Game Theory for Managing Security in Chemical Industrial Areas

Large inventories of hazardous chemicals which can cause catastrophic consequences if released maliciously, the presence of chemical agents which can be stolen and be used either in later terrorist attacks or in making chemical and biochemical weapons, along with the key role of chemical plants in the economy and the public welfare and as an integral element in the supply chain have made the security of chemical plants a great concern especially since 9/11 terrorist attacks in the US. Aside from the importance of chemical plants themselves as potentially attractive targets to terrorist attacks, the usage of chemicals in more than half of the terrorist attacks worldwide further emphasizes the security assessment and management of chemical plants.

Game theory is a mathematical tool for supporting decision making in a multiple players situation where one player’s utility will be determined not only by his own decision, but also by other players’ decisions. An illustrative example of this situation is the Rock/Scissors/Paper game (“RSP” game). In an RSP game, whether a player wins or loses depends on both what he plays and what his opponent plays. This is a well-known game between mostly children with very simple rules. Two ‘players’ hold their right hands out simultaneously at an agree signal to represent a rock (closed fist), a piece of paper (open palm), or a pair of scissors (first and second fingers held apart). If the two symbols are the same, it’s a draw. Otherwise rock blunts scissors, paper wraps rock, and scissors cut paper, so the respective winners for these three outcomes are rock, paper and scissors. The RSP game is what is called a ‘two-player zero-sum non-cooperative’ game. There are obviously many other types of game and the field of game theory is very powerful to provide (mathematical) insights into strategic decision-making.

In this chapter, we introduce a game theoretic model for protecting a chemical plant from intelligent attackers. The model is named Chemical Plant Protection Game, abbreviated as “CPP Game” [1]. The CPP Game is developed based on the general intrusion detection approach in chemical plants. To this end, the general intrusion detection approach is firstly introduced. We develop and explain the CPP Game by modelling its players, strategies, and payoffs. Afterwards in Sect. 3.3, different equilibrium concepts are used to predict the outcome of the CPP Game [2]. An analysis of the inputs and outputs of the game is provided in Sect. 3.4, from an industrial practice point of view [3]. Finally, conclusions are drawn at the end of this chapter.

In this chapter, the Chemical Plant Protection game is extended to deal with input parameters with distribution-free uncertainties [1] The so-called interval CPP game is defined. Two algorithms, namely, the interval bi-matrix game solver (IBGS) and the interval CPP game solver (ICGS), are proposed.

In this chapter, we model attackers with bounded rationality in the Chemical Plant Protection game. Three different behaviour models of attackers are investigated, namely, the epsilon-optimal attacker, the monotonic-optimal attacker, and the MiniMax attacker. All these attacker models are integrated to the Stackelberg CPP game, which means that the defender moves first, and the attackers follow. Furthermore, the monotonic-optimal attacker is investigated in the Interval CPP game with only one type of attacker, and a game solution named Monotoic MaxiMin Solution for the Interval CPP game (MoSICP) is defined [1]. The MoSICP solution incorporates both bounded rational attackers and distribution-free uncertainties into the CPP game. The epsilon-optimal attacker model, being related to the defender’s distribution-free uncertainties, and the MiniMax attacker model, being the most conservative model, are therefore investigated in the Bayesian Stackelberg CPP game framework, instead of in the Interval CPP game framework. The defender is still assumed to behave rationally to maximize her payoff.

Due to economies of scale and all kinds of collaboration benefits, chemical plants are usually geographically clustered, forming chemical industrial parks or so-called ‘chemical clusters’. Some examples of such clusters are the Antwerp port chemical cluster in Belgium, the Rotterdam port chemical cluster in the Netherlands, the Houston chemical cluster in the US, or the Tianjin chemical cluster in China. Besides fixed security countermeasures within every plant, the patrolling of security guards is also scheduled, for securing these chemical facilities at different points and times, e.g. at night. The patrolling can either be single-plant oriented, which can be completely scheduled by the plant itself, or it can be multiple-plants oriented, which should be scheduled by an institute at a higher level than the single-plant level, for instance a multiple plant council (MPC) [1] Both types of patrolling have a drawback of not being able to deal with intelligent attackers. Some patrollers follow a fixed patrolling route, and in this case the adversary is able to predict the patroller’s position at a certain time. Other patrollers purely randomize their patrolling, without taking into consideration the hazardousness level that each installation/facility/plant holds, and if this is the case, the adversary may focus to attack the most dangerous installations/facilities/plants since all installations/facilities/plants are equally patrolled.

Two case studies are carried out in this chapter, for illustrating the single plant protection game (the CPP game) and for demonstrating the multi-plant patrolling game (the CCP game), respectively.

Chemicals-using industries have an important role in modern society for providing the basic ingredients (fuels, chemicals, intermediates and consumer products) for our modern day lives and luxury. However, they also pose huge threats to society due to the mere use and storage of large amounts of hazardous materials with sometimes extreme processing conditions. The prevention of unintentionally caused events, which is the field of occupational safety and process safety, has been significantly improved in the process industries. Conversely, the physical protection of chemical plants and areas from malicious attacks, being the field of physical security, has not received enough attention yet by both academic researchers and industrial practitioners.