Development of generic scenarios of industrial accidents triggered by floods: a first step toward decreasing the vulnerability of industrial facilities

Abstract

Interactions between natural events and industrial installations may lead to dangerous phenomena such as: fires, explosions, or toxic dispersions. The industrial sector is often unprepared for these Natech events, increasing its vulnerability. The final purpose of this study is to help industrial facilities to decrease their vulnerability to accidents triggered by floods. In order to attain this objective, the present article deals with the development of reference bowties that reconfigure the scenarios of these accidents. According to bibliographical research, Natech events are mainly caused by the releases of hazardous materials. This means that flood water can affect numerous parts of an industrial installation but only equipment in which hazardous substances can be found, are critical: storage units, process units and pipe networks. It is their damage that causes the LOC (Loss of Containment), responsible for generating dangerous phenomena. Studies on past accidents also showed that depending on the levels of flood severity, flood water can cause different categories of structural damage to critical equipment and thus affect the intensity of the LOC. Consequently, for each critical equipment identified, a list of structural damages (critical events) is associated depending on three levels of flood severity. Then, for each critical event, a bow-tie is elaborated. The development of nine generic bow-ties is the first step of a three step methodology of risk assessment aiming at the development of a checklist to be used in order to decrease the vulnerability of facilities to accidents triggered by floods.

Development of generic bow-tie diagrams of accidental scenarios triggered by flooding of industrial facilities (Natech)

“…This means that flood water can affect numerous parts of an industrial installation but only some equipment is critical. It is mainly the damage to hazardous materials containing equipment that causes the loss of containment (LOC) that in turn leads to dangerous phenomena (El Hajj et al, 2012). Therefore, damages to critical equipment leading to LOC are considered as the critical events.…”

“…These sub-systems were grouped according to the nature of hazards generated: SS1 Critical equipment (as mentioned earlier: atmospheric storage tanks, atmospheric reactors and pipes containing substances in liquid phase) (El Hajj et al, 2012), SS2 Building structures and non-critical equipment (other industrial equipment not classified as critical), SS3 Electrical equipment and SS4 Control, heating and cooling systems (Fig. 4).…”

Risk analysis in Natech events: State of the art

“…This framework suggests a complete analysis taking into account that it frames qualitative and quantitative aspects. After this work, (C. El Hajj et al, 2012) designed bow-tie diagrams to perform risk analysis in facilities that are vulnerable to flooding. These bow-tie diagrams were designed as the first step of a three-step methodology that consists of a checklist used to reduce the vulnerability of industrial facilities to lightning, earthquake, and flood events.…”

“…For instance, (Cruz & Okada, 2008) proposed a method for preliminary assessment of Natech risk in urban areas to mitigate 2018 emissions of hazardous materials from nearby industrial facilities, or facilities which store chemicals or equipment that contain hazardous materials in urban areas; however, subsequent to this work, more robust methodologies which include qualitative analyses have been developed. These include: framework for the assessment of Natech (Krausmann, Cozzani, Salzano, & Renni, 2011), methodology Bow-tie to perform risk analysis in flood-prone facilities (C. El Hajj, Piatyszek, & Laforest, 2012), fragility curves for pipelines that include land use planning (Lanzano, Salzano, de Magistris, & Fabbrocino, 2013), systematic methodology for analysis of flood risks (Carine El Hajj, Piatyszek, Tardy, & Laforest, 2015); semi-quantitative analyses such as: computational tool RAPID-N (Girgin & Krausmann, 2013), methodology to carry out risk assessments through the analysis of cascade effects (Kadri, Birregah, & Chatelet, 2014), geographic information system-based methodology (Soto & Renard, 2015). In terms of quantitative analyses, (Ernesto Salzano, Garcia Agreda, Di Carluccio, & Fabbrocino, 2009), developed comprehensive quantitative risk assessment (QRA) methodologies, inserting domino effect risk and the risk associated with nature in the QRA.…”