Genserik Reniers

An external domino effects investment approach to improve cross-plant safety within chemical clusters.

Every company situated within a chemical cluster faces the risk of being struck by an escalating accident at one of its neighbouring plants (the so-called external domino effect risks). These cross-plant risks can be reduced or eliminated if neighbouring companies are willing to invest in systems and measures to prevent them. However, since reducing such multi-plant risks does not lead to direct economic benefits, enterprises tend to be reluctant to invest more than needed for meeting minimal legal requirements and they tend to invest without collaborating. The suggested approach in this article indicates what information is required to evaluate the available investment options in external domino effects prevention. To this end, game theory is used as a promising scientific technique to investigate the decision-making process on investments in prevention measures simultaneously involving several plants. The game between two neighbouring chemical plants and their strategic investment behaviour regarding the prevention of external domino effects is described and an illustrative example is provided. Recommendations are formulated to advance cross-plant prevention investments in a two-company cluster.

Engineering risk management

Keywords: Risk ; Management Reference EPFL-BOOK-187445 Record created on 2013-07-09, modified on 2017-11-13

Using graph theory to analyze the vulnerability of process plants in the context of cascading effects

Dealing with large quantities of flammable and explosive materials, usually at high-pressure high-temperature conditions, makes process plants very vulnerable to cascading effects compared with other infrastructures. The combination of the extremely low frequency of cascading effects and the high complexity and interdependencies of process plants makes risk assessment and vulnerability analysis of process plants very challenging in the context of such events. In the present study, cascading effects were represented as a directed graph; accordingly, the efficacy of a set of graph metrics and measurements was examined in both unit and plant-wide vulnerability analysis of process plants. We demonstrated that vertex-level closeness and betweenness can be used in the unit vulnerability analysis of process plants for the identification of critical units within a process plant. Furthermore, the graph-level closeness metric can be used in the plant-wide vulnerability analysis for the identification of the most vulnerable plant layout with respect to the escalation of cascading effects. Furthermore, the results from the application of the graph metrics have been verified using a Bayesian network methodology.

Domino effects within a chemical cluster: A game-theoretical modeling approach by using Nash-equilibrium

Every company situated within a chemical cluster faces domino effect risks, whose magnitude depends on every companys own risk management strategies and on those of all others. Preventing domino effects is therefore very important to avoid catastrophes in the chemical process industry. Given that chemical companies are interlinked by domino effect accident links, there is some likelihood that even if certain companies fully invest in domino effects prevention measures, they can nonetheless experience an external domino effect caused by an accident which occurred in another chemical enterprise of the cluster. In this article a game-theoretic approach to interpret and model behaviour of chemical plants within chemical clusters while negotiating and deciding on domino effects prevention investments is employed.

MISTRAL : a game-theoretical model to allocate security measures in a multi-modal chemical transportation network with adaptive adversaries

In this paper we present a multi-modal security-transportation model to allocate security resources within a chemical supply chain which is characterized by the use of different transport modes, each having their own security features. We consider security-related risks so as to take measures against terrorist acts which could target critical transportation systems. The idea of addressing security-related issues, by supporting decisions for preventing or mitigating intentional acts on transportation infrastructure, has gained attention in academic research only recently. The decision model presented in this paper is based on game theory and it can be employed to organize intelligence capabilities aimed at securing chemical supply chains. It enables detection and warning against impending attacks on transportation infrastructures and the subsequent adoption of security countermeasures. This is of extreme importance for preventing terrorist attacks and for avoiding (possibly huge) human and economic losses. In our work we also provide data sources and numerical simulations by applying the proposed model to a illustrative multi-modal chemical supply chain.

A game-theoretical approach for reciprocal security-related prevention investment decisions

Every company situated within a chemical cluster faces important security risks from neighbouring companies. Investing in reciprocal security preventive measures is therefore necessary to avoid major accidents. These investments do not, however, provide a direct return on investment for the investor-company and thus plants are hesitative to invest. Moreover, there is likelihood that even if a company has fully invested in reciprocal security prevention, its neighbour has not, and as a result the company can experience a major accident caused by an initial (minor or major) accident that occurred in an adjacent chemical enterprise. In this article we employ a game-theoretic approach to interpret and model behaviour of two neighbouring chemical plants while negotiating and deciding on reciprocal security prevention investments.

Vulnerability of industrial facilities to attacks with improvised explosive devices aimed at triggering domino scenarios

Process- and chemical plants may constitute a critical target for a terrorist attack. In the present study, the analysis of industrial accidents induced by intentional acts of interference is carried out focusing on accident chains triggered by attacks with home-made (improvised) explosives. The effects of blast waves caused by improvised explosive devices are compared with those expected from a net equivalent charge of TNT by using a specific methodology for the assessment of stand-off distances. It is demonstrated that a home-made explosive device has a TNT efficiency comprised between 0.2 and 0.5. The model was applied to a case study, demonstrating the potentiality of improvised explosives in causing accident escalation sequences and severe effects on population and assets. The analysis of the case-study also allowed obtaining suggestions for an adequate security management.

Risk-based design of process plants with regard to domino effects and land use planning.

Land use planning (LUP) as an effective and crucial safety measure has widely been employed by safety experts and decision makers to mitigate off-site risks posed by major accidents. Accordingly, the concept of LUP in chemical plants has traditionally been considered from two perspectives: (i) land developments around existing chemical plants considering potential off-site risks posed by major accidents and (ii) development of existing chemical plants considering nearby land developments and the level of additional off-site risks the land developments would be exposed to. However, the attempts made to design chemical plants with regard to LUP requirements have been few, most of which have neglected the role of domino effects in risk analysis of major accidents. To overcome the limitations of previous work, first, we developed a Bayesian network methodology to calculate both on-site and off-site risks of major accidents while taking domino effects into account. Second, we combined the results of risk analysis with Analytic Hierarchical Process to design an optimal layout for which the levels of on-site and off-site risks would be minimum.

Resilience of chemical industrial areas through attenuation-based security

This paper investigates the possibility of attenuation-based security within chemical industrial areas. Representing chemical industrial areas as mathematical networks, we prove by case-study that the resilience to disaster of such areas may follow a power-law distribution. Furthermore, we examine what happens to the network when highly hazardous installations would be intelligently protected against malicious acts: the network disintegrates into separate smaller networks. Hence, islands are formed with no escalation danger in between. We conclude that it is possible to protect chemical industrial areas in such a way that they are more resilient against terrorism.

Knock-on accident prevention in a chemical cluster

Empirical research has revealed that safety managers acknowledge the importance of cross-company cooperation for knock-on risk reduction. This paper presents a decision support software tool, called DomPrevPlanning(C), to prevent knock-on or domino accidents. The computer automated toolkit only requires simple input data to identify the weak spots in an industrial area. This way, a ranking list is generated, indicating where to take safety and security measures in a complex chemical surrounding consisting of hundreds of chemical installations. The software is tested on a real-case sub-cluster from the Antwerp chemical cluster, the second largest chemical cluster in the world.