The effectiveness evaluation for security system based on risk entropy model and Bayesian network theory

With the rapid development of industrialization and urbanization, security situation becomes more and more serious in China. Many cities have established security systems in order to maintain social stability. Although various security facilities have played a positive role in the field of social security prevention, there are some defects with poorly designed and personnel protection, physical protection, and electronic protection formation non-integration. The effectiveness evaluation of security system focuses on the experience and management level, and lack of scientific quantitative evaluation of the theoretical methods and technical means. So, we need to assess their effectiveness in a scientific and objective manner. In this paper, we first put forward the concept of risk entropy according to the Shannon information theory and establish the risk entropy model of security system. We then apply Bayesian network and Fuzzy Comprehensive Evaluation methods to determine the parameters in the model analysis. We not only solve the problem of interference on subject factors in traditional qualitative method for security system, but also improve the assessment objectively. In the next stage, we conduct the simulation experiment according to the specific examples of security system. The case study shows that the assessment results are consistent with the reality and this method can be reasonably used for the effectiveness evaluation for security system. We believe this model can break through bottlenecks in the technology of objective effectiveness evaluation and bear important scientific significance in the applications of scientific prediction and effective prevention for public safety emergency response system.