An Intelligent Situation Matching and Course-of-Action Selection Method Based on the PREA–OODA Framework

In intelligent command and control, the selection of a course of action (COA) during the transition from the deduced situation to the immediate situation in the PREA loop still relies heavily on human judgment, resulting in limitations in both efficiency and decision consistency. To address this issue, this study first develops a multidimensional feature-vector model for the structured representation of situation characteristics associated with COAs, incorporating four key dimensions: enemy situation, friendly situation, battlefield environment, and mission requirements. A situation-matching algorithm based on feature similarity is then designed to enable rapid matching between the immediate situation and the situational background of candidate COAs. Specifically, Gaussian similarity, semantic similarity, and ordinal distance measures are employed for numerical, nominal, and ordinal features, respectively. On this basis, a parallel simulation-based framework for COA adaptation and effectiveness evaluation is established, in which dynamic optimization of COAs is supported through the transition among four states: physical, descriptive, predictive, and prescriptive. A maritime joint amphibious landing operation case study is conducted to demonstrate the effectiveness of the proposed method.