Guidelines for quantitative risk assessment

1 PUBLICATIEREEKSGEVAARLIJKE STOFFENG uidelines for quantitative risk assessment3 Publication Series on Dangerous Substances (PGS 3) Guidelines for quantitative risk assessmentPreamblePREAMBLES tarting from June 1st2004, the Advisory Council on Dangerous Substances (Adviesraad Gevaarlijke Stoffen - AGS) was installed by the Cabinet. At the same time the Committee for the Prevention of Disasters (Commissie voor de Preventie van Rampen- CPR) was abolished. CPR issued several publications, the so-called CPR- Guidelines (CPR-richtlijnen), that are often used in environmental permits, based on the Environmental Protection Law, and in the fields of of labour safety, transport safety and fire CPR- Guidelines have been transformed into the Publication Series on Dangerous Substances (Publicatiereeks Gevaarlijke Stoffen PGS).

2 The aim of these publications is generally the same as that of the CPR- Guidelines . All CPR- Guidelines have been reviewed, taking into account the following questions:1. Is there still a reason for existence for the guideline or can the guideline be abolished;2. Can the guideline be reintroduced without changes or does it need to be PGS 3 edition of the Guidelines for quantitative risk assessment hasn t been changed in regard of the first 1999 on behalf of my colleagues at the Ministries of Transport, Social Affairs and of the Interior,The State Secretary of Housing Spatial Planning and the Environment (VROM). Drs. van GeelDecember 2005 Guideline for quantitative risk assessment Purple book CPR 18 EPart one: Establishmentsdr. Uijt de Haagdr.

3 AleRVIMP refacePREFACEThis rep ort documents the methods to calculate the risks due to dangerous substances in theNetherlands using the models and data available. Calculation of the risks relates, on the one hand,to stationary installations and, on the other, to transp ort and related rep ort consists of two p arts. Part 1, describing the methods to calculate the risks ofstationary installations, was written by the National Institute of Public Health and theEnvironment (RIVM ) under a sup ervisory committee of rep resentatives from the subcommissionon Risk Evaluation of the Committee for the Prevention of Disasters (CPR-RE). Part 2, drawnup under the resp onsibilit y of t he M inist ry of T ransp ort , Public Works and Wat er M anagement ,describes the calculation of the risks connected with the transp ort of dangerous goods, based ont he ap p roach develop ed in accordance wit h t he M inist ry of Housing, Sp at ial Planning and t heEnvironment and set down in the last few y ears in various the rep ort describes the p resent-day calculation methods (in p ractice, no bettermethods are currently available), discussions on a number of subjects in the sup ervisorycommittee led to the conclusion that additional research would be necessary to guarantee thequality of the calculation methods in the future.

4 Three subjects for study were indicated:A. The failure frequencies of stationary installations. Failure frequencies are based on the so-called COVO study from 1981. Additional failure frequencies have been determined in variousst udies carried out for t he Dut ch government over t he y ears. Recent ly, new st udies have beenp ublished, rep orting different figures - mostly higher - for a number of failure frequencies. Amore det ailed st udy on t he failure frequencies will be carried out , concent rat ing esp ecially onthe original data T he met eorological model. Disp ersion calculat ions are carried out as p art of t he risk analy sesusing generally accep t ed met eorological models and t he corresponding meteorological he nat ional model used in air p ollut ion calculat ions has recent ly been adap t ed t o include newinsight s.

5 At t he moment , met eorological st at ist ics are not sufficient ly available t o ap p ly t hisnew meteorological model to risk analy ses. The relevance of the new model to risk analy sesshould be ascertained; furthermore, the consequences which the new model, including themodel p arameters, could have on the results of calculating risks should be examined. The studyon t hese consequences will be st art ed up in t he short t Differences in risk calculations for transp ort and for stationary installations. The method tocalculate the risks of transp orting dangerous goods is comp arable to the calculation methodap p lied to stationary installations. During the last few y ears, the basic p rincip les of riskanaly sis have been discussed and established with the p arties involved.

6 Since develop ments inthe risk calculation methods for transp ort and stationary installations were sep arate, severaldifferences exist bet ween t he basic p rincip les in risk calculat ions for t ransp ort and forstationary installations. These differences relate, among other asp ects, to the frequency ofcatastrop hic failure of tank wagons relative to stationary tanks and to certain loss ofcont ainment Committee for the Prevention of Disasters considers it imp ortant to have reliable riskcalculations for stationary installations and for transp ort of dangerous goods; these should, as faras p ossible, also be founded on similar basic p rincip les. It is t herefore advisable t o analy se t hebasic p rincip les of the calculation methods and to study the consequences of removing thedifferences in the calculation methods.

7 Both the M inistries mentioned above can then decidewhether these differences should actually be discussions show that the methods of risk analy sis are still being furt her develop ed. T heCommit t ee for t he Prevent ion of Disast ers is p leased t hat wit h t he p ublicat ion of t his rep ort asubst ant ial cont ribut ion will be made t o t he furt her develop ment of t his risk analy sis inst rument .The Committee thanks the government exp erts, research institutes and industry for theircontributions. The Committee for the Prevention of Disasters is convinced that the rep ort willbe of great value for all those dealing with risk analy sis and risk management. Table of LE O F C O N T EN T S1. INTRODUCTION2 . S ELEC T IO N O F T HE IN S TALLAT IO N S FO R T HE Q EXCLUSION OF PARTICULAR 2.

8 THE SELECTION Definition of installations in an Calculation of the indication number, Calculation of the selection number, Selection of Sp ecific p PROCEDURE TO ASSESS THE OBLIGATION TO MAKE A SAFETY AN EXAMPLE . LO S S O F C O N TAIN MEN T EVEN T LOSS OF CONTAINMENT EVENTS AT Stationary p ressurised tanks and Stationary atmosp heric tanks and Pip Pump Heat Pressure relief LOCs for storage in Storage of exp Transp ort units in an . MO D ELLIN G S O URC E T ERM AN D D IS PERS IO PROPERTIES OF OUTFLOW REPRESSION Blocking sy Other rep ression sy POOL VAPOUR CLOUD Coup ling outflow and vap our cloud disp M odelling t he vap our cloud disp Release inside a Fires and p lume of Direct Delay ed Substances both toxic and EFFECTS OF IGNITION OF A VAPOUR RUPTURE OF METEOROLOGICAL MODEL TO CALCULATE THE PROBABILITY OF DELAYED METEOROLOGICAL.

9 MO D ELLIN G EXPO S URE AN D D DAMAGE Probit Toxic exp Pressure effects for a vap our cloud exp Survey of the p op ulation p Fraction indoors and . C ALC ULAT IO N AN D PRES EN TAT IO N O F RES ULT CALCULATION OF THE INDIVIDUAL RISK AND THE SOCIETAL Definition of the Individual Risk Societal Risk Definition of ignition events for flammable Probabilit y of deat h, Pd, and fract ion of deat hs, Fd, for t oxic subst Probabilit y of deat h, Pd, and fract ion of deat hs, Fd, for PRESENTATION OF THE PROBABILITY THAT THE GRID POINT IS COVERED BY THE CLOUD, SAMPLE CALCULATION OF THE INDIVIDUAL RISK AT A GRID Q UAN TITATIVE EN VIRO N MEN TAL RIS K AN ALYS IS8 . T HE US E O F N EW MO D ELS IN A Q RATable of UN C ERTAIN TY IN A Q SOURCES OF Starting p M Parameter Use of the QUANTIFICATION OF REFERENCESGLO S S INTRODUCTIONA quantitative Risk assessment (QRA) is a valuable tool for determining the risk of the use,handling, transp ort and storage of dangerous substances.

10 QRAs are used to demonstrate the riskcaused by the activity and to p rovide the comp etent authorities with relevant information toenable decisions on t he accep t abilit y of risk relat ed t o develop ment s on sit e, or around theestablishment or transp ort the results of a QRA in the decision-making p rocess are to be used, they must be verifiable,rep roducible and comp arable. These requirements necessit at e QRAs made on t he basis of similarstarting-p oints, models and basic data. Ideally, differences in QRA results should only arise fromdifferences in p rocess- and site-sp ecific information. A number of documents for attainingcomp arabilit y in t he QRA calculat ions have been p ublished over t he y ears. T he Commit t ee forthe Prevention of Disasters (CPR) has issued three rep orts describing the methods to be used in aQRA calculation, namely the Red Book , the Yellow Book and the Green Book.