A Method to Extract Causality for Safety Events in Chemical Accidents from Fault Trees and Accident Reports

Du, Junwei; Zhao, Hanrui; Yu, Yangyang; Hu, Qiang

doi:10.1155/2020/7132072

Cited by 4 publications

(1 citation statement)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A higher speed limit leads riders to fail to give way when turning [3]. A long waiting time at red lights [23], low density of motorized vehicles [24], and riding at the peak of morning [25] are in correlation with disobeying traffic signals. Among human factors except behavior, male riders and young riders are more likely to commit dangerous behaviors.…”

Section: Literature Reviewmentioning

confidence: 99%

Analyzing Safety Concerns of (e-) Bikes and Cycling Behaviors at Intersections in Urban Area

Wang

Chen

2022

Sustainability

View full text Add to dashboard Cite

Extensive effort has been devoted to examining the causal relationship between contributing factors and injury severities. Given the important role of riders’ behaviors in traffic conflicts, this paper aims to analyze the causal effects of traffic conflicts resulting from riders’ behaviors at intersections. The authors collected video data on 152 traffic conflicts caused by riders’ dangerous behaviors in Jiangning District, China. This paper proposes a Bayesian-structural equation modeling (BSEM) approach. Based on the obtained BSEM path coefficient diagram, the factor loadings and path coefficients are analyzed to unveil the potential influence of factors, including personal features, dangerous behavior tendency, temporal and spatial characteristics of dangerous behavior, and the external environment. The results show that compared to human factors, environmental factors have a less direct impact on the severity of traffic conflicts; instead, they have an indirect positive impact on traffic conflicts by affecting behaviors. That is, if riders judge that road conditions are not suitable to conduct dangerous behaviors, they become more cautious in view of current road conditions and time revenue. Furthermore, dangerous cycling behaviors that continue to encroach on the time and space of motorized vehicles are prone to be more dangerous. The dangerous behaviors that continuously encroach on the time and space of motor vehicles (e.g., disobeying traffic signals and riding in a motorway) are significant predictors of serious conflicts. Considering the heterogeneity of riding behavior, these findings could be applied to develop effective education and intervention programs for preventing riders’ high-risk behaviors and improving the traffic environment.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Analyzing Safety Concerns of (e-) Bikes and Cycling Behaviors at Intersections in Urban Area

Wang

Chen

2022

Sustainability

View full text Add to dashboard Cite

show abstract

An information support framework for chain reaction of major emergencies based on causality eventic graph

Zhang,

Li,

Sheng

2024

DTA

View full text Add to dashboard Cite

PurposeWith the rapid development of society, major emergencies occur frequently, posing a serious threat to people’s lives and property. This study focuses on the chain reaction of major emergencies, aiming to improve the overall situational awareness and capabilities for managing major emergencies in complex scenarios.Design/methodology/approachWe proposed an information support framework for the chain reaction of major emergencies based on a causality eventic graph (CEG). The framework consists of three modules: the data layer, the analysis layer and the service layer. The data layer focuses on the perception and collection of major emergency information. The analysis layer includes key components such as causality recognition, causality extraction, event fusion and generalization. In this layer, we developed several deep learning (DL)-based models using a joint extraction approach to obtain causal pairs. The service layer depicts the event evolution logic from both industry and public perspectives.FindingsThe empirical study has demonstrated the feasibility and effectiveness of the proposed information support framework. First, the BERT-BiLSTM-CRF model achieved the best performance in the causality extraction task. Second, the SBERT model was found to be more suitable for event fusion. Third, the analysis results of CEG revealed that the impact of pandemics on industries in turn affects other industries as well as people’s livelihoods and vice versa. The framework shows a better information support, discovery and reasoning effect. This study, however, still has several limitations. We focused primarily on causal logic and did not fully explore other logical relationships. In future work, we will incorporate more logical relations to further refine the eventic graph (EG).Originality/valueThis study proposes a novel information support framework for the chain reaction of major emergencies, leveraging CEG to provide targeted and hierarchical information to industry and public stakeholders. It solves several problems in the emergency management of major emergencies.

show abstract

Constructing the Sequential Event Graph for Event Prediction towards Cyber-Physical Systems

Liu

Cai

Han

et al. 2021

2021 IEEE 24th International Conference on Computer Supported Cooperative Work in Design (CSCWD)

View full text Add to dashboard Cite

A Method to Extract Causality for Safety Events in Chemical Accidents from Fault Trees and Accident Reports

Cited by 4 publications

References 33 publications

Analyzing Safety Concerns of (e-) Bikes and Cycling Behaviors at Intersections in Urban Area

Analyzing Safety Concerns of (e-) Bikes and Cycling Behaviors at Intersections in Urban Area

An information support framework for chain reaction of major emergencies based on causality eventic graph

Constructing the Sequential Event Graph for Event Prediction towards Cyber-Physical Systems

Contact Info

Product

Resources

About