of decision maker. (2)The deficiency of low accuracy of safety hazard ranking results caused by the large and discontinuous span of evaluation values. (3)The traditional LEC method applies equal distribution of indicator weights, ignoring the priority order of indicators. (4)There is no consideration of the psychological characteristics of limited rationality of decision makers in the security risk assessment process, such as loss avoidance, reference dependence, and other psychological behavioral factors.
To effectively remedy the inherent defects in the above traditional LEC methods, this paper proposes a probabilistic linguistic LEC risk assessment method based on prospect theory and the linguistic ordered weighted average (LOWA) operator. Firstly, the method uses probabilistic linguistic term sets (PLTS) instead of exact discontinuous values to portray the uncertainty of makers’ evaluation evaluation information. Secondly, we use probabilistic linguistic multiplicative analytic hierarchy process (PL-MAHP) and two-way projection minimum deviation model of positive and negative ideal points to determine the subjective and objective weights of indicators, and the subjective and objective weights are integrated and optimized based on the optimization idea of game theory. Then, considering the psychological behaviors of loss avoidance and reference dependence of decision maker in the process of risk assessment, the prospect theory is introduced into the traditional LEC method, and the LOWA operator is used to determine the prospect reference points to reduce the subjectivity of reference point selection, thus realizing the risk ranking of safety hazards. Finally, in order to test the rationality and effectiveness of this method, we apply it to the risk assessment calculations of safety hazards of hazardous chemical road transportation. On this basis, we verify the feasibility, effectiveness and superiority of the method through the sensitivity analysis of subjective parameters and the comparison analysis with the existing decision-making methods and reference point selection methods.
This paper summarizes and analyzes the shortcomings of the traditional LEC method based on three perspectives: The description forms of evaluation information, the weight model of indicators and the decision-making method of limited rationality. In order to adapt to the language expression habits of experts and effectively avoid the omission of evaluation information, the LEC method is extended to the probabilistic linguistic environment to characterize the hesitancy and uncertainty of decision maker. Considering the influence of positive and negative ideal points on decision points, this paper constructs a minimum deviation weight model based on the two-way projection closeness of positive and negative ideal points. In view of the finite rationality characteristics of decision maker, the prospect theory and the LOWA operator are introduced into the LEC method to improve the objectivity of reference point selection, and finally achieve good risk assessment results.

Key words: LEC method, probabilistic linguistic term set, game theory, prospect theory, LOWA operator, transportation safety of dangerous chemicals

摘要： 考虑到作业条件危险性评价法(LEC)存在的评价信息遗漏、风险量化分析能力弱、忽略决策者有限理性特征等缺陷,提出一种基于前景理论和语言有序加权平均(LOWA)算子的概率语言LEC风险评估方法。首先,该方法采用概率语言术语集(PLTS)代替精确的非连续性数值来刻画决策者评价信息的不确定性;其次,利用概率语言连乘层次分析法(PL-MAHP)和正、负理想点双向投影贴近度最小偏差模型确定指标的主、客观权重,并基于博弈论的思想对主、客观权重进行组合优化;然后,考虑到风险评估过程中决策者的损失规避和参照依赖等心理行为,在前景理论决策视角下,引入LOWA算子确定前景参照点以提升参照点选取的科学性,进而实现安全隐患的风险排序;最后,通过危化品运输安全隐患的风险评估算例,验证了该方法的合理性和有效性。

关键词: LEC方法, 概率语言术语集, 博弈论, 前景理论, LOWA算子, 危化品运输安全