Playing Bayesian Stackelberg game model for optimizing the vulnerability level of security incident system in petrochemical plants

Abstract

Constrained by the limited security resources of petrochemical plants, optimizing the vulnerability level of the security incident system plays a significant role in preventing and controlling security incidents. Previous research for security vulnerability has concentrated on the protection system, neglecting the influence of interconnections among vulnerability elements. The intelligence and struggle features of Security Incident System Vulnerability (SISV) elements are critical for optimizing the vulnerability level and characterizing the evolution of intelligence of security incident participants. By studying the game characteristics among SISV elements, this paper proposes the Bayesian Stackelberg SISV game model based on the security incident protection model established. In the game model, the strategies for protective vulnerability and aggressive vulnerability are defined as setting the security layer sub-units and the protective vulnerability levels, and setting the intrusion routes and the aggressive vulnerability levels, respectively. The model optimizes the security system vulnerability levels within the different security layer sub-units, based on the available security resources. Despite being demonstrated via an illustrative case, the defender's return on investment is significantly higher by using the SISV game model, the methodology can easily be tailored to a wide variety of petrochemical plants.