Quantitative assessment of domino and NaTech scenarios in complex industrial areas

Highlights

• Quantitative Area Risk Analysis (QARA) is a powerful decision support tool.

• QARA may be extended to assess NaTech and first and second-level domino scenarios.

• NaTech scenarios are likely to give a relevant contribution to industrial risk assessed in QARA.

• Second-level domino scenarios give a non-negligible contribution to industrial risk.

• QARA extension to domino and NaTech is sufficiently is ready to become a standard tool.

Abstract

Since the late 80s the application of quantitative risk assessment to the issue of land-use planning with respect to major accident hazards emerged as a topic to be addressed within the safety assessment of chemical and process plants. However, in the case of industrial clusters or complex industrial areas specific methodologies are needed to deal with high-impact low-probability (HILP) events. In the present study, innovative methodologies developed for the quantitative assessment of risk due to domino and NaTech scenarios are presented. In recent years a set of models for the calculation of equipment damage probability were developed. A specific effort was dedicated to the improvement of models for the calculation of equipment damage probability in these accident scenarios. In the present study, the application of these models to case-studies was analyzed. The results of the improved models obtained for NaTech quantitative assessment were compared to previous results in the literature. A specific innovative approach was developed to multi-level quantitative assessment of domino scenarios, and its potential was analyzed. The results were examined also evidencing the role and the progress with respect to the pioneering work started on these topics by Franco Foraboschi.

Introduction

Quantitative risk assessment (QRA) is among the tools more widely applied worldwide to provide quantitative information on the risk caused by conventional accidents in chemical and process plants (Center for Chemical Process Safety [CCPS], 2000; Mannan, 2005). QRA consists of a set of methodologies for estimating the risk posed by a given system in terms of human loss or, in some cases, economic loss (CCPS, 2000; Mannan, 2005; Uijt de Haag & Ale, 1999). The application of QRA to extended and complex industrial areas or to chemical clusters required the development of specific methodologies to account for issues as the distribution of population and the transportation of hazardous substances to and from the area of concern. These extensions of the QRA methodology were named “Quantitative Area Risk Assessment” or QARA in the literature (Antonioni, Spadoni, & Cozzani, 2009; Cozzani, Antonioni, & Spadoni, 2006; Cozzani, Bonvicini, Spadoni, & Zanelli, 2007; Cozzani, Gubinelli, Antonioni, Spadoni, & Zanelli, 2005; Egidi, Foraboschi, Spadoni, & Amendola, 1995; Spadoni, Contini, & Uguccioni, 2003; Spadoni, Egidi, & Contini, 2000).

Two pilot studies may be recognized as the precursors of QARA: the integrated studies of Canvey Island (Health and Safety Executive [HSE], 1978, 1981) and of the Rijnmond area (Rijnmond Public Authority, 1982). In 1988 a project aiming at the quantitative analysis of risk in the Ravenna industrial area was started, named “Analisi dei Rischi Industriali e Portuali nell'Area di Ravenna” or ARIPAR project (Egidi et al., 1995). Within the ARIPAR project, a software allowing the quantitative calculation of individual and societal risk was developed. The Aripar software, developed by University of Bologna, EC Joint Research Centre and Snamprogetti was at the time the first dedicated software tool allowing the calculation of iso-risk and F–N plots for industrial areas, including hazardous materials transportation systems (Bellezza, Contini, Binda, & Spadoni, 1998; Spadoni et al., 2003).

Since then, the methodologies, the models and the software tools available for QRA and QARA where continuously improved (http://www.dnv.com/binaries/Safeti_1012_tcm4-76524.pdf; http://www.tno.nl/RISKCURVES; Godoy, Santa Cruz, & Scenna, 2007; Pasman & Reniers, 2013). Also in the framework of the application of the Seveso Directives (Directives 82/501/EEC, 96/82/EC, and 2012/18/EU), these techniques became a reference for decision making of operating companies and of public authorities. In particular the second Seveso Directive (96/82/EC), introducing specific requirements for land-use planning with respect to major accident hazard called for decision support methods often based on QARA in complex industrial areas: e.g. a number of QARA assessments were carried out in Italy to control and reduce the risk of extended industrial areas as those of Livorno and Piombino (Spagli, Ceccanti, & Mossa Verre, 2006; Uguccioni, Bellocchi, & Spadoni, 2006), Porto Marghera (Luria & Aspinall, 2003), Messina (Milazzo, Lisi, Maschio, Antonioni, & Spadoni, 2010; Morra, Lisi, Spadoni, & Maschio, 2009), Priolo-Melilli (Bartolozzi, Bajardi, Vasile, & Marino, 2010).

The growing public concern caused by high-impact low-probability (HILP) accident scenarios, and the increasing attention devoted in the scientific and technical literature to domino effect (Reniers & Cozzani, 2013) and NaTech events (Krausmann, Cozzani, Salzano, & Renni, 2011; Krausmann & Cruz, 2008; Petrova & Krausmann, 2011), called for the extension of the conventional QARA methodology. Domino effect was responsible of several catastrophic accidents that took place in the chemical and process industry (Abdolhamidzadeh, Abbasi, Rashtchian, & Abbasi, 2011; CCPS, 2000; Khan & Abbasi, 1999; Mannan, 2005). Severe NaTech events were documented as a consequence of earthquakes, floods, hurricanes and lightning (Cozzani, Campedel, Renni, & Krausmann, 2010; Krausmann, Cozzani, et al., 2011; Krausmann, Renni, Campedel, & Cozzani, 2011; Renni, Krausmann, & Cozzani, 2010; Young, Balluz, & Malilay, 2004).

In recent years, a common framework was developed to include the assessment of external hazard factors in risk assessment. A comprehensive approach was obtained to allow the quantitative assessment of domino and NaTech, extending the QARA potentialities to the quantitative assessment of the contribution to industrial risk due to such scenarios.

In the present study the overall procedures for the quantitative assessment of domino and NaTech scenarios are revised. Recent developments, as the possibility to include higher order domino scenarios are presented. Progress on models for equipment damage assessment is discussed. Several case-studies are analyzed, to show the importance of these developments for the calculation of overall risk figures. The results are also analyzed in the perspective of the pioneering work in the field carried out by Franco Paolo Foraboschi, to which this contribution in the “process safety pioneers” special issue is dedicated.