Risk Chain of Gas Leakage in Communities Based on Eventic Graph and Complex Network

Gas leakage accidents are becoming increasingly common due to aging pipeline segments and improper operation. When viewed from a community perspective, gas leakage has the potential to cause widespread harm by triggering chain reactions that affect other infrastructure systems and community functions. This could lead to serious consequences such as casualties and property damage. To explore the sequence of events triggered by gas leakage in communities and evaluate the associated risks, this paper introduces a risk chain analysis model using event-logic (eventic) graph and complex network theory. The model involves extracting events and their relationships from historical accident reports through text mining, creating a risk evolution graph using Gephi software, and developing a risk assessment model that considers the consequences of events, impacts of event relationships, and transition probabilities of events. By applying this model to analyze 493 accident reports related to gas leakage in the community, 296,057 risk chains were identified. Furthermore, we conducted a comparative analysis of the risks associated with different risk chains to identify the most dangerous ones and the most critical links within them. Our results reveal that gas leakage has a significant impact on fire protection and public security systems, with regional fires and mass panic being the most and second most dangerous evolutionary outcomes, respectively. The research results can provide suggestions and guidance for reducing the risk level after gas leaks from two aspects: targeted emergency response and evolution-based risk containment.