Jean-Marie Flaus

Industrial emergency planning modeling: A first step toward a robustness analysis tool

The purpose of this paper is to present a model-based approach to the analysis of the robustness of industrial emergency plans, established by the European Union SEVESO II Directive. Robustness is defined in terms of the capacity of the mechanism to respond to deteriorated conditions. Analysis of emergency plans has been so far based mainly upon lessons learned from past major accidents or exercises, which do not allow for an integral analysis of the response mechanism. The proposed methodology is based upon a systemic, hierarchical and generic model of an internal or external industrial emergency plan, using the FIS modeling approach. The process generally found within an industrial emergency plan is identified through the model. Potential failures are estimated through an a priori analysis of the plan model and an a posteriori analysis of lessons learned from exercises and past accidents. Assessment of the plans functions is carried out via assessment checklists, structured via the systemic model for each of the plans process. This approach can hence be used as a toolbox both for the assessment of existing plans and the development of industrial emergency plans.

Model‐Driven and Risk‐Based Performance Analysis of Industrial Emergency Plans

This paper presents a methodology for the assessment of the performance of industrial hazardous materials emergency plans. This approach is based on a structuro-functional model of the emergency plan, which highlights the plans functions, resources, support services and interactions. A resource taxonomy is used to manage the complexity of the emergency response system. The model can be used both as a planning guide and for the analysis of the performance of the emergency response system, based on the risk assessment of the system. The failure probability is estimated through the plans functions and resources fault trees. The failure severity of each function is determined by using the facilitys hazard study. The failure criticality of each function is hence obtained.

Robustness analysis of industrial emergency plans: a model-based methodology

The purpose of this paper is to present a methodology for the analysis of the robustness of industrial internal and external emergency plans, established by the European Union SEVESO II Directive. This methodology is based upon a systemic, hierarchical and generic model of an internal or external industrial emergency plan. The process generally found within an internal and/or external industrial emergency plan is identified through the model, thus allowing for a modular representation of the plan. Each process integrates the functions/activities performed, the resources necessary for performing these functions/activities, the resources generated or affected by these functions, the supports required for each resource and the interactions (inputs and outputs) of each process. An explicit specification of these elements for a generic plan was provided through an ontological approach. An analysis of the plan model is performed to estimate a priori the failures that can potentially occur in the emergency response phase, when the plan is put into action. An extensive experience feedback analysis has also been performed to identify a posteriori the failures that have already occurred during the application of the plan. Some 160 accident reports have been consulted, and 31 internal and external emergency plan exercises were followed. This double analysis (a priori and a posteriori) has led to putting in place assessment checklists, structured via the systemic model for each of the plan’s process. Each checklist provides an indicator of the level of accomplishment (performance indicator) for the respective function. The logical combination of the function performance indicators results in a comprehensive assessment of the robustness of the industrial emergency plan. This methodology can hence be used as a toolbox, both for the assessment of existing plans and also for the iterative development of industrial emergency plans.