A comprehensive risk prediction method for defense mission planning based on probabilistic reasoning and hierarchical analysis

Abstract

Most existing risk prediction methods focus on constructing risk element sets and analyzing their uncertainties but do not deeply explore the correlation types and intensities of factors, resulting in large errors in the comprehensive risk prediction results. In this paper, a new integrated risk prediction method is proposed based on the correlation types of tasks in defense task planning and execution. The approach mainly includes the following steps: First, based on the difference of the sequence and mode of action of link tasks, three correlation types (hierarchical, synergistic, and independent) are defined among them, and various correlation measurement techniques are proposed to model these abstract correlation relations and provide data basis for constructing risk decision graphs. Secondly, the rotation extraction strategy is introduced to excavate the internal correlation law between link tasks and generate their hierarchical topology to ensure the rational distribution of their hierarchy positions in defense missions. Then, the intra-layer risk weight is determined based on the centrality of each node in the topology structure, and then the comprehensive risk prediction weighting graph is constructed. Finally, the path analysis is used to assess the rationality of the hierarchical topology structure of the link tasks, and the validity of the proposed method is verified using the test sample set. The results show that compared with other approaches, the predicted results of the proposed method more closely approximate the actual outcomes.

Keywords

Defense mission planning / risk prediction / associative relationships / hierarchical topology / decision mapping