Resilience-based optimal fire fighting in process plants

In the recent years, fire domino effects have been involved in some extremely severe accidents in the chemical and process industries. In the need of a better understanding of the prevention of fire escalation scenarios, the present study focuses on emergency firefighting which compared to passive and active fire protection measures has received less attention. In the present study, a resilience-based firefighting methodology has been introduced to increase the absorptive capacity of process plants in withstanding the expanding disruptions of fire escalation scenarios. The proposed resilience metric allows to measure the resilience and performance deficiencies and facilitates the identification of optimal intervention actions that lead to the lowest loss of resilience. The modeling of fire escalation scenarios has been conducted using a Bayesian network approach while the modeling and investigating the effect of different firefighting strategies has been performed by extending the Bayesian network to an influence diagram. The area above the resilience curve, which is equal to the loss of resiliency, has been considered as a metric and discriminating factor to identify the optimal firefighting strategy. The results of this innovative investigation allow to better understand the effectiveness and reliability of the emergency protection interventions in a chemical plant and to study how the domino effect risk assessment can be enriched by new concepts optimizing the distribution of human and material resources in fire fighting operations.