Modelling vicious networks with P-graph causality maps

P-graph causality maps were recently proposed as a methodology for systematic analysis of intertwined causal chains forming network-like structures. This approach uses the bipartite representation of P-graph to distinguish system components (objects represented by O-type nodes) from the functions they perform (mechanisms represented by M-type nodes). The P-graph causality map methodology was originally applied for determining structurally feasible causal networks to enable a desirable outcome to be achieved. In this work, the P-graph causality map methodology is extended to the analysis of vicious networks (i.e., causal networks with adverse outcomes). The maximal structure generation algorithm is first used to assemble the problem elements into a complete causal network; the solution structure generation algorithm is then used to enumerate all structurally feasible causal networks. Such comprehensive analysis gives insights on how to deactivate vicious networks through the removal of keystone objects and mechanisms. The extended methodology is illustrated with an ex post analysis of the 1984 Bhopal industrial disaster. Prospects for other applications to sustainability issues are also discussed.

Automated summary

The P-graph causality map methodology distinguishes system components and functions to determine structurally feasible causal networks for achieving desired outcomes. This approach has been extended to analyzing vicious networks and provides insights on deactivating them by removing key objects and mechanisms. The methodology is illustrated with an ex post analysis of the 1984 Bhopal industrial disaster, with potential applications to sustainability issues.