Hierarchical Coordinated Bayesian Modeling of Complex Systems of Systems
Yacov Y. Haimes · 2018
The authors incorporate multiple decompositions from multiple perspectives supported and populated with the Bayesian data analysis. This modeling theory, philosophy and methodology, integrates all the direct and indirect relevant information from different levels of the hierarchies while placing more emphasis on relevant direct data. The authors coordinate the results from different decompositions and perform quantitative modeling of complex systems of systems (Complex SoS) supported with multiple databases. Then, they build on hierarchical coordinated Bayesian modeling (HCBM) and the partitioning multiobjective risk method (PMRM) for risk-based decision analysis of Complex SoS. The authors demonstrate embedding Bayes' theorem with Bellman's principle of optimality in dynamic programming for the purpose of resource allocation for intelligence gathering in countering terrorism. Finally, they discuss a two-track intelligence collection strategy: One team aims to maximize the posterior probabilities of occurrence and one team aims to minimize the posterior probabilities of occurrence.