Quantitative risk analysis (QRA) is one of the most generally used safety analysis measures for risk management in process industries. Currently, earthquakes occur worldwide, resulting in significant damage. Despite the importance of considering the danger of earthquakes, however, seismic effects are often not included in risk analysis owing to difficulties in considering the multi-hazard nature and domino effects of earthquakes.

In this study, an improved methodology for QRA was proposed to consider the seismic effects including domino effects, and multi-hazard impacts of an earthquake by using a Bayesian network (BN). This analysis was applied to a topside CO₂ injection system for underground storage, which is susceptible to seismic effects. Because frequency analysis is based on a causal relationship, the BN can be used to simultaneously consider domino effects and multi-hazard risks. As a result, the societal risk integral, one of the factors in risk analysis, was $9.667\times {10}^{-4}/\mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r}$ in modified QRA; this value shows an increase of 3.9% compared with the societal risk integral in conventional QRA. Furthermore, the value can be increased to 35% in the sensitivity analysis depending on annual exceedance probability (AEP). This result shows the importance of considering seismic effects, including both the domino effect and multi-hazard impacts, in QRA. A risk reduction method was additionally applied to mitigate the process risk.

定量风险分析（QRA）是过程工业风险管理中最常用的安全分析措施之一。当前，地震在世界范围内发生，造成了严重的破坏。尽管考虑到地震危险性很重要，但是由于难以考虑地震的多灾种性质和多米诺骨牌效应，因此地震分析通常不包括在地震风险分析中。

在这项研究中，提出了一种改进的QRA方法，以利用贝叶斯网络（BN）考虑地震效应，包括多米诺骨牌效应和地震的多灾种影响。该分析被应用于易受地震影响的用于地下存储的顶侧CO ₂注入系统。由于频率分析基于因果关系，因此可以使用BN来同时考虑多米诺效应和多灾种风险。结果，作为风险分析因素之一的社会风险积分是$9.667\times {10}^{--4}/\mathrm{ÿ}\mathrm{Ë}\mathrm{一个}\mathrm{[R}$在修改后的QRA中；与传统QRA中的社会风险积分相比，该值增加了3.9％。此外，根据年度超出概率（AEP），在敏感性分析中，该值可以增加到35％。该结果表明在QRA中考虑地震影响（包括多米诺骨牌效应和多灾种影响）的重要性。此外，还采用了降低风险的方法来减轻过程风险。