Modelling vicious networks with P-graph causality maps.

Area of Science:

• Systems analysis
• Causal network analysis
• Industrial safety

Background:

• P-graph causality maps offer a systematic method for analyzing complex causal chains.
• The original methodology focused on achieving desirable outcomes.
• Analysis of adverse causal networks is crucial for preventing industrial disasters and addressing sustainability issues.

Purpose of the Study:

• To extend the P-graph causality map methodology for analyzing vicious (adverse) causal networks.
• To identify keystone objects and mechanisms within adverse networks for deactivation.
• To provide a framework for understanding and mitigating industrial accidents and sustainability challenges.

Main Methods:

• Utilizing P-graph's bipartite representation (O-type nodes for objects, M-type nodes for mechanisms).
• Employing maximal structure generation to assemble complete causal networks.
• Applying solution structure generation to enumerate all feasible causal networks.

Main Results:

• The extended methodology successfully identifies critical components in adverse causal networks.
• Analysis provides insights into deactivating vicious networks by removing key elements.
• The approach was validated through an ex post analysis of the 1984 Bhopal disaster.

Conclusions:

• The extended P-graph causality map methodology is effective for analyzing and intervening in adverse causal networks.
• This approach can enhance industrial safety and inform sustainability strategies.
• Further applications in environmental and societal risk management are promising.