Microsoft Word - Manuscript_The Assessment of Risk Caused by Fire and explosion_ A Domino Effect-Based Study The Assessment of Risk Caused by Fire and Explosion in Chemical Process Industry: A Domino Effect-Based Study Farid Kadri Univ. Lille Nord de France, F-59000 Lille, France UVHC, TEMPO Lab., "Production, Service, Information" Team, F-59313 Valenciennes, France Farid.kadri@univ-valenciennes.fr Eric Chatelet, Patrick Lallement University of Technology of Troyes (UTT), France Laboratory of Systems Modeling and Dependability UMR STMR/ICD CNRS N°6279 {eric.chatelet, partrick.lallement}@utt.fr Abstract In the field of risks analysis, the domino effect has been documented in technical literature since 1947. The accidents caused by the domino effect are the most destructive accidents related to industrial plants. Fire and explosion are among the most frequent primary accidents for a domino effect due to the units under pressure and the storage of flammable and dangerous substances. Heat radiation and overpressure are one of major factors leading to domino effect on industrial sites and storage areas. In this paper we present a method for risk assessment of domino effects caused by heat radiation and overpressure on industrial sites. This methodology is based on the probabilistic models and the physical equations. It allows quantifying the effect of the escalation vectors (physical effects) in industrial plants, the three areas defined in this study may be useful in the choice of safe distances between industrial equipments. The results have proven the importance of domino effect assessment in the framework of risk analysis. Keywords: Domino effect; Quantitative risk assessment; Explosions; Fires; Storage areas. 1. Introduction The accidents caused by the domino effect are those that cause the most catastrophic consequences. The consequences of these latter are at various levels and may affect not only the industrial plants, but also people, environment and economy. The probability of domino effect is increasingly high due to development in industrial plants, the proximity of such establishments and their inventories of dangerous substances. The potential risk of domino effect is widely recognized in the legislation since the first "Seveso-I" Directive (82/501/EEC), which required the assessment of domino effects in the safety analysis of industrial sites whose activities are subject to this directive. Furthermore, the "Seveso-II" (96/82/EC) extended these requirements to the assessment of domino effects not only within the site under consideration, but also to nearby plants [1]. An inventory of the past domino accidents [2], reveals that explosion are the most frequent cause of domino effect (57%), followed by fires (43%). A study of 225 accidents involving domino effects [3], shows that storage areas are the most probable starters of a domino effect (35%), followed by process plant (28%). Also, the Journal of Risk Analysis and Crisis Response, Vol. 3, No. 2 (August 2013), 66-76 Published by Atlantis Press Copyright: the authors 66 willieb Typewritten Text willieb Typewritten Text Received 7 November 2012 willieb Typewritten Text Accepted 19 February 2013 willieb Typewritten Text Farid Kadri, Eric Chatelet and Patrick Lallement most frequent accident sequences are explosion-fire (27.6%), fire-explosion (27.5%) and fire-fire (18%). To address the problem posed by the assessment and/or analysis of domino effects in industrial sites, several methods and software tools have been developed [4, 5, 6, 7]. An analytic methodology for the quantitative assessment of industrial risk due to accidents triggered by seismic events has been developed [8]. This procedure is based on the use of available data (historical data) to assess the expected frequencies and magnitude of seismic events. A method for assessing domino effects based on Monte Carlo simulation has been developed by [9]., the authors developed an algorithm, which is based on conducting several hypothetical experiments to simulate the actual behavior of a multi-unit system. Recently, a review of methodologies and software tools used in the literature to the study of the cascading events [10], shows that, in the last decade, the available methodologies for the assessment of domino effects caused by heat load and overpressure to process equipments are based on the probit models [11, 12, 13]. The objective of this article is to present a methodology for the quantitative assessment of domino effects caused by heat radiation and overpressure to industrial/chemical plants and storage sites. Next-subsection is dedicated to a brief definition of the domino effect and its main features, potential sources of domino effects and the propagation process. Next, brief analysis of previous works is presented. In the third section we present a methodology for quantitative assessment of domino accidents in industrial sites. The fourth section uses a case study to illustrate the proposed model and to present typical results. The last section concluded this paper. 1.1. Domino effect and escalation There is no generally accepted definition of what constitutes domino effects in the context of accidents in the industrial plants, although various authors have provided suggestions [14, 15, 16, 17]. A domino accidental event may be considered as an accident in which a primary event propagates to nearby equipment (units), triggering one or more secondary events resulting in overall consequences more severe than those of the primary event [18]. The concept of escalation is a process that promotes the degradation of property (materials, equipments, systems industrials, ecosystems) and injury to people during development of the domino effect (increase damages). Thus, in the industrial field, we consider that any event spreading from equipment and/or industrial unit to another or from one site to another site should be classified as a domino event. According to the case histories concerning past domino accidents, all the accidental sequences where a relevant domino effect has took place have three common features [19]:  A primary accidental scenario, which initiates the domino accidental sequence;  The propagation of the primary event, due to an escalation vectors, generated by the physical effects of the primary scenario, that results in the damage of at least one secondary target;  One or more than one secondary accidental scenarios or events, involving the same or different plant units causing the propagation of the primary event. 1.2. Potential sources of domino effects Potential sources of domino effects are of different nature and are also linked to various initiating events. In general, they are distinguished by the nature of risks, from natural or anthropogenic. In the latter category, there are technological and organizational risks (unintentional) and the risks of malevolence (intentional), knowing that the purpose of study of domino effects takes into account the combination of these two risks. It is therefore possible to propose the decomposition of the nature of risks as follows: a) Natural origins (geological origins and/or atmospheric mainly) [20, 21]:  Climate origin: forest fires, runoff and floods, avalanches, hurricanes and tornadoes, storms;  Geological origin: landslides and earthquakes, tsunamis, volcanic eruptions and other natural emissions (gas, etc.). Published by Atlantis Press Copyright: the authors 67 The Assessment of Risk fire and explosion in Chemical Process Industry b) Human origins (organizational and malevolence) [22, 23, 24]:  Organizational origin: Humans failures (incorrect human action, lack of human action), defects in design, procedures and/or organizational;  Malevolence origin, thefts, sabotage and/or revenge action, damage of any kind attacks. These actions may touch or affect the material, but also the personal or sensitive information. c) Technological origin (fire, explosion and toxic releases):  Fire: pool fire, flash fire, fireball and jet fire;  Explosion: confined vapor cloud explosions (CVCE), boiling liquid expanding vapor explosion (BLEVE), vented explosion, vapor cloud explosion (VCE), dust explosion and mechanical explosion;  Toxic chemicals release: from process or storage sites and transportation accidents. These risks can be combined which significantly complicates the analysis. Sometimes, the very different nature of risks involves varied propagation processes. This also leads to the exploitation of different analysis methods (deterministic, probabilistic and quantitative methods). 1.3. The propagation process The propagation process is directly related to the potential source and the initiating event, but also to its immediate environment (field of danger). It is described by a physical-chemical process, but also informational whose evolution conditions are guided by features such as: physical (atmospheric, geological, hydrological) and material (buildings, sites, facilities, roads,...), ecological (vegetation, animals), informational (detections, observations and information systems) and human (individual behavior, organization and logistics, local demography). For more detailed about the propagation of danger from potential source to a potential target and the concepts of "source" and "target" and systemic approaches, it is advisable to refer to references [25, 26, 27]. 2. Domino effect analysis In the framework of domino effect analysis, the risk of explosion and fire, characterized by the possibility of an accident in an industrial site may lead to damage and serious consequences for the surrounding process equipment, people, goods and environment. These latter can generate four main events that may affect and/or cause the failure of the surrounding process equipments/units:  Overpressure/blast waves;  Heat load;  Projection of fragments (missiles);  Toxic release. Although several studies were dedicated to the assessment of domino effect caused by fires and explosions, only few models based on very simplistic assumptions are available for the assessment of equipment damage caused by heat load and overpressure in the framework of domino effect. The more simple approach proposed for the assessment of damage to equipment caused by fires and explosions. Several authors propose to consider zero probability of damage to equipment if the physical effect is lower than a threshold value for damage, and to assume a probability value of one if the physical effect is higher than a threshold value for damage [28,29, 30, 31]. A quantitative study, however, of the domino effect has been made by [32]. They have described possible approaches for quantifying the consequences of domino effects resulting from events giving rise to thermal radiation. A first approach evaluating the frequency accidental explosions was proposed by [33]. They provided a methodology for predicting domino effects from pressure equipment fragmentation. A simplified model proposed by [34].allows to assess the damage probability of process equipment caused by blast wave. The model is based on "experimental" evaluation of equipment displacement with the subsequent deformation and breakage of connections. The author defines the "probit function" (Y) relating equipment damage to the peak static overpressure (P^0) as follow: ln (1) where Y is probit function for equipment damage, P^0is peak static overpressure (Pa), a and b are the probit coefficients. Published by Atlantis Press Copyright: the authors 68 Farid Kadri, Eric Chatelet and Patrick Lallement The probit approach has been followed by [18, 35, 36], the authors have been published articles in which they analyzed and reviewed the existing models to develop a probabilistic model for damage evaluation of specific categories of industrial equipments. The damage probability model proposed by the authors takes into account four categories of industrial equipments (atmospheric vessels, pressurized vessels, elongated vessels, and small equipments). The probit coefficients and thresholds for overpressure damage probabilities for four equipment categories are represented in the table 1. Table 1. Probit coefficients for different equipment categories [36] Equipment category a b Threshold Atmospheric vessels -18.96 +2.44 22 kPa Pressurized vessels -42.44 +4.33 16 kPa Elongated equipments -28.07 +3.16 31 kPa Small equipments -17.79 +2.18 37 kPa To estimate the time to failure ttf of industrial equipments exposed to fire. A well known simplified model proposed by [37] is based on the probit approach. The authors proposed damage probability models that take into account the categories of industrial equipments. Table 2 presents the thresholds and probit models for two equipment categories. A methodology for domino effect analysis has been developed by [4] and, some applications in [38, 39]. The authors have cited that the intensity of heat radiation of 37 kW/m2 is sufficient to cause severe damage to process equipment in other installations that operate under atmospheric conditions. Also, a peak overpressure of 70 kPa is enough to cause severe damage to process equipment and may generate new accidents, either associated to new explosions or new events involving fires. A systematic procedure for the quantitative assessment of the risk caused by domino effect to industrial plants has been developed by [19]. This methodology aims to calculate the propagation probability of primary scenarios, the expected frequencies of domino events, and allowed to estimate the contribution of domino scenarios to individuals as well as societal risk. On industrial sites/storage areas, the heat load and overpressure generated by BLEVE explosions of tanks containing gas or highly pressurized liquids are threats to other surrounding equipment and can lead to successive explosions and fires. Several studies have been done on modeling the impact of BLEVE explosions on industrial installations [40, 41, 42, 43, 44]. Boiling liquid expanding vapor explosions (BLEVEs) are among the diverse major accidents which can occur in process industries. It is usually associated with the explosion of tanks containing flammable liquids (LPG). Therefore, to the effects of the BLEVE, one must add those corresponding to the fireball often occurring immediately after the explosion. On the whole, then, the physical effects from this type of explosion are usually i) thermal radiation, ii) overpressure (blast) and ii) fragments projection. The BLEVEs mechanism, the causes and consequences are presented by [45, 46]. Different formulas are used to quantify the heat radiation generated by fire. The radiation from fireball or pool fire on a receptor body located at a distance r from the center of this latter may be expressed by the following equation [47]: Table 2. Probability models and threshold values for the heat radiation, Y is the probit function, ttf is the time to failure (sec), V is the vessel volume (m3), and I is the amount of heat radiation received by the target vessel ( / ) [36] Equipment category Threshold Correlation Atmospheric vessels 15 / 12.54 1.847 10 1.128 2.667 10 9.887 Pressurized vessels 50 / 12.54 1.847 10 0.947 8.835 . Published by Atlantis Press Copyright: the authors 69 The Assessment of Risk fire and explosion in Chemical Process Industry I r (2) where I(r) is the heat radiation flow (kW/m ), F is the fraction of the generated heat radiated from the flame surface, m is the combustion velocity per unit surface area of the pool [ kg/ m .s)], τ , is the atmospheric transmissivity coefficient, H is a combustion heat (kJ/kg), D is the pool diameter. In experiments with explosives framework, the equivalent mass of TNT (m ) was used to evaluate the effects of potential damage of a quantity of fuel (hydrocarbon) given. The combustion energy available in a cloud of steam was converted into an equivalent mass of TNT (kg). m may be evaluated assuming that an exploding fuel mass behaves like exploding TNT on equivalent energy basis. Hence, the equivalent mass of TNT is estimated by using the following equation [48]: m ∆ (3) where μ is the explosion efficiency (0.03 to 0.1), m is the mass of fuel involved in the explosion (Kg), ∆H is the energy of explosion of the flammable gas (energy/mass) (MJ/kg), E is the energy of explosion of TNT (MJ/kg). In an explosion, the peak overpressure may be estimated using the following equation: P r . . . . . . (4) where is the peak of overpressure (kPa), and is atmospheric pressure (101.3 kPa), is a scaled distance ( / ) which may be estimated using an equivalent mass TNT ( ) as follow: (5) where r is distance from the center of the explosion. Note that, can be calculated by setting the threshold of peak of overpressure for each equipment categories. 3. Methodology An industrial site and storage areas contains many storage equipments/units under pressure that may be subjected to an external and/or internal incident. The escalation vectors (physical affects) generated after a unit rupture (explosion), may affect the surrounding units, building, personnel and environment. If the affected targets are damaged, these latter, may also explode and generate another threats to other surrounding facilities and so on. This accident chain is a domino effect and may lead to catastrophic consequences in an industrial plant. 3.1 Domino system We define a domino system as a system which consists at least of two subsystems ( , ), a source subsystem and a target subsystem (see Fig. 1):  A source subsystem: its failure may generate a danger (physical effects) that may affect other surrounding subsystems (heat load, overpressure, fragments, toxic releases), and  A target subsystem: it may be affected by the failure of sub-system sources. In addition to these physical effects, we may include the influencing factors that can influence or aggravate the target system damage (malicious acts, human and organizational factors, intervention system and weather conditions). In the case of domino effect analysis, the failure of a subsystem depends on the dynamic characteristics of the escalation vectors (input vector), threshold values and the aforementioned influence factors. Then, the domino system can be described by the following vector function: , , (6)  , …, : is a real vector (input vector) with p dimension in a space of physical state at time t. may be divided into two types of parameters, random physical parameters (physical effects) and influence factors (intervention system and human factor);  , …, : is a real vector (input vector) with g dimension, represents the deterministic input parameters of the system (physical characteristics of system like thresholds); Published by Atlantis Press Copyright: the authors 70 Farid Kadri, Eri  outp depe 3.2 Represen An industria ( , ,…, analysis, each three main sta  Stat corre syste resp affec by a  Stat esca corre affec  Stat vect thres To study the as starting p initiating eve least one faile transitions sta ic Chatelet and P , …, put with k dim ending on inpu ntation of syste al system com ). In the f h unit can be c ates: te 1: In norma esponding to em are less ectively. In cted by the e a primary even te 2: While th alation vecto esponding th cted, and te 3: While or(s) is gre shold value. domino accid point, the fai ent. Based on ed subsystem. ates in the cas Fig. 1. DoFig. 1. Do Patrick Lallement : is the v mensions, is ut parameters em states mposed of sev framework of characterized 1, 2, 3, al operation, t the input pa than the t this state, th escalation vec nt, he intensity o or(s) is hreshold value the value o ater than it dental sequen ilure of at le n the assump . Figure 2 pres e of two subsy omino system omino system ector of syst random varia . veral subsyste f domino ef by the follow the output val arameters of threshold val he unit may ctor(s) genera or the value(s) equal to e, the unit s of the escalat s correspond nce, one can t east one unit ption, there is sents the possi ystems/units. tem able ems ffect wing lues the lues be ated ) of its says tion ding take t as s at ible 3.3 In to thr out cor fai of Wh sys for Aft pro cal Th vec cal 3.4 Wh sub dam wh bet the dam eac 3 Failure prob normal opera the input par reshold values tput of syst rresponding t led. Then, the the system ma here is the stem) and is r which , fter calculatin obability lculated by the he total failur ctors that aff lculated with t 4 Domino effe hile failure bsystem, the mage radii (a hole system. tween a minim e impacted zo mage level is ch impacted z F bability ation, the outp rameters of th s respectiv tem for an en threshold val e failure funct ay define as fo e threshold c s the output of 0 , then ng the failu for each es e following eq re probability fects the targe the following ⋃ ect probability probability probability o ffected zones Domino effe mum of two z ones are pres increased in one. Fig. 2. Transiti put values c he system are vely. While the ntry point is g lue , the tion that desc ollows: criterion (defi f target system the system sa ure function, scalation vec quation: , 0 y for all t et subsystem equation: 0 y/affected zon is know of domino e ) may be eva ect consists ones. The dam ented in the involved area ion between sta corresponding e less than the e value of any greater than its system says cribes the state (7) fined for each m. If it exists ays failed. , the failure ctor may be (8) the escalation ( ) may be (9) nes wn for each effect and the aluated for the in interaction mage radii and figure 3. The as, but also on ates g e y s s e ) h i e e ) n e ) h e e n d e n Published by Atlantis Press Copyright: the authors 71 According to affected zone equipment lo probability this zone the and iii) saf process equip The probabil accidental seq where n is involving in probability th 4. The case-s The above d study in orde storage area. this case stud inventory are o the figure 3 es i) zone of c ocated in this 1, ii) zo failure probab fety zone; th pment 0 lity of each do quence) may b the number the domino hat each unit fr study efined metho er to assess Figure 4 sho dy. The type e shown in the Fig.3. The a 3, we can de certain destruc s area are fai one of possibl bility is betwe he failure pro 0. omino scenario be calculated ∏ r of the fai o sequence, from sequence dology was u domino effec ows the lay-o of equipmen table 3 bellow affected zones efine three m ction; all proc iled with fail le destruction een 0 obability of o (domino as follows: ( led sub-syste is the jo e i fails. used in the ca ct in the case out considered ts/units and th w. The A main cess lure n; in 1, the (10) ems oint ase- e of d in heir Tab Tank TK1 TK2-7 4.1 We rup gen ove aff Som the bee cas In we pro exp can des saf Assessment of Ris ble 3. Equipme k Typ Pressurize 7 Atmospher 1 Effects on su e assume tha pture (catastro nerate three e erpressure wa fect the surrou me simplifica e effects of he en considered se are tabulate the case-stud ere considere obability in fu plosion of the n define thre struction, ii) fety zone. Fig sk fire and explo ent considered i e Sub ed tank L ric tank Eth urrounding e at a primary ophic failure) escalation vec ave and iii) f unding equipm ations are used eat radiation a d. The influen ed in the table dy, only prim ed. The figu unction of the TK1 in case ee types of zone of pos g.4. Lay-out use osion in Chemical in the case-stud bstance Cont LPG 1 hanol 3 quipments y scenario ha of one tank. ctors; i) heat fragments, the ments. d in the prese and overpress nce parameter 4. mary and seco ure 5 shows distance resu of overpressu zone: i) zon ssible destruc ed for the case s l Process Industry dy tent (t) Failu 150 9 315 as caused the The latter can t radiation, ii) ese latter may ent study, only sure wave has rs used in this ondary events s the failure ulting from the ure effects. We ne of certain ction, and iii) study ry ure frequency 9 10 10 e n ) y y s s s e e e n ) Published by Atlantis Press Copyright: the authors 72 Fa Farid Kadri, Eri We remark th the area limit failure proba overpressure failure proba radiation. The failure p escalation v waves) are r figure 6 pres catastrophic f heat radiation (zone of pos probability Table 5. Proba ailed tank Es TK1 H TK1 H TK1 O TK1 O TK2 H TK2 H TK2 O TK2 O Fig. 5. Failure Taic Chatelet and P hat the whole ted by the radi ability waves, and ability probability, ectors (heat represented in sents the affe failure (ruptur n, Z1 (zone o sible destruct 0.9 10 ability due t over scalation Vector Heat radiation Heat radiation Overpressure Overpressure Heat radiation Heat radiation Overpressure Overpressure e probability res the case able 4. The infl Random : E m: Mass : Explo : Atmo : Frac D: Pool d Patrick Lallement process equip ius of 132 m 9.4 10 the radius o 10 in th due to the e radiation an n the followi ected zones g re) of the tan of certain des tion) estimate 0 respective to the effects of rpressure Target tank TK2-6 TK4 TK2-6 TK4 TK2-5-6 TK3-7 TK2-5-6 TK3-7 sulting from the of overpressur uence paramete m parameters Explosion energ s involved in th osion efficiency ospheric transm ction of the ge diameter pment that are have failed w in the case of 420 m w he case of h effects of the t nd overpress ing table 5. T generated by ks in the case struction) and ed for the fail ly. f heat radiation Failure proba 1.21 10 1.32 10 7.51 10 2.2 10 2.84 10 1.75 10 9.46 10 8.91 10 e rupture of the e ers used in the c gy he explosion y missivity enerated heat e in with of with heat two sure The the e of d Z2 lure and ability 15 4.2 To tha wa dom tab Tab Scen 1 2 3 4 Fig. 6 TK1 in case of heat rad Probabilis ~ 50 2. Domino eff o estimate the at the two ev aves) are ind mino sequen bulated in the t ble 6. The prob arios Domin 1 T 2 T 3 T 4 T 6. Affected zo diation and over stic distributio ~ ~ 0. ~ ~ ~ 2 fect scenarios e domino eff vents (heat r dependents. ce, for table 6. ability for each no effect sequenc TK1-TK4-TK5 TK4-TK6-TK7 TK4-TK6-TK7 TK5-TK4-TK1 ones Z1 and Z rpressure waves on 4.50,0.15 80,0.04 0.65,0.18 0.20,0.80 0.26,0.08 2 ,0.26 fect sequences radiation and The probabi each domino h considered dom ce Failure pr 1.17 1.08 8.75 2.02 Z2 in the case o s, R is spherical 2 s, we assume d overpressure ility of each o scenario is mino scenario obability ( ) 7 10 8 10 5 10 2 10 of heat radiati l tank rayon. e e h s ion. Published by Atlantis Press Copyright: the authors 73 The Assessment of Risk fire and explosion in Chemical Process Industry 5. Conclusion A quantitative method for the assessment of domino effects in industrial sites has been developed in this paper. It allows quantifying the effect of heat load and over pressuring waves in industrial plants and/or storage areas. Based on this method, we can evaluate the failure probability for each subsystem (unit), after the probability of domino scenario (domino sequence) may be evaluated for all the system. The three areas defined in this study (zone of certain destruction, zone of possible destruction, and safety zone) may be useful in the choice of safe distances between industrial equipments. Domino effect caused by fragments is not studied in this paper. However, the projectiles generated by an explosion of a tank (unit) containing gas or highly pressurized liquids are threats to other surrounding equipment and can lead to successive explosions and a chain of accidents. Hence, domino effect caused by fragments must be considered to evaluate the total failure probability for each equipment resulting from the combination of these events (heat load, overpressure and fragments). Also, heat radiation and overpressure effects can affect not only the industrial equipments but also environment and people. So, a human vulnerability models to the heat radiation and overpressure effects should be developed to estimate the individual and societal risk. The analysis above shows the importance of domino effect assessment in the framework of risk analysis. Hence, it shows that must much more importance be attached to the study of this phenomenon. Finally, domino effects need more scientific investigations, particularly in terms of quantitative assessment of risks and damage with probabilistic and deterministic modeling. Acknowledgements This work has been developed in the DISC project and supported by the ANR (Agence Nationale de la Recherche) from the French Research Ministry (http://www.agence-nationale-recherche.fr). References 1. Council Directive 96/82/EC on the control of major- accident hazards involving dangerous substances. Official Journal of the European Communities, January, 1997. 2. Abdolhamidzadeh B., Abbasi T., Rashtchian D., and Abbasi S.A., 2011, Domino effect in process-industry - An inventory of past events and identification of some patterns. Journal of Loss Prevention in the Process Industries, 24(5), 575–593. 3. Darbra R.M., Palacios A., and Casal J., 2010, Domino effect in chemical accidents: Main features and accident sequences. Journal of Hazardous Materials, pages 556- 573. 4. Khan F.I., and Abbasi S.A., 1998, Models for domino effect analysis in chemical process industries. Process Safety Progress-AIChE, 17(2):107-113. 5. Tixier J., Rault-Doumax J., Dandrieux A., Dimbour J.P., and Dusserre G., 2002, GeOsiris : Domino effects software. GOsiris: Logiciel de gestion des effets domino. In 13-ESREL, 1-5. 6. Lee, J.Y, Lee JW., Ko, J., and En Sup Yoon, ES., 2005. Optimization for allocating the explosive facilities in order to minimize the domino effect using nonlinear programming. Korean Journal of Chemical Engineering, 22(5):649-656. 7. Reniers G. L.L., and Dullaert W., 2007, DomPrevPlanning: User-friendly software for planning domino effects prevention. Safety Science, 45(10):1060-1081. 8. Antonioni, G., Spadoni G., and Cozzani V., 2007, A methodology for the quantitative risk assessment of major accidents triggered by seismic events. Journal of Hazardous Materials, 147(1-2):48-59. 9. Abdolhamidzadeh B., Abbasi, T., Rashtchian, D., and Abbasi, S.A., 2010, A new method for assessing domino effect in chemical process industry. Journal of Hazardous Materials, 182(1-3):416-426. 10. Kadri F., Châtelet E., and Elegbede C., 2013, Domino effect analysis and assessment of industrial sites: A review of methodologies and software tools, International Journal Of Computers & Technology (accepted, 2013). 11. Cozzani, V., and Salzano E., 2004, The quantitative assessment of domino effects caused by overpressure part I. Probit models. Journal of Hazardous Materials, A107:67-80. 12. Mingguang, Z., and Juncheng J., 2008, An improved probit method for assessment 13. of domino effect to chemical process equipment caused by overpressure. Journal of Hazardous Materials, (158), 208-286. 14. Landucci G., Gubinelli, G., Antonioni, G., and Cozzani V., 2009, The assessment of the damage probability of storage tanks in domino events triggered by fire. Accident Analysis and Prevention, 41(6):1206-1215. 15. Delvosalle, CH., 1996, Domino effects phenomena: Definition, Overview and Classification, European Seminar on Domino Effects. Leuven, Belgium, Federal Ministry of Employment, Safety Administration, Published by Atlantis Press Copyright: the authors 74 Farid Kadri, Eric Chatelet and Patrick Lallement Direction Chemical Risks, Brussels, Belgium, pages 5- 15. 16. Gledhill J. and Lines, I., 1998, Development of methods to assess the significance of domino effects from major hazard sites, CR Report 183, Health and Safety Executive. 17. Khan FII., and Abbasi S.A., 1999, Major accidents in process industries and an analysis of causes and consequences. Journal of Loss Prevention in the Process Industries, 12(5):361-378. 18. CCPS (Centre for Chemical Process Safety), 2000, Guidelines for Chemical Process Quantitative Analysis, Second Edition, American Institute of Chemical Engineers, New York. 19. Cozzani,V., and Salzano E., 2004, Threshold values for domino effects caused by blast wave interaction with process equipment. Journal of Loss Prevention in the Process Industries,(17)437- 447. 20. Cozzani V., et al., 2005, The assessment of risk caused by domino effect in quantitative area risk analysis. Journal of Hazardous Materials, 127:14-30. 21. [20] Krausmann, E., and Mushtaq F., 2008, A qualitative Natech damage scale for the impact of floods on selected industrial facilities. Natural Hazards, 46(2):179-197. 22. Steinberg L.J., Sengul H., and Cruz, A.M., 2008, Natech risk and management: an assessment of the state of the art. Natural Hazards, 46(2):143-152. 23. DiMattia D.G., Khan F.L., and Amyotte P.R., 2005, Determination of human error probabilities for offshore platform musters. Journal of Loss Prevention in the Process Industries, 18:488-501. 24. Mohaghegh, Z., and Mosleh, A., 2009, incorporating organizational factors into probabilistic risk assessment of complex socio-technical systems: Principles and theoretical foundations. Safety Science, 47:1139-1158. 25. Piwowar J., Châtelet E., and Laclémence P., 2009, An efficient process to reduce infrastructure vulnerabilities facing malevolence. Reliability Engineering and System Safety, 94:1869-1877. 26. Kervern G.Y., 1995, Eléments Fondamentaux des Cindyniques, Editions Economica, Paris 27. Perilhon P., 2000, MOSAR, techniques de l'ingénieur, traité sécurité et gestion des risques. SE 4 060, pages 1- 16. 28. Le Moigne J. L., 1994, Théorie du Système Général, théorie de la modélisation. Ed. PUF, Paris (4e édition). 29. Lees F.P., and Ang M. L., 1989, Safety Cases within the Control of Industrial Major Accident Hazards (CIMAH) Regulations. Butterworths Scientific in London. 30. Bagster D.F., and Pitblado R.M., 1991, The Estimation of Domino Incident Frequencies- An Approach. Trans IChemE, 69:195-199. 31. Purdy G., Pitblado R.M., and Bagster D.F., 1992, Tank Fire Escalation Modeling and Mitigation. In Internaional Symposium on Loss Prevention and Safety promotion in the process Industries. 32. Pettitt G. N., 1993, Evaluating the probability of major hazardous incidents as a result of escalation events. Journal of Loss Prevention Process Industries, 6(1):37- 46. 33. Lath P., Gautam G., and Raghavan K. V., 1992, Strategies for quantification of thermally initiated cascade effects. Journal of Loss Prevention Process Industries, 5(1):15-21. 34. [33] Scilly N. F., and Crowther J. H., 1992, Methodologies for Predicting Domino Effects from Pressure Vessel Fragmentation. In International Conference on Hazard Identification and Risk Analysis, Human Factors and Human Reliability in Process Safety, Florida, CCPS, AIChE, 15-17. 35. Eisenberg N. A., Lynch C. J., and Breeding R.J., 1975, Vulnerability Model: A Simulation System for Assessing Damage Resulting from Marine Spills, Report CG-D-136-75, Enviro Control Inc., Rockville, MD. 36. Cozzani, V., and Salzano, E., 2004, The quantitative assessment of domino effect caused by overpressure part II, case studies. Journal of Hazardous Materials, A107:81-94. 37. Cozzani, V., Gubinelli G., and Salzano, E., 2006, Escalation thresholds in the assessment of domino accidental events. Journal of Hazardous Materials, 129:1-21. 38. Cozzani, V., Antonioni G., and Spadoni G., 2006, Quantitative assessment of domino scenarios by a GIS- based software tool. Journal of Loss Prevention in the Process Industries, 19(5):463-477. 39. Khan F.I., and Abbasi S.A., 2001, An assessment of the likelihood of occurrence, and the damage potential of domino effect (chain of accidents) in a typical cluster of industries. Journal of Loss Prevention in the Process Industries, 14(4):283-306. 40. Khan F.I, and Abbasi S.A., Estimation of probabilities and likely consequences of a chain of accidents (domino effect) in Manali Industrial Complex. Journal of Cleaner Production, 9(6):493-508. 41. Baum M. R., 1988, Disruptive failure of pressure vessels: preliminary design guidelines for fragment velocity and the extent of the hazard zone. Journal of Pressure Vessel Technology, 110(2):168-176. 42. Birk A.M., 1996, Hazards from propane BLEVEs: an update and proposal for emergency responders. Journal of Loss Prevention in the Process Industries, 9:173-181. 43. Venart J.E.S., 2000, Boiling liquid expanding vapor explosions (BLEVE); possible failure mechanisms and their consequences. in: Proceedings of the IChemE Symposium Series, 147:121-137. 44. Planas-Cuchi, E., Salla J.M., and Casal J., 2004, Calculating overpressure from BLEVE explosions. Journal of Loss Prevention in the Process Industries, 17(6):431-436. 45. Ogle R.A., Ramirez J.C.,, and Smyth, S.A., 2012, Calculating the explosion energy of a boiling liquid expanding vapor explosion using exergy analysis. Process Safety Progress, 31(1):51-54. 46. Tellez C., and Pena J.A., 2002, Boiling-liquid expanding vapor explosion (BLEVE): an introduction to consequence and vulnerability analysis. Chem. Eng.Educ, 36:206-211. 47. Abbasi, T., and Abbasi S.A., 2007, The boiling liquid expanding vapor explosion (BLEVE): Mechanism, Published by Atlantis Press Copyright: the authors 75 The Assessment of Risk fire and explosion in Chemical Process Industry consequence assessment, management. Journal of Hazardous Materials, 141(3):489-519. 48. Van den Bosh C. J. H. and Weterings R. A. P. M., 1996, Methods for the Calculation of Physical Effects due to releases of hazardous materials (liquids and gases) Yellow Book. The Hague. 49. Daniel A.C., and Louvard J.F., 2002, Chemical Process Safety: Fundamentals with Applications. Prentice Hall International Series in the Physical and Chemical Engineering Sciences, 2002. Published by Atlantis Press Copyright: the authors 76