Applying game theory for adversarial risk analysis in chemical plants

Abstract

Since the 9/11 attack in New York in 2001, a lot of attention has been paid to the protection of critical infrastructures. Chemical industries are without doubt critical infrastructures due to their extreme importance for society in combination with their vulnerability. They play important roles in modern-life society, from producing and providing daily necessities such as food and energy, to making modern medicine. They are thus truly essential to our modern way of living. Process plants usually store dangerous goods in large quantities, which may pose an important threat to themselves as well as to their surroundings. Moreover, due to a variety of benefits of scale, process plants tend to build their factories geographically together, potentially aggravating the danger. Therefore, the importance of protecting industrial process plants (including those in the chemical industry, the food industry, the energy industry, and others) cannot be overestimated.
Risks caused by human behaviours with the intention to cause losses are defined as security risks. For instance, thieves intentionally intruding a plant for stealing valuable materials, or terrorists maliciously setting a fire on a chemical facility to cause societal fear. Initiators of security events (henceforth, attackers) would intelligently observe the defender’s defence plan and then schedule their attack accordingly. Literature has actually shown how resources can be misallocated if intelligent interactions between the defender and the attacker are not considered.
Game theory was developed in the economic domain for modelling both cooperative and competitive behaviours in a multiple actors system. In the last 100 years, game theory has been theoretically improved and practically applied to various domains, such as evolutionary biology, computer science etc. These researches have demonstrated the capability of game theory in modelling intelligent interactions. Several security management systems based on game theory have been developed and deployed in reality, such as the ARMOR system for the Los Angeles airport, the PROTECT system for the US coast guard, etc.
In this research, game theory is employed to study the protection of chemical industrial areas. Four models are proposed: i) DAMS – an agent-based modelling and simulation approach for assessing domino effects in chemical plants; ii) CPP game – a game theoretic model for single plant protection; iii) CCP game – a game theoretic model for multiple plants protection, by optimizing patrolling; and iv) PPG – a game theoretic model aiming at optimizing pipeline patrolling within or between chemical plants. These models are briefly explained hereafter.