Threat assessment of air targets, using probabilistic Rule based model

Threat assessment is a fundamental task in command and control systems in military domain. This process must be fully real time and with high precision. Expert systems as decision  support systems, guide the commander to the best decision in minimum time. in this paper, by hybrid of  Bayesian networks and Belief rule based systems we propose a Probabilistic Rule based model to improve model construction and inference process. In proposed model, number of parameters in compare of two main method in threat assessment, Bayesian networks and rule based systems such as fuzzy, Is too much less and construction of model and inference process, Is much faster. This is an important feature in threat assessment and weapon allocation systems. In addition, weighting causes and interaction between them, Simple diagnosis and fast correction, are important advantages of proposed model.Simulation results for a 4 aerial target scenario, show that proposed model is better in compare with related method, in must validation criteria.