IOGP-Report 434-06-1 - Ignition Probabilities

REPORT SEPTEMBER
434-06 2019
RISK ASSESSMENT DATA DIRECTORY
Ignition Probabilities
responsible activity
Acknowledgements
Safety Committee
Photography used with permission courtesy of

©Opla/iStockphoto and ©Zsolt Biczó/iStockphoto (Front cover)
©Photo_Concepts/iStockphoto (Back cover)
Feedback
IOGP welcomes feedback on our reports: [email protected]
Disclaimer
Whilst every effort has been made to ensure the accuracy of the information
contained in this publication, neither IOGP nor any of its Members past present or
future warrants its accuracy or will, regardless of its or their negligence, assume
liability for any foreseeable or unforeseeable use made thereof, which liability is
hereby excluded. Consequently, such use is at the recipient’s own risk on the basis
that any use by the recipient constitutes agreement to the terms of this disclaimer.
The recipient is obliged to inform any subsequent recipient of such terms.
This publication is made available for information purposes and solely for the private
use of the user. IOGP will not directly or indirectly endorse, approve or accredit the
content of any course, event or otherwise where this publication will be reproduced.
Copyright notice
The contents of these pages are © International Association of Oil & Gas Producers.
Permission is given to reproduce this report in whole or in part provided (i) that
the copyright of IOGP and (ii) the sources are acknowledged. All other rights are
reserved. Any other use requires the prior written permission of IOGP.
These Terms and Conditions shall be governed by and construed in accordance

with the laws of England and Wales. Disputes arising here from shall be exclusively
subject to the jurisdiction of the courts of England and Wales.
REPORT SEPTEMBER
434-06 2019
RISK ASSESSMENT DATA DIRECTORY
Revision history
VERSION DATE AMENDMENTS
1.0 September 2019 First release

Contents
Abbreviations 5
1. Introduction 6
2. Summary of recommended data 7

2.1 Ignition Probability Curves 7
2.2 Ignition timing 22
2.3 LNG Releases 24
2.4 Blowout Ignition Probabilities 24
2.5 Onshore Gas Pipeline Ignition Probabilities 25
3. Guidance on use of the UKOOA model 27

3.1 General Validity 27
3.2 Alternative Approaches 27
3.3 Uncertainties 32
4. Review of primary data source 33
5. Alternative ignition models 35

5.1 Ignition modelling of process leaks in Norwegian offshore O&G industry – MISOF, OLF & TDIIM 35
5.2 BEVI Model 41
5.3 The CCPS Model 44
6. Recommended data sources for further information 45
7. References 46
4
Abbreviations
FPSO Floating Production Storage and Offloading (Installation)
LPG Liquefied Petroleum Gas
NAP Normal Atmospheric Pressure
NUI Normally Unmanned Installation
OLF Oljeindustriens landsforening
QRA Quantitative Risk Assessment
SIMOPS Simultaneous Operations
TDIIM Time Dependant Internal Ignition Model
UKOOA United Kingdom Offshore Operators Association
5
1. Introduction
The data presented in section 2 provide estimates of the probabilities of hydrocarbon

releases igniting to result in an explosion and/or a sustained fire. This data may be applied
to any of the leak types described in the Process Release Frequencies datasheet1.
The values presented relate to “total” ignition probability, which can be considered as the
sum of the probabilities of immediate ignition and delayed ignition. Immediate ignition
can be considered as the situation where the fluid ignites immediately on release through
auto-ignition or because the accident which causes the release also provided an ignition
source. Delayed ignition is the result of the build-up of a flammable vapour cloud which is
ignited by a source remote from the release point. It is assumed to result in flash fires or
explosions and also to burn back to the source of the leak resulting in a jet fire and/or a
pool fire.
The datasheets presented in section 2.1 provide probabilities which are considered
appropriate for use in QRA studies where a relatively coarse assessment is acceptable.
Later sections refer to a more detailed approach for QRAs using the “full” UKOOA model
where this is considered to be required.
Some details are also provided of other ignition models in use for specific situations or
preferred by national regulators.
Specific information on the ignition probabilities of blowouts and onshore pipelines are
given in section 2.4 and section 2.5 respectively.
1 With the exception of “zero pressure” releases, where the limited inventory and hence cloud size would result in a lower ignition
probability than would be predicted using this approach.
6
2. Summary of recommended data
2.1 Ignition Probability Curves

Data presented in this section come in the form of 28 mathematical functions drawn from
the UKOOA look-up correlations (see section 4.0) which relate ignition probabilities in air2
to release rates for typical scenarios both onshore and offshore and published in a report
by the Energy Institute [1]. The various scenarios are summarised in Table 2-1, Table 2-2
and Table 2-3. The functions themselves are given in both tabular and graphical form in the
data sheets which follow.
The curves of ignition probability vs. release rate comprise a number of sections, each of
which is a straight line when plotted on log-log axes.
These curves represent “total” ignition probability. The method assumes that the immediate
ignition probability is 0.001 and is independent of the release rate. As a result, all the
curves start at a value of 0.001 relating to a release rate of 0.1 kg/s. Users of the data may
wish to adopt this value and obtain delayed ignition probabilities by subtracting 0.001 from
the total ignition probability, e.g., an ignition probability value of 0.004 obtained from the
look-up correlations can be considered as an immediate ignition probability of 0.001 and a
delayed ignition probability of 0.003.
The definition of “immediate” in this context has tended to be interpreted in different ways
and is often confused with “early” ignition which may be defined, for example, as an ignition
occurring within 30 seconds or 60 seconds of the start of the release. This confusion may
lead to incorrect interpretation of the curves. Further discussion is provided in Section 2.2.
The overpressure resulting from a delayed ignition will depend on the size of the cloud
when it is ignited together with other factors such as degree of confinement, degree of
congestion and ignition location. The time of ignition is related to the rate at which the
flammable cloud grows and this is dependent on the release rate, ventilation conditions
and the distribution of ignition sources. Delayed ignitions which occur while the cloud
is still relatively small will have low overpressures which will not be sufficient to cause
damage to structures or equipment. Likewise, the low overpressure will not result in harm
to persons although the effects of radiation and changes in the composition of gasses
which the worker breaths may do. The later the ignition, the greater the probability workers
will have moved away from the vicinity of the explosion.
For more detailed risk assessments, users may wish to sub-divide the consequences of
delayed ignitions to account for the variation in cloud size at time of ignition. This may
include associating a proportion of delayed ignitions with overpressures great enough to
cause certain levels of harm or damage. The recommended approach is to carry out a
probabilistic explosion analysis which calculates overpressure exceedance curves and use
these within the risk assessment.
2 Ignition probabilities in other atmospheres, e.g. oxygen enriched or chlorine, are outside the scope of this datasheet.
7
Although a distinction is made between “immediate” and “delayed” ignitions in this

model, it may be more convenient to consider only the total ignition probability and then to
calculate the conditional probability of the resulting overpressure exceeding a particular
magnitude and the conditional probability of escape prior to ignition as part of a separate
piece of analysis.
Apart from situations such as a release with a high water cut or a short duration release
which stops before ignition occurs, it should be assumed that a sustained jet fire or spray
fire will result irrespective of the time of ignition. The consequence assessment should not
double count fatalities due to the initial flash fire which burns back to the fire.
Guidance on the timing of ignitions is given section 2.2. but users should not consider
ignitions occurring before a certain time to be equivalent to an immediate ignition and then
to apply a probability other than the 0.001 inherent in this model.
However, historically, to increase the efficiency of QRA studies it is often assumed that
ignitions that occur immediately and during the early stages of the dispersion will result in
fire scenarios which dominate the consequence. For delayed ignitions, it is assumed that
either flash fire or explosion overpressure is the dominating consequence based on steady
state modelling. This is an acceptable practice because it leads to an overall conservative
assessment compared with historical incident data. For more detailed studies, a time
dependent approach to the consequence modelling may be more appropriate.
The dispersion modelling underpinning the ignition model, and hence the look-up
correlations, is not valid for LNG releases. An approach for dealing with such releases is
given in Section 2.3. Similarly, the approach is not appropriate for high pressure natural gas
releases where reference should be made to Section 2.5.
Table 2-1: Onshore Ignition Scenarios
Scenario
Look-up Release Type Application
No.
1 Pipe Liquid Industrial Releases of flammable liquids that do not have any significant flash fraction
(Liquid Releases from onshore (10% or less) if released from onshore cross-country pipelines running through
pipeline in industrial area) industrial or urban areas.
2 Pipe Liquid Rural Releases of flammable liquids that do not have any significant flash fraction
(Liquid Releases from onshore (10% or less) if released from onshore cross-country pipelines running through
pipeline in industrial area) rural areas.
3 Pipe Gas LPG Industrial Releases of flammable gases other than buried natural gas pipelines, vapour
(Gas or LPG release from onshore or liquids significantly above their normal (Normal Atmospheric Pressure
pipeline in an industrial area) (NAP)) boiling point from onshore cross-country pipelines running through
industrial or urban areas.
For buried natural gas pipelines the method given in section 2.5 is
recommended.
4 Pipe Gas LPG Rural Releases of flammable gases other than buried natural gas pipelines, vapour
(Gas or LPG release from onshore or liquids significantly above their normal (NAP) boiling point from onshore
pipeline in a rural area) cross-country pipelines running through rural areas.
For buried natural gas pipelines the method given in section 2.5 is
recommended.
5 Small Plant Gas LPG Releases of flammable gases, vapour or liquids significantly above their normal
(Gas or LPG release from small (NAP) boiling point from small onshore plants (plant area up to 1200 m2, site
onshore plant) area up to 35,000 m2).
8
Scenario
No.
6 Small Plant Liquid Releases of flammable liquids that do not have any significant flash fraction
(Liquid release from small onshore (10% or less) if released from small onshore plants (plant area up to 1200 m2,
plant) site area up to 35,000 m2) and which are not bunded or otherwise contained.
7 Small Plant Liquid Bund Rural Releases of flammable liquids that do not have any significant flash fraction
(Liquid release from small onshore (10% or less) if released from small onshore plants (plant area up to 1200 m2,
plant where the spill is bunded) site area up to 35,000 m2) and where the liquid releases from the plant area are
suitably bunded or otherwise contained.
8 Large Plant Gas LPG Releases of flammable gases, vapour or liquids significantly above their normal
(Gas or LPG release from large (NAP) boiling point from large onshore outdoor plants (plant area above 1200
onshore plant) m2, site area above 35,000 m2).
9 Large Plant Liquid Releases of flammable liquids that do not have any significant flash fraction
(Liquid release from large onshore (10% or less) if released from large onshore outdoor plants (plant area above
plant) 1200 m2, site area above 35,000 m2) and which are not bunded or otherwise
contained.
10 Large Plant Liquid Bund Rural Releases of flammable liquids that do not have any significant flash fraction
(Liquid Released from large onshore (10% or less) if released from large onshore outdoor plants (plant area above
plant where spill is bunded) 1200 m2, site area above 35,000 m2) and where the liquid releases from the
plant area are suitably bunded or otherwise contained.
11 Large Plant Congested Gas LPG Releases of flammable gases, vapour or liquids significantly above their normal
(Gas or LPG released from a large (NAP) boiling point from large onshore plants (plant area above 1200 m2, site
confined or congested onshore plant) area above 35,000 m2), where the plant is partially walled/roofed or within a
shelter or very congested.
12 Tank Liquid 300m x 300m Bund Releases flammable liquids that do not have any significant flash fraction (10%
(Liquid release from a large confined or less) if released from very large onshore outdoor storage area ‘tank farm’
or congested onshore plant) (e.g., spill in a large multi-tank bund over 25,000 m2 area).
See curve No. 30 “Tank Liquid – diesel, fuel oil’ if liquids are stored at ambient
conditions below their flash point.
13 Tank Liquid 100m x 100m Bund Releases of flammable liquids that do not have any significant flash fraction
(Liquid release from onshore tank (10% or less) if released from onshore outdoor storage area ‘tank farm’ (e.g.,
farm where spill is limited by small spill in a large tank bund containing four or fewer tanks, or any other bund less
or medium sized bund) than 25,000 m2 area).
See curve No. 30 “Tank Liquid – diesel, fuel oil’ if liquids are stored at ambient
conditions below their flash point.
14 Tank Gas LPG Plant Releases of flammable gases, vapour or liquids significantly above their normal
(gas or LPG release from onshore (NAP) boiling point from onshore outdoor storage tanks located in a ‘tank farm’
tank farm within the plant) entirely surrounded by plants. For tank farms adjacent to plants use curve No.
15 “Tank Gas LPG Storage Industrial” or Curve No. 16 “Tank Gas LPG Storage
Only Rural” look-up correlations. Releases from process vessels or tanks
inside plant areas should be treated as plant releases.
15 Tank Gas LPG Storage Industrial Releases of flammable gases, vapour or liquids significantly above their normal
(Gas or LPG released from onshore (NAP) boiling point from onshore outdoor storage tanks located in a ‘tank farm’
tank farm sited adjacent to a plant or adjacent to plants or situated away from plants in an industrial or urban area.
away from the plant in an industrial
area)
16 Tank Gas LPG Storage Only Rural Releases of flammable gases, vapour or liquids significantly above their normal
(Gas or LPG released from onshore (NAP) boiling point from onshore outdoor storage tanks located in a ‘tank farm’
tank farm sited adjacent to a plant or adjacent to plants or situated away from plants in a rural area.
away from the plant in an industrial
area)
Source: Energy Institute [1]
9
Table 2-2: Offshore Ignition Scenarios
Scenario
No.
17 Offshore Process Liquid Releases of flammable liquids that do not have any significant flash fraction
(Liquid release from offshore process (10% or less) if released from within offshore process modules.
module)
18 Offshore Process Liquid NUI Releases of flammable liquids that do not have any significant flash fraction
(Liquid release from offshore process (10% or less) if released from within offshore process modules or decks on
area on NUI) NUIs.
19 Offshore Process Gas Open Deck NUI Releases of flammable gases, vapour or liquids significantly above their normal
(Gas release from offshore process (NAP) boiling point from an offshore process weather deck/ open deck on NUIs.
open deck area on NUI) Can also be used for open/uncongested weather decks with limited process
equipment on larger attended integrated platforms.
20 Offshore Process Gas Typical Releases of flammable gases, vapour or liquids significantly above their
(Gas release from typical offshore normal (NAP) boiling point from within offshore process modules or decks
process module) on integrated deck/conventional installations). Process modules include
separation, compression, pumps, condensate handling, power generation, etc.
If the module is mechanically ventilated or very congested – see curve No. 22
“Offshore Process Gas Congested or Mechanical Vented Module”.
21 Offshore Process Gas Large Module Releases of flammable gases, vapour or liquids significantly above their normal
(gas release from typical offshore (NAP) boiling point from within large offshore process modules or decks on
process module) integrated deck/conventional installations (module greater than 1000 m2 floor
area). Process modules include separation, compression, pumps, condensate
handling, power generation, etc. If the module is mechanically ventilated
or very congested – see curve No. 22 ‘Offshore Process Gas Congested or
Mechanical Vented Module’.
22 Offshore Process Gas Congested or Releases of flammable gases, vapour or liquids significantly above their
Mechanical Vented Module normal (NAP) boiling point from within offshore process modules or decks
(Gas released from a mechanically on integrated deck/conventional installations: applies where the module is
ventilated or very congested offshore enclosed and has a mechanical ventilation system or is very congested (volume
process module) blockage ratio => 0.14 and less than 25% of area of the end walls open for
natural ventilation)
23 Offshore Riser Releases from offshore installation risers in the air gap area where there is
(Gas release from typical offshore little chance of the release entering process areas on the installation (e.g.,
riser in air gap) solid decks, wind walls). Applies to partial flashing oil or gas releases.
May also be used for blowouts with well positioned diverters directing any
release away from the installation (see also curve No. 27 “Offshore Engulf –
blowout riser”).
Note that this correlation relates to situations where the distribution and
intensity of ignition sources are typical for offshore installations. It should not
be applied to incidents of ships colliding with risers since the collision itself will
provide an intense ignition source which will have a high probability of ignition.
In this case, a conservative value in the range of 0.9 to 1 is recommended.
24 Offshore FPSO Gas Releases of flammable gases, vapour or liquids significantly above their normal
(Gas release from offshore FPSO (NAP) boiling point from within offshore process modules or decks on FPSOs.
process module) This correlation was specifically developed for weathervaning FPSOs where any
release would be directed along the deck of the FPSO.
See curve No. 25 “Offshore FPSO Gas Wall” if the release is from an area
downwind of a transverse wall across the FPSO deck.
25 Offshore FPSO Gas Wall Releases of flammable gases, vapour or liquids significantly above their normal
(Gas release from offshore FPSO (NAP) boiling point from within offshore process modules or decks on FPSOs. This
process module behind a transverse correlation applies if the release is from an area downwind of a transverse wall
solid wall) across the FPSO deck. This correlation applies to weathervaning FPSOs only.
26 Offshore FPSO Liquid Releases of flammable liquids that do not have any significant flash fraction
(Liquid release from typical offshore (10% or less) if released from within offshore process modules or decks on
FPSO process module) FPSOs. This correlation applies to weathervaning FPSOs only.
10
Scenario
No.
27 Offshore Engulf – blowout – riser Releases from drilling or well working blowouts or riser failures under open
(Major release which can engulf an grated deck areas where the release could engulf the entire installation and
entire offshore installation) reach into platform areas: applies to partial flashing oil or gas releases. (see also
curve No. 23 “Offshore Riser” for riser releases and blowouts with diverters).
Note that this correlation relates to situations where the distribution and
intensity of ignition sources are typical for offshore installations. It should not
be applied to incidents of ships colliding with risers since the collision itself will
provide an intense ignition source which will have a high probability of ignition.
In this case, a conservative value in the range of 0.9 to 1 is recommended.

Note: Curve Nos. 28 and 29 related to Cox, Lees and Ang formulation which were included in the Energy Institute report for
comparison and are not reproduced in this report.
Table 2-3: Special (Derived) Ignition Scenarios
Scenario
No.
30 Tank Liquid – diesel fuel oil Releases of combustible liquids stored at ambient pressure and at temperatures
(Liquid Release from onshore tank below their flash point (e.g., most gas, oil, diesel and fuel oil storage tanks)
farm of liquids below their flash point, from onshore outdoor storage area “tank farm”. This look-up correlation can
e.g., diesel or fuel oil) be applied to releases from tanks and low pressure transfer lines or pumps in
the tank farm/ storage area. However, it should not be used for high-pressure
systems (over a few barg): in these situations, use curve No. 12 “Tank Liquid 300
m x 300 m Bund” or curve No. 13 “Tank Liquid 100 x 100 m Bund”
11
Data Sheet 1: Scenarios 1 – 4
1 - Pipe Liquid Industrial 2 - Pipe Liquid Industrial 3 - Pipe Liquid Industrial 4 - Pipe Liquid Industrial
Liquid release from onshore pipeline Liquid release from onshore pipeline Gas or LPG release from onshore Gas or LPG release from onshore
in industrial area in rural area pipeline in an industrial area) pipeline in a rural area
Releases of flammable liquids that do not Releases of flammable liquids that do not Releases of flammable gases other than Releases of flammable gases other than
have any significant flash fraction (10% or have any significant flash fraction (10% or buried natural gas pipelines, vapour or natural gas pipelines, vapour or liquids
less) if released from onshore cross country less) if released from onshore cross country liquids significantly above their normal significantly above their normal (NAP)
pipelines running through industrial or pipelines running through rural areas (NAP) boiling point from onshore cross boiling point from onshore cross country
urban areas country pipelines running through pipelines running through rural areas.
industrial or urban areas.
See section 2.4 for recommendations for
See section 2.4 for recommendations for natural gas pipelines.
natural gas pipelines.
Release Rate (kg/s) Ignition Probability Release Rate (kg/s) Ignition Probability Release Rate (kg/s) Ignition Probability Release Rate (kg/s) Ignition Probability
0.1 0.0018 0.1 0.0018 0.1 0.0011 0.1 0.0011
0.2 0.0027 0.2 0.0027 0.2 0.0018 0.2 0.0012
0.5 0.0044 0.5 0.0037 0.5 0.0036 0.5 0.0014
1 0.0065 1 0.0041 1 0.0060 1 0.0015
2 0.0096 2 0.0045 2 0.0101 2 0.0016
5 0.0160 5 0.0050 5 0.0199 5 0.0018
10 0.0236 10 0.0055 10 0.0332 10 0.0020
20 0.0348 20 0.0060 20 0.0554 20 0.0035
50 0.0580 50 0.0067 50 0.1091 50 0.0073
100 0.0700 100 0.0070 100 0.1821 100 0.0126
200 0.0700 200 0.0070 200 0.3041 200 0.0220
500 0.0700 500 0.0070 500 0.5989 500 0.0459
1000 0.0700 1000 0.0070 1000 1.0000 1000 0.0800
1
Ignition probability
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Pipe Liquid Industrial (1) Pipe Liquid Rural (2)
Pipe Gas LPG Industrial (3) Pipe Gas LPG Rural (4)
12
5 - Small Plant Gas LPG 6 - Small Plant Gas LPG 7 - Small Plant Gas LPG
Gas or LPG release from small onshore Liquid release from small onshore Liquid release from small onshore
plant plant plant where spill is bunded
Releases of flammable gases, vapour or Releases of flammable liquids that do not Releases of flammable liquids that do not
liquids significantly above their normal have any significant flash fraction (10% or have any significant flash fraction (10% or
(NAP) boiling point from small onshore less) if released from small onshore plants less) if released from small onshore plants
plants (plant area up to 1200m2, site area (plant area up to 1200m2, site area up to (plant area up to 1200m2, site area up to
up to 35,000m2) 35,000m2) and which are not bunded or 35,000m2) and where the liquid releases
otherwise contained. from the plant area are suitably bunded or
otherwise contained.
Release Rate (kg/s) Ignition Probability Release Rate (kg/s) Ignition Probability Release Rate (kg/s) Ignition Probability
0.1 0.0011 0.1 0.0011 0.1 0.0011
0.2 0.0014 0.2 0.0014 0.2 0.0014
0.5 0.0020 0.5 0.0019 0.5 0.0019
1 0.0025 1 0.0024 1 0.0024
2 0.0074 2 0.0042 2 0.0042
5 0.0204 5 0.0088 5 0.0088
10 0.0339 10 0.0155 10 0.0130
20 0.0564 20 0.0272 20 0.0130
50 0.1107 50 0.0570 50 0.0130
100 0.1842 100 0.1000 100 0.0130
200 0.3065 200 0.1000 200 0.0130
500 0.6000 500 0.1000 500 0.0130
1000 0.6000 1000 0.1000 1000 0.0130
1
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Small Plant Gas LPG (5)
Small Plant Liquid (6)
Small Plant Liquid Bund (7)
13
8 - Large Plant Gas LPG 9 - Large Plant Gas LPG 10 - Large Plant Gas LPG 11 - Large Plant Gas LPG
Gas or LPG release from large onshore Liquid release from large onshore plant Gas or LPG release from a large
Liquid release from large onshore plant
plant where spill is bunded confined or congested onshore plant
Releases of flammable gases, vapour or Releases of flammable liquids that do not Releases of flammable liquids that do not Releases of flammable gases, vapour or
liquids significantly above their normal have any significant flash fraction (10% or have any significant flash fraction (10% or liquids significantly above their normal
(NAP) boiling point from large onshore less) if released from large onshore outdoor less) if released from large onshore outdoor (NAP) boiling point from large onshore
outdoor plants (plant area above 1200m2, plants (plant area above 1200m2, site area plants (plant area above 1200m2, site area plants (plant area above 1200m2, site area
site area above 35,000m2) above 35,000m2) and which are not bunded above 35,000m2) and where the liquid above 35,000m2), where the plant is
or otherwise contained. releases from the plant area are suitably partially walled/ roofed or within a shelter
bunded or otherwise contained. or very congested.
0.1 0.0011 0.1 0.0011 0.1 0.0011 0.1 0.0011
0.2 0.0014 0.2 0.0014 0.2 0.0014 0.2 0.0014
0.5 0.0020 0.5 0.0020 0.5 0.0019 0.5 0.0020
1 0.0025 1 0.0025 1 0.0024 1 0.0025
2 0.0050 2 0.0045 2 0.0042 2 0.0058
5 0.0125 5 0.0097 5 0.0088 5 0.0176
10 0.0250 10 0.0173 10 0.0155 10 0.0407
20 0.0500 20 0.0310 20 0.0272 20 0.0942
50 0.1250 50 0.0670 50 0.0500 50 0.2860
100 0.2500 100 0.1200 100 0.0500 100 0.4815
200 0.5000 200 0.1300 200 0.0500 200 0.6000
500 0.6500 500 0.1300 500 0.0500 500 0.7000
1000 0.6500 1000 0.1300 1000 0.0500 1000 0.7000
1
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Large Plant Gas LPG (8) Large Plant Liquid (9)
Large Plant Liquid Bund (10) Large Plant Confined Gas LPG (11)
14
Data Sheet 4: Scenarios 12, 13 & 30
12 - Tank Liquid 300x 300m Bund 13 - Tank Liquid 300x 300m Bund 30 - Tank Liquid 300x 300m Bund
Liquid release from onshore tank farm Liquid release from onshore tank farm
Liquid release from onshore tank farm
where spill is limited by a small or of liquids below their flash point, e.g.
where spill is limited by a large bund
medium sized bund diesel or fuel oil
Releases of flammable liquids that do not Releases of flammable liquids that do not Releases of combustible liquids stored at
have any significant flash fraction (10% or have any significant flash fraction (10% or ambient pressure and at temperatures
less) if released from very large onshore less) if released from onshore outdoor below their flashpoint (e.g. most gas oil,
outdoor storage area 'tank farm' (e.g. spill storage area 'tank farm' (e.g. spill in a large diesel, fuel oil storage tanks) from onshore
in a large multi-tank bund over 25,000m2 tank bund containing 4 or fewer tanks, or outdoor storage area 'tank farm'. This look-
area). See '30 - Tank Liquid - diesel, fuel oil' any other bund less than 25,000m2 area). up correlation can be applied to releases
if liquids are stored at ambient conditions See '30 - Tank Liquid - diesel, fuel oil ' if from tanks and low-pressure transfer lines
below their flashpoint. liquids are stored at ambient conditions or pumps in the tank farm/ storage area.
below their flashpoint. However it should not be used for high-
pressure systems (over a few barg): in these
situations use look-up '12 - Tank Liquid 300
m x 300 m Bund' or '13 - Tank Liquid 100 m
0.1 0.0011 0.1 0.0011 0.1 0.0010
0.2 0.0011 0.2 0.0011 0.2 0.0010
0.5 0.0012 0.5 0.0012 0.5 0.0010
1 0.0013 1 0.0013 1 0.0010
2 0.0016 2 0.0016 2 0.0011
5 0.0024 5 0.0024 5 0.0011
10 0.0037 10 0.0037 10 0.0014
20 0.0068 20 0.0068 20 0.0021
50 0.0153 50 0.0150 50 0.0024
100 0.0281 100 0.0150 100 0.0024
200 0.0517 200 0.0150 200 0.0024
500 0.1160 500 0.0150 500 0.0024
1000 0.1200 1000 0.0150 1000 0.0024
1
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Tank Liquid 300x300m Bund (12)
Tank Liquid 100x100m Bund (13)
Tank Liquid - diesel, fuel oil (30)
15
14 - Tank Gas LPG Storage Plant 15 - Tank Gas LPG Storage Plant 16 - Tank Gas LPG Storage Plant
Gas or LPG release from onshore tank Gas or LPG release from onshore tank
Gas or LPG release from onshore tank
farm sited adjacent to a plant or away farm sited adjacent to a plant or away
farm within the plant
from the plant in an industrial area from the plant in a rural area
Releases of flammable gases, vapour or Releases of flammable gases, vapour or Releases of flammable gases, vapour or
liquids significantly above their normal liquids significantly above their normal liquids significantly above their normal
(NAP) boiling point from onshore outdoor (NAP) boiling point from onshore outdoor (NAP) boiling point from onshore outdoor
storage tanks located in a 'tank farm' storage tanks located in a 'tank farm' storage tanks located in a 'tank farm'
entirely surrounded by plant(s). For tank adjacent to plants or situated away from adjacent to plants or situated away from
farms adjacent to plants use the following plants in an industrial or urban area plants in a rural area
'storage industrial or rural' look-up
correlations. Releases from process vessels
or tanks inside plant areas should be
treated as plant releases.
0.1 0.0011 0.1 0.0011 0.1 0.0011
0.2 0.0011 0.2 0.0011 0.2 0.0011
0.5 0.0011 0.5 0.0011 0.5 0.0011
1 0.0012 1 0.0012 1 0.0012
2 0.0032 2 0.0026 2 0.0025
5 0.0121 5 0.0073 5 0.0071
10 0.0334 10 0.0162 10 0.0154
20 0.0917 20 0.0359 20 0.0204
50 0.3492 50 0.1026 50 0.0296
100 0.9600 100 0.2270 100 0.0392
200 1.0000 200 0.3555 200 0.0520
500 1.0000 500 0.6434 500 0.0755
1000 1.0000 1000 1.0000 1000 0.1000
1
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Tank Gas LPG Storage Plant (14)
Tank Gas LPG Storage Industrial (15)
Tank Gas LPG Storage Rural (16)
16
Data Sheet 6: Scenarios 17 & 18
17 - Offshore Process Liquid 18 - Offshore Process Liquid

Liquid release from offshore process Liquid release from offshore process
module area on NUI
Releases of flammable liquids that do not Releases of flammable liquids that do not
have any significant flash fraction (10% or have any significant flash fraction (10% or
less) if released from within offshore less) if released from within offshore
process modules process modules or decks on NUIs
Release Rate (kg/s) Ignition Probability Release Rate (kg/s) Ignition Probability
0.1 0.0011 0.1 0.0011
0.2 0.0015 0.2 0.0015
0.5 0.0021 0.5 0.0021
1 0.0028 1 0.0028
2 0.0037 2 0.0037
5 0.0053 5 0.0053
10 0.0070 10 0.0070
20 0.0092 20 0.0092
50 0.0133 50 0.0100
100 0.0175 100 0.0100
200 0.0175 200 0.0100
500 0.0175 500 0.0100
1000 0.0175 1000 0.0100
1
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Offshore Process Liquid (17) Offshore Process Liquid NUI (18)
17
19 - Offshore Process Gas Opendeck 20 - Offshore Process Gas Opendeck 21 - Offshore Process Gas Opendeck 22 - Offshore Process Gas Opendeck
NUI NUI NUI NUI
Gas release from a mechanically
Gas release from offshore process Gas release from typical offshore Gas release from large offshore
ventilated or very congested offshore
open deck area on NUI process module process module
process module
Releases of flammable gases, vapour or Releases of flammable gases, vapour or Releases of flammable gases, vapour or Releases of flammable gases, vapour or
liquids significantly above their normal liquids significantly above their normal liquids significantly above their normal liquids significantly above their normal
(NAP) boiling point from an offshore (NAP) boiling point from within offshore (NAP) boiling point from within large (NAP) boiling point from within offshore
process weather deck/ open deck on NUIs. process modules or decks on integrated offshore process modules or decks on process modules or decks on integrated
Can also be used for open/ uncongested deck / conventional installations. (Process integrated deck / conventional installations deck / conventional installations: applies
weather decks with limited process modules include separation, compression, (module greater than 1000m2 floor area). where the module is enclosed and has a
equipment on larger attended integrated pumps, condensate handling, power (Process modules include separation, mechanical ventilation system, or is very
platforms. generation, etc.). If the module is compression, pumps, condensate handling, congested (volume blockage ratio => 0.14
mechanically ventilated or very congested – power generation, etc.). If the module is and less than 25% of area of the end walls
see curve 22 'Offshore Process Gas mechanically ventilated or very congested – open for natural ventilation)
Congested or Mechanical Vented Module'. If see curve 22 'Offshore Process Gas
module or deck is very large, greater than Congested or Mechanical Vented Module'.
1000m2 floor area, see below.
0.1 0.0011 0.1 0.0011 0.1 0.0011 0.1 0.0011
0.2 0.0011 0.2 0.0019 0.2 0.0020 0.2 0.0024
0.5 0.0012 0.5 0.0038 0.5 0.0043 0.5 0.0068
1 0.0012 1 0.0065 1 0.0077 1 0.0150
2 0.0022 2 0.0110 2 0.0138 2 0.0174
5 0.0050 5 0.0183 5 0.0300 5 0.0213
10 0.0091 10 0.0240 10 0.0366 10 0.0247
20 0.0168 20 0.0315 20 0.0445 20 0.0287
50 0.0250 50 0.0400 50 0.0500 50 0.0350
100 0.0250 100 0.0400 100 0.0500 100 0.0400
200 0.0250 200 0.0400 200 0.0500 200 0.0400
500 0.0250 500 0.0400 500 0.0500 500 0.0400
1000 0.0250 1000 0.0400 1000 0.0500 1000 0.0400
1
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Offshore Process Gas Opendeck NUI (19)
Offshore Process Gas Typical (20)
Offshore Process Gas Large Module (21)
Offshore Process Gas Congested_Mech Vented Module (22)
18
24 - Offshore FPSO Gas 25 - Offshore FPSO Gas 26 - Offshore FPSO Gas

Gas release from offshore FPSO that
Gas release from typical offshore FPSO Liquid release from typical offshore
‘weather-vanes’, process module
process module that ‘weather-vanes’ FPSO process module
behind a transverse solid wall
Releas es of flammable gas es , vapour or Releas es of flammable gas es , vapour or Releas es of flammable liquids that do not
liquids s ignificantly above their normal liquids s ignificantly above their normal have any s ignificant flas h fraction (10% or
(NAP) boiling point from within offs hore (NAP) boiling point from within offs hore les s ) if releas ed from within offs hore
proces s modules or decks on FPSOs . This proces s modules or decks on FPSOs . This proces s modules or decks on FPSOs or
correlation was specifically developed for correlation applies if the release is from an FPUs .
weather-vaning FPSOs where any releas e area downwind of a transverse wall across
would be directed along the deck of the the FPSO deck. This correlation was
FPSO. s pecifically developed for s ituations on
weather-vaning FPSOs where equipment
immediately behind the wall, etc. may be in
the permanent ‘wind s hadow’ of a blas t/ fire
wall or other object, and hence experience
0.1 0.0011 0.1 0.0011 0.1 0.0011
0.2 0.0012 0.2 0.0016 0.2 0.0015
0.5 0.0012 0.5 0.0038 0.5 0.0023
1 0.0013 1 0.0088 1 0.0032
2 0.0030 2 0.0207 2 0.0045
5 0.0092 5 0.0639 5 0.0069
10 0.0213 10 0.1500 10 0.0095
20 0.0493 20 0.1500 20 0.0132
50 0.1500 50 0.1500 50 0.0202
100 0.1500 100 0.1500 100 0.0280
200 0.1500 200 0.1500 200 0.0280
500 0.1500 500 0.1500 500 0.0280
1000 0.1500 1000 0.1500 1000 0.0280
1
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Offshore FPSO Gas (24)
Offshore FPSO Gas Wall (25)
Offshore FPSO Liquid (26)
19
Data Sheet 9: Scenarios 23 & 27
23 - Offshore Riser 27 - Offshore Riser

Gas release from typical offshore riser Major release which can engulf an
in air gap entire offshore installation
Releases from offshore installation risers in Releases from drilling or well working
the air gap area where there is little blowouts or riser failures where the release
likelihood of the release entering process could engulf the entire installation and
areas on the installation (e.g. solid decks, reach in to platform areas: applies to
wind walls). Applies to partial flashing oil partial flashing oil or gas releases. (See
or gas releases. May also be used for also 'Offshore Riser' for riser releases and
blowouts with well-positioned diverters blowouts with diverters).
directing any release away from the
installation (see also Offshore Engulf –
blowout - riser).
Release Rate (kg/s) Ignition Probability Release Rate (kg/s) Ignition Probability
0.1 0.0011 0.1 0.0011
0.2 0.0016 0.2 0.0017
0.5 0.0026 0.5 0.0031
1 0.0037 1 0.0049
2 0.0053 2 0.0078
5 0.0086 5 0.0141
10 0.0124 10 0.0222
20 0.0178 20 0.0350
50 0.0250 50 0.0636
100 0.0250 100 0.1000
200 0.0250 200 0.1000
500 0.0250 500 0.1000
1000 0.0250 1000 0.1000
1
0.1
0.01
0.001
0.1 1 10 100 1000 10000
Release Rate (kg/s)
Offshore Riser (23) Offshore Engulf_blowout_riser (27)
20
Notes:
1) A flammable substance stored above its auto-ignition temperature is likely to ignite on
release and should be modelled as having an overall (total) ignition probability of one. A
suitable split between immediate and delayed ignition should be considered based on
the nature of the fluid and its release characteristics.
2) Very reactive substances are unlikely to be found in oil and gas processing operations
but if present it is suggested that the values given in the look-up correlations are
doubled, subject to a maximum of 1. Such substances include hydrogen, acetylene,
ethylene oxide and carbon disulphide.
3) High flash point (>55°C) liquids stored at or near atmospheric pressure and
significantly below 55°C are significantly less likely to ignite than suggested in the
look-up correlations. It is suggested that the ignition probability from the look-up
correlations is multiplied by a factor of 0.1, subject to a minimum of 0.001 and taking
account of the 0.001 immediate ignition probability.
Using the correlation for scenario 1, “Pipe Liquid Industrial” as an example, the lower
curve in Figure 2-1 would be appropriate for high flashpoint liquids.
Figure 2-1: Revised Ignition Probability Curve for High Flash Point Liquids
4) For liquids with flash fractions above 10% it is suggested that the ignition probability
is estimated by combining the relevant liquid ignition probability with a suitable gas/
LPG ignition probability. The appropriate release rates should be obtained from the
flash fraction, e.g., a 10 kg/s release with a 20% flash fraction should give rise to an
equivalent 2 kg/s gas release and 8 kg/s liquid release.
The two probabilities can be combined using the following equation:
Pign = 1 - [(1 - Pigngas/LPG )(1 - Pignliquid )]
Alternatively, the higher of the two ignition probabilities can be used on the basis that
the areas covered by the liquid and gas are likely to have considerable overlap.
Further options for determining intermediate correlations are given in [2]
21
5) Since the correlations are based on typical combinations of ignition sources, it follows
that they should not be used in situations where particularly strong sources such as
fired heaters are present or where there are a larger than normal number of sources.
In this case the full UKOOA ignition model [1] is more appropriate.
2.2 Ignition timing

The time between the start of a hydrocarbon release and the time it ignites may be
important for two reasons:
• Providing the release has been detected, the time until ignition influences the
probability that workers can relocate to a safer area prior to ignition.
• The time has an influence on the size of a gas cloud or liquid pool when it ignites and
therefore the resulting consequences, particularly the explosion overpressure.
The nature of many accidents makes it difficult to obtain reliable information on the time of
ignition, particularly because it may not be known how long a leak has existed prior to detection.
The UKOOA ignition model [1] draws a distinction between “immediate” and “delayed”
ignitions. These two terms are often confused with “early” and “late” ignitions. Within the
UKOOA model an immediate ignition allows for the situation where the cause of the leak
could also cause the leak to ignite. In this situation, it is reasonable to assume that the
consequence will be a jet fire or liquid spray fire. This is not the same as an “early” ignition
which occurs within a short period, e.g., 30 seconds, after the release. “Early” ignitions are
an arbitrary sub-set of delayed ignitions that may not allow time for workers to leave the
area but may still include situations where the gas cloud has developed to a size capable of
generating high overpressures and a subsequent jet fire.
The guidance in the model [1] refers to distributions of 30:70 (early:late) or 50:50 (early:late)
as being reasonable. It is important that this is not interpreted as the ratio of “immediate”
to “delayed”. The model assumes an immediate ignition probability of 0.001 irrespective of
the release size.
The guidance [1] suggests that if a time dependent distribution is required then the
distribution given in for plant and transport scenarios and presented graphically in Figure
2-2 is used. These are based on historical data for large releases.
Table 2-4: Suggested Ignition Timing Distribution Given Ignition Occurs
Relative Cumulative Probability

Time (secs)
Plant Transport
1 0.22 0.53
10 0.29 0.53
30 0.36 0.53
100 0.63 0.6
1000 0.94 0.86
>1000 1.00 1
22
Figure 2-2: Ignition Timing Distribution Given Ignition
For small release rates, the immediate ignition probability of 0.001 may equate to more
than 22% of the total ignition probability.
Where the cumulative probability of ignition derived from the timing distribution is lower
than the probability of immediate ignition (0.001), then it is suggested that this is treated
as immediate ignition. The historical timing distribution table and curve can be used if
a time-dependant distribution is required and for leaks of significant size. However, this
approach does have the disadvantage that all ignitions in the early part of the cumulative
timing distribution will be treated as ‘immediate’ ignition, especially where the total ignition
probability is only slightly higher than the probability of immediate ignition.
Figure 2-2
An alternative approach, which provides a smoother timing distribution, is to use the
probability of immediate ignition (0.001) directly and then apply the cumulative timing
distribution to that part of the total ignition probability greater than the immediate ignition
probability, i.e., the cumulative timing distribution is applied to the delayed ignition
probability contribution.
Either of these approaches are valid and have their pros and cons. The differences between
these methods are small, and are only likely to be noticeable at low release rates/ low
ignition probabilities and so should not have a significant impact on the overall QRA results.
Alternatively, the full UKOOA Ignition Probability Model may be used or, if a more accurate
assessment is required, recourse made to models involving the calculation of cloud build
and ignition source location.
The sparsity of data and the quality with which it is practical to report ignition cases results
in large uncertainties in the probabilistic distribution of the ignition times. In many risk
analysis scenarios it may be appropriate to assume that the time of ignition is sufficiently
small such that it can effectively be assumed to occur at time t=0 provided a suitable
proportion of the modelled consequences are compatible with delayed ignitions.
23
2.3 LNG Releases

The dispersion modelling underpinning the ignition modelling and therefore the look-up
correlations is not valid for LNG releases because it does not take account of the heat
transfer and evaporation effects which can affect both the quantity of gas generated and the
way it disperses. Although the UKOOA model does not represent flashing cryogenic liquid
releases, there are some options in establishing an estimate of ignition probability:
1) Separate calculations could be undertaken to estimate an equivalent gas release and
use this with the look-up correlations.
2) The full model uses a relatively simple dispersion model but this can be bypassed by
inserting the areas covered by the flammable cloud directly into the model where it
combines with the data on ignition source strength relevant to those areas.
3) The UKOOA model could be dispensed with altogether and alternative models used
to calculate the rate of gas evolution, the dispersion of the cloud based on weather
conditions and the land use of the areas covered by the flammable cloud. Ignition
source densities for the land use types are available in [1].
The degree of conservatism in each of these approaches is dependent on the models used
for estimating gas evolution and dispersion.
2.4 Blowout Ignition Probabilities

An alternative to the blowout ignition probabilities given by the UKOOA look-up correlations
can be obtained from Lloyd’s Register’s interpretation of the blowout data provided by
SINTEF [3]. This is given in Table 2-5. These are mostly based on historic incidents.
This table is based on operations of North Sea standard. For the ignition probabilities
relevant to operations not of North Sea standard refer to the current SINTEF report [4] for
guidance. The most significant category is that for drilling and heavy interventions which
has an overall probability of ignition is 30%. In cases where there are no historic ignitions,
alternative values are given based on the MISOF model. More information is available in [3].
Table 2-5: Ignition probabilities for Blowouts and Well Releases on Offshore Installations from [3]
Early ignition Delayed ignition Very Delayed

Release Type Basis
(< 5 min) (5 – 60 min) ignition (> 60 min)
Historic 0 0 0
Producing Well Blowouts
MISOF 0.02 0.03 -
Shallow Gas Blowout Historic 0.07 0.11 0.07
Drilling and Heavy

Historic 0.12 0.01 0.19
Interventions1
Historic 0 0 0.17
Light Interventions2
MISOF 0.05 0.05/0.0053 -
Well Releases (all

Historic 0.04 0 0
operations)
1. Including all operations that will be performed from drill floor: drilling, completion, heavy workover including snubbing.
2. Including interventions that are likely performed from an intervention deck, i.e., wireline and coiled tubing
3. The ignition probability for a delayed ignition is reduced from 5% to 0.5% if a flame arrestor system on diesel engine air intakes
are used.
24
Sufficient data are not available to determine the influence of drilling or well intervention
activities combined with ongoing production SIMOPS on the ignition probabilities. .
Should this be required, a detailed analysis of the gas cloud dispersion and the presence
of potential ignition sources is recommended since this is a situation where the ignition
probability may be higher.
2.5 Onshore Gas Pipeline Ignition Probabilities

A trend has been observed from analysis of historical data for onshore gas transmission
pipeline rupture incidents where the ignition probability (Pign) increases linearly with pd2,
where p = pipeline operating pressure (bar gauge) and d = pipeline diameter (m) [5,6]. The
correlation derived for rupture releases takes the form:
Pign = 0.0555 + 0.0137 pd2; 0 ≤ pd2 ≤ 55
and
Pign = 0.81; pd2 > 55
The various ignition possibilities, together with the release types, should be drawn
out logically on an event tree to obtain overall probabilities. Appropriate values for the
probability of immediate or delayed ignition (and, if delayed, the assumed time(s) of ignition)
should be applied.
For rupture releases, the total probability of ignition (Pign calculated as detailed above)
is generally apportioned as 0.5 for immediate ignition and 0.5 for delayed ignition, where
delayed ignition occurs after 30 seconds (see Section 2.5.1).
Puncture releases use the same form of the ignition probability relationship as for rupture
releases; however, in this instance d = release hole diameter (m) and the coefficient of the
pd2 value is halved, with an upper bound on ignition probability of 0.43. This is based on a
rupture being a double-ended release and therefore with two sources whereas a puncture
has a single release source through the hole in the pipeline.
Pign = 0.0555 + 0.00685 pd2; 0 ≤ pd2 ≤ 55
and
Pign = 0.43; pd2 > 55
It has been observed for punctures that the consequences of immediate and delayed
ignition are essentially the same and, therefore, no distinction is made between immediate
and delayed ignition for puncture releases [6].
2.5.1 Ignition timing for onshore gas transmission pipelines

The time of ignition is particularly important for risk analysis of high pressure gas pipelines,
because of the rapid depressurisation that follows a pipeline rupture and the highly
transient nature of the initial gas release rate. As a result, the corresponding fire following
pipeline rupture is much larger at earlier times and, hence, the consequences are more
severe for people and property in the vicinity of the incident if immediate ignition occurs.
25
To provide information on the time of ignition observed in actual incidents involving onshore
gas transmission pipelines, a simple analysis was performed of the time to ignition for the
limited number of cases where information on the ignition time was recorded [6]. Inevitably,
because of the variable nature of the information recorded on incidents, the time of ignition
is subject to a significant degree of uncertainty. Nevertheless, by analysing the detailed
descriptions of incidents where information was available, it was possible to assign an
estimated time of ignition within certain time bands and the results are presented in Table
2-6 together with the probability of ignition occurring with each band.
Table 2-6: Time to Ignition Analysis for Ignited Rupture Incidents from [5]
Time from failure Number of rupture Probability of ignition Cumulative probability

to ignition, t (s) incidents within timeframe of ignition
0 ≤ t ≤ 30 27 0.64 0.64
30 < t ≤ 60 2 0.05 0.69
60 < t ≤ 120 2 0.05 0.74
t > 120 11 0.26 1.00
Total 42 1.00 -
As shown in Table 2-6, 64% of the incidents were estimated to have ignited within the first
30 seconds. Uncertainty in the time of ignition has generally been addressed by simplifying
the information in Table 2-6 to represented it as two possible ignition times for the risk
calculations: either immediate ignition or ignition after a delay of 30 seconds, each with an
equal likelihood. This appears to be an appropriate representation in the light of the above
data and suitably cautious, bearing in mind that ignition at later times results in lower
consequences according to the risk calculations, due to the rapid reduction in the gas flow
rate.
26
3. Guidance on use of the UKOOA model
3.1 General Validity

The correlations presented for the UKOOA model in Section 2.1 are considered to provide
an acceptable approach for use in typical QRA studies. For more detailed analysis it is
recommended that the full spreadsheet UKOOA ignition model is used so that the specific
circumstances with regard to layout and ignition sources can be more accurately represented.
The correlations were developed for UKOOA (now Oil & Gas UK) member companies with
the intention of providing representative probabilities for installations operating in UK
waters. They may be applied to the analysis of hydrocarbon releases in other regions which
comply with recognised industry good practice, as it is applied in the UKCS.
The foreword to the Energy Institute report [1] states that the model and look-up
correlations “are not suited to the ignition probability assessment of refrigerated liquefied
gases, vapourising liquid pools, sub-sonic gas releases, or non-momentum driven
releases, such as those following catastrophic storage vessel failure.” This is because the
dispersion models built into the overall method do not take account of the thermal effects
which can dominate the source term.
Despite this note, flashing liquid releases are covered by a number of the correlations and
analysts may further modify them by combining them with a gas or LPG ignition probability
in suitable proportions as suggested in note 4 of section 2.1. Atmospheric storage tanks are
dealt with in the Storage Incident Frequencies data sheet. Low momentum and sub-sonic
gas releases are uncommon in process systems.
3.2 Alternative Approaches

The initial task for the analyst is to determine which of the scenarios given in Table 2-1
to Table 2-2 and Table 2-3 best matches the scenario under consideration. There may
be situations where the scenario under consideration lies between two of the described
scenarios, in which case the analysts may attempt to interpolate between two curves.
The data presented in the tables in Section 2.1 can be used in three ways:
1) Estimate from the graphs
2) Obtain probability based on the tabulated values
3) Use values in Table 3-1 to calculate the probability. Note that, in interpolating between
the data points, it is necessary to take logarithms of the release rate and probabilities,
interpolate between these to find the logarithm of the required probability and then
obtain the value itself, i.e.,:
log Pign = logPignlower + (logQ – logQlower) (logPignupper – logPignlower )

(logQupper – logQlower)
27
where Pign is the required ignition probability corresponding to release rate Q
Pignlower is the ignition probability at a release rate of Qlower (the lower bound of the
relevant curve section), and
Pignupper is the ignition probability at a release rate of Qupper (the upper bound of the
relevant curve section)
The third of these options is the recommended approach and the analyst may find it
convenient to construct a spreadsheet or some other computer programme to carry this out.
The data used to generate the lines on the graphs in the datasheets (Section 2.1) are shown
in Table 3-1. This is as reported in [2] and has been derived from Table 2.9 in the Institute of
Energy report [1] which provides further explanation on the derivation of the correlations. This
specifies the release rates and ignition probabilities relating to each of the points bounding
the segments as indicated in Figure 3-1. This includes some detail of the curves below 0.1
kg/s which is the lower bound of the release rate axis in the graphs presented above.
Ignition Probability
Release Rate
Figure 3-1 Typical Ignition Probability Curve
A further approach is to use the equation of the form Pign = aQb which applies to the release
range. These are presented in Table 3-2.
A more accurate assessment may be obtained using the full UKOOA ignition model which
is described in [1]. This has been implemented in a spreadsheet tool which is made
available on a CD which
Figure 3-1 accompanies the report. This allows the user to input specific data
relating to release conditions, platform layout and ignition sources. However, this requires
more effort on the part of the analyst and the availability of more installation specific data
compared with the relative ease with which the look-up functions can be used.
28
Table 3-1: Data for Look-up Correlations
Point 1 Point 2 Point 3 Point 4 Point 5 Point 6 Point 7

Case
Case Description
No.
Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign
1 Pipe Liquid Industrial 0.01 0.00100 0.03493 0.00100 0.100 0.00180 70.000 0.07000 100000 0.07000
2 Pipe Liquid Rural 0.01 0.00100 0.03787 0.00100 0.100 0.00180 0.300 0.00350 70.000 0.00700 70.000 0.00700 100000 0.00700
3 Pipe Gas LPG Industrial 0.01 0.00100 0.08791 0.00100 0.100 0.00110 1000 1.00000 100000 1.00000
4 Pipe Gas LPG Rural 0.01 0.00100 0.04799 0.00100 0.100 0.00110 10.000 0.00200 1000.000 0.08000 23408.5 1.00000 100000 1.00000
5 Small Plant Gas LPG 0.01 0.00100 0.07654 0.00100 0.100 0.00110 1.000 0.00250 3.000 0.01400 498.991 0.60000 100000 0.60000
6 Small Plant Liquid 0.01 0.00100 0.07548 0.00100 0.100 0.00110 1.000 0.00240 100.000 0.10000 100000 0.10000
7 Small Plant Liquid Bund 0.01 0.00100 0.07548 0.00100 0.100 0.00110 1.000 0.00240 8.053 0.01300 100.000 0.01300 100000 0.01300
8 Large Plant Gas LPG 0.01 0.00100 0.07654 0.00100 0.100 0.00110 1.000 0.00250 100.000 0.25000 260.000 0.65000 100000 0.65000
9 Large Plant Liquid 0.01 0.00100 0.07654 0.00100 0.100 0.00110 1.000 0.00250 100.000 0.12000 109.990 0.13000 100000 0.13000
10 Large Plant Liquid Bund 0.01 0.00100 0.07548 0.00100 0.100 0.00110 1.000 0.00240 42.492 0.05000 100.000 0.05000 100000 0.05000
11 Large Plant Confined Gas LPG 0.01 0.00100 0.07654 0.00100 0.100 0.00110 1.000 0.00250 70.000 0.43000 325.028 0.70000 100000 0.70000
12 Tank Liquid 300x300m Bund 0.01 0.00100 0.05250 0.00100 0.100 0.00105 1.000 0.00125 7.000 0.00270 519.617 0.12000 100000 0.12000
13 Tank Liquid 100x100m Bund 0.01 0.00100 0.05250 0.00100 0.100 0.00105 1.000 0.00125 7.000 0.00270 49.035 0.01500 100000 0.01500
14 Tank Gas LPG Storage Plant 0.01 0.00104 0.00160 0.00100 0.100 0.00110 1.000 0.00116 100.000 0.96000 102.838 1.00000 100000 1.00000
Tank Gas LPG Storage

15 0.01 0.00104 0.00160 0.00100 0.100 0.00110 1.000 0.00116 100.000 0.22700 988.106 1.00000 100000 1.00000
Industrial
16 Tank Gas LPG Storage Rural 0.01 0.00104 0.00160 0.00100 0.100 0.00110 1.000 0.00116 10.000 0.01540 52551.5 0.50000 100000 0.50000
17 Offshore Process Liquid 0.01 0.00100 0.07882 0.00100 0.100 0.00110 100.000 0.01750 100000 0.01750
Point 1 Point 2 Point 3 Point 4 Point 5 Point 6 Point 7

Case
Case Description
No.
Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign Q (kg/s) Pign
18 Offshore Process Liquid NUI 0.01 0.00100 0.07882 0.00100 0.100 0.00110 24.731 0.01000 100.000 0.01000 100000 0.01000
Offshore Process Gas

19 0.01 0.00101 0.00803 0.00100 0.100 0.00110 1.000 0.00120 30.000 0.02400 31.423 0.02500 100000 0.02500
Opendeck NUI
20 Offshore Process Gas Typical 0.01 0.00100 0.08833 0.00100 0.100 0.00110 3.000 0.01500 10.000 0.02400 37.008 0.04000 100000 0.04000
Offshore Process Gas Large

21 0.01 0.00100 0.08933 0.00100 0.100 0.00110 5.000 0.03000 30.000 0.05000 100000 0.05000
Module
Offshore Process Gas

22 Congested Mech Vented 0.01 0.00100 0.09194 0.00100 0.100 0.00110 1.000 0.01500 50.000 0.03500 92.624 0.04000 100000 0.04000
Module
23 Offshore Riser 0.01 0.00100 0.08340 0.00100 0.100 0.00110 30.000 0.02200 38.267 0.02500 100000 0.02500
24 Offshore FPSO Gas 0.01 0.00100 0.02688 0.00100 0.100 0.00110 1.000 0.00130 50.000 0.15000 100000 0.15000
25 Offshore FPSO Gas Wall 0.01 0.00100 0.08393 0.00100 0.100 0.00110 0.300 0.00200 10.000 0.15000 100000 0.15000
26 Offshore FPSO Liquid 0.01 0.00100 0.08160 0.00100 0.100 0.00110 100.000 0.02800 100000 0.02800
Offshore Engulf_blowout_
27 0.01 0.00100 0.08642 0.00100 0.100 0.00110 100.000 0.10000 100000 0.10000
riser
28 Cox, Lees, Ang - Gas 0.01 0.00081 0.50000 0.01000 100.000 0.30000 100000 0.30000
29 Cox, Lees, Ang - Liquid 0.01 0.00215 0.50000 0.01000 100.000 0.08000 100000 0.08000
30 Tank Liquid - diesel, fuel oil 0.01 0.00100 0.10000 0.00100 1.000 0.00103 7.000 0.00117 25.551 0.00240 100000 0.00240
Table 3-2: Explicit Formulae for Release Rate Ranges
Release Release
Case Case Case Case
Rate Range Equation Rate Range Equation
No. Description No. Description
(kg/s) (kg/s)
1 Pipe Liquid 0.1 - 69.9 Pign = 0.00652 Q0.559 12 Tank Liquid 0.1 - 1 Pign = 0.00125 Q0.076
Industrial 300x300m
> 70 Pign = 0.070 Bund 1-7 Pign = 0.00125 Q0.396
2 Pipe Liquid 0.1 - 0.3 Pign = 0.00725 Q0.605 7 - 519 Pign = 0.00049 Q0.881
Rural
0.3 - 70.3 Pign = 0.00408 Q0.127 > 519 Pign = 0.120
> 70.3 Pign = 0.007 13 Tank Liquid 0.1 - 1 Pign = 0.00125 Q0.076
100x100m
3 Pipe Gas LPG 0.1 - 997 Pign = 0.00604 Q0.740 Bund 1-7 Pign = 0.00125 Q0.396
Industrial
> 1000 Pign = 1.000 7 - 49 Pign = 0.00049 Q0.881
4 Pipe Gas LPG 0.1 - 10 Pign = 0.00148 Q0.130 > 49 Pign = 0.015
Rural
10 - 23417 Pign = 0.00032 Q0.801 14 Tank Gas LPG 0.1 - 1 Pign = 0.00116 Q0.023
Storage Plant
> 23417 Pign = 1.000 1 - 103 Pign = 0.00116 Q1.459
5 Small Plant 0.1 - 1 Pign = 0.00250 Q0.357 > 103 Pign = 1.000
Gas LPG
1-3 Pign = 0.00250 Q1.568 15 Tank Gas 0.1 - 1 Pign = 0.00116 Q0.023
LPG Storage
3 - 498 Pign = 0.00624 Q0.735 Industrial 1 - 100 Pign = 0.00116 Q1.146
> 498 Pign = 0.600 100 - 992 Pign = 0.01152 Q0.647
6 Small Plant 0.1 - 1 Pign = 0.00240 Q0.339 > 992 Pign = 1.000
Liquid
1 - 99.9 Pign = 0.00240 Q0.810 16 Tank Gas LPG 0.1 - 1 Pign = 0.00116 Q0.023
Storage Rural
> 99.9 Pign = 0.100 1 - 10 Pign = 0.00116 Q1.123
7 Small Plant 0.1 - 1 Pign = 0.00240 Q0.339 10 - 52890 Pign = 0.00604 Q0.406
Liquid Bund
1 - 8.05 Pign = 0.00240 Q0.810 > 52890 Pign = 0.500
> 8.05 Pign = 0.013 17 Offshore 0.1 - 99.5 Pign = 0.00277 Q0.401
Process Liquid
8 Large Plant 0.1 - 1 Pign = 0.00250 Q0.357 > 100 Pign = 0.018
Gas LPG
1 - 260 Pign = 0.00250 Q 1.000
18 Offshore 0.1 - 24.6 Pign = 0.00277 Q0.401
Process Liquid
> 260 Pign = 0.650 NUI > 100 Pign = 0.010
9 Large Plant 0.1 - 1 Pign = 0.00250 Q0.357 19 Offshore 0.1 - 1 Pign = 0.00120 Q0.038
Liquid Process Gas
1 - 110 Pign = 0.00250 Q 0.841
1 - 31.4 Pign = 0.00120 Q0.881
Opendeck NUI
> 110 Pign = 0.130 > 31.4 Pign = 0.025
10 Large Plant 0.1 - 1 Pign = 0.00240 Q 0.339
20 Offshore 0.1 - 3 Pign = 0.00645 Q0.768
Liquid Bund Process Gas
1 - 42.5 Pign = 0.00240 Q 0.810
3 - 37.1 Pign = 0.00977 Q0.390
Typical
> 42.5 Pign = 0.050 > 37.1 Pign = 0.040
11 Large Plant 0.1 - 1 Pign = 0.00250 Q 0.357
21 Offshore 0.1 - 5 Pign = 0.00770 Q0.845
Confined Gas Process Gas
LPG 1 - 70 Pign = 0.00250 Q 1.212
5 - 30 Pign = 0.01896 Q0.285
Large Module
70 - 327 Pign = 0.11166 Q0.317 > 30 Pign = 0.050
> 327 Pign = 0.700
31
Release Release
Case Case Case Case
Rate Range Equation Rate Range Equation
No. Description No. Description
(kg/s) (kg/s)
22 Offshore 0.1 - 1 Pign = 0.01500 Q1.135 26 Offshore FPSO 0.1 - 99.6 Pign = 0.00324 Q0.469
Process Gas Liquid
Congested_ 1 - 91.8 Pign = 0.01500 Q0.217 > 100 Pign = 0.028
Mech Vented
> 91.8 Pign = 0.040 27 Offshore 0.1 - 99.9 Pign = 0.00495 Q0.653
Module
Engulf_
blowout_riser > 100 Pign = 0.100
23 Offshore Riser 0.1 - 38.3 Pign = 0.00369 Q0.525
> 30 Pign = 0.025 28 Cox, Lees, Ang 0.5 - 100 Pign = 0.01560 Q0.642
- Gas
0.1 - 1 Pign = 0.00130 Q0.073 > 100 Pign = 0.300
24 Offshore FPSO 1 - 50 Pign = 0.00130 Q1.214 29 Cox, Lees, Ang 0.5 - 101 Pign = 0.01313 Q0.392
Gas - Liquid
> 50 Pign = 0.150 > 100 Pign = 0.080
25 Offshore FPSO 0.1 - 0.3 Pign = 0.00385 Q0.544 30 Tank Liquid - 0.1 - 1 Pign = 0.00103 Q0.011
Gas Wall diesel, fuel oil
0.3 - 10 Pign = 0.00881 Q1.231 1-7 Pign = 0.00103 Q0.068
> 10 Pign = 0.150 7 - 25.5 Pign = 0.00040 Q0.555
> 25.5 Pign = 0.002
3.3 Uncertainties
The assessment of ignition probability is subject to a large degree of uncertainty. The
spreadsheet model produced under phase I of the joint industry project for the full
UKOOA model is itself subject to uncertainties in the analytical approach taken and in
the data used. The adoption of the lookup correlations based on the UKOOA model [1]
introduces more uncertainties because a compromise has to be made in selecting the most
appropriate curve and these curves themselves are approximations to the curves produced
by the “full” UKOOA model.
Ignition probabilities are influenced by design layout, the number and separation of ignition
sources, the quality of maintenance of equipment, and thereby the control of ignition
sources.
Despite these uncertainties, the approach is considered to be an advance on previous

formulations which relate ignition probability to release rate only, and with no regard for the
location and strength of ignition sources, the nature of the fluids or the layout of the plant.
32
4. Review of primary data source
The data presented in Section 2 is largely a reproduction of data from the Energy Institute
report [1], published on behalf of the joint industry project sponsors UKOOA (Now Oil and
Gas UK), the HSE and the Energy Institute. The report reviews existing models and develops
a new model which could be applied to both onshore and offshore scenarios. The work was
undertaken in two phases.
The first phase saw the development of a model for assigning ignition probabilities in QRA
studies to further the understanding of scenario specific ignition probabilities. The work
was undertaken by AEA Technology (now ESR) and co-ordinated by a joint industry steering
group drawn from UKOOA member representatives, the HSE and consultants working in
the field of onshore and offshore QRA.
The report [1] summarised the current status of knowledge and research in the field
of ignition probability estimation in support of QRA. It evaluated this, together with the
usefulness of the UK HSE’s hydrocarbon release database as a basis to develop an
improved ignition model for use in QRA. The end result was a spreadsheet model for
estimating the ignition probability of process leaks offshore and most typical onshore
hydrocarbon leak scenarios. The spreadsheet models the ignition probability by considering
the size of the gas cloud which would be formed by the release and taking into account the
number and type of ignition sources reached by the cloud, at flammable concentration. The
complexity of the model is such that users are required to obtain and enter a significant
amount of data relating to the platform configuration and the distribution of ignition
sources.
The second phase saw the development of look-up correlations for representative
scenarios which could be used in QRA studies without the need for the user to gather the
data required for the full model. The release types considered included:
• Gas releases
• LPG (flashing liquefied gas) releases
• Pressurised liquid oil releases – leading to a spray release with flashing/ evaporation/
aerosol formation
• Low pressure liquid oil releases – leading to a spreading pool only (no aerosol
formation or flashing)
• Release rates from 0.1 to 1000 kg/s – (graphs shown in the data sheets are extended
to 10000 kg/s where the probability function does not reach a maximum below 1000
kg/s)
A large number of analyses were carried out to produce graphs of ignition probability
against release rate. Figure 4-1 shows a typical set of curves. Similar curves were grouped
into the scenarios listed in Table 2-1 to Table 2-3 and a representative curve assigned to
them. These curves consist of a number of segments each of which appears as a straight
line when plotted on logarithmic axes. It is these curves which are depicted in the data
sheets.
33
Figure 4-1: Example of Ignition Probability Curve Calculated by UKOOA ignition model
Prior to the introduction of the UKOOA ignition model approach outlined above, the
formulation attributed to Cox, Lees and Ang [7] was widely used. This gained acceptance
largely because of the proportion of analysts using it rather than because of the rigour of
the theory underlying it. Ignition probabilities predicted by this method were in excess of
what was found to occur in practice and this was partly responsible for instigating the work
which resulted in the UKOOA ignition model.
References in this report to “UKOOA (spreadsheet) model” and “UKOOA look-up

correlations” relate respectively to the output from the two phases of the project [1].
34
5. Alternative ignition models
5.1 Ignition modelling of process leaks in Norwegian offshore O&G

industry – MISOF, OLF & TDIIM
Three models have generally been used within the Norwegian offshore oil and gas
industry for estimation of ignition probability upon hydrocarbon process leaks. They can be
summarised as:
• MISOF ignition model [8]
• OLF ignition model [9]
• TDIIM/JIP ignition model [11]3
The mathematical and physical framework of the models is quite similar for these three
models and they can be regarded as steps in an evolution process where the TDIIM/JIP
ignition model established the initial basis and the MISOF model represents the most
updated version which is likely to become the most widely used for studies in in Norway.
While these are part of the same evolution, the MISOF model uses updated statistical
material giving it a fundamentally different basis for the MISOF model opposed to the
previous models. Furthermore, the MISOF model is intended to reflect new knowledge
on the properties of the potential ignition sources as well as the behaviour of the ignition
control barrier at offshore installations. For instance, it deals with the isolation of
equipment upon detection in a more appropriate way. The MISOF model is aligned with
the PLOFAM leak frequency model [11], together providing the best estimate of the fire
and explosion frequency for an offshore installation at the Norwegian Continental Shelf.
However, MISOF can also be used in conjunction with alternative leak frequency models.
The TDIIM/JIP model has been used since the late 1990s. The OLF model was proposed in
2007 and has been used regularly by most stakeholders. MISOF was proposed in 2016 [8]
with its ignition intensity values being updated in 2018. This section will refer to all three
models but provide more information on the TDIIM/JIP and MISOF models.
The main principles and key parameters of the models can be summarised as follows:
• The three models are intended to predict the ignition probability following a
hydrocarbon leak from the process system in an offshore oil and gas production and/
or processing facility. They calculate the ignition probability of a flammable cloud of
a given size or area to which various ignition sources are exposed for a given period.
The size and duration of the flammable gas cloud needs to be modelled separately as
input to the models.
3 Only ignition in the source module is described here. The JIP report on ignition modelling also contain a framework for dispersion
modelling in order to predict probability of gas exposure to ignition sources, as well as model for predicting ignition probability due
to external ignition. sources. However, in the presentation of ignition modelling included here, it is focused on updated description of
internal ignition modelling as described in ref(DNV report no 99-3193 / Scandpower report no 27.29.03) rather than the full JIP scope
described in ref(DNV report no. 96-3629)
35
• The three models support transient (time dependent) modelling of ignition probability.
The ignition probability contribution upon a leak is calculated in time steps in order
to predict the accumulated ignition probability at any given time following release of a
hazardous substance.
• Immediate ignition: This is an ignition that occurs simultaneously with the initial
release and is caused by mechanisms that are typically related to the cause of the
leak. The mechanism may be sparks generated from a rupture, flow generated
electric sparks or other causes such as external impact causing simultaneous
leak and ignition. This is consistent with the UKOOA model. The consequence upon
immediate ignition is commonly modelled as fire without resulting in significant
explosion overpressure. The three models propose ignition probabilities upon
initiation of the leak as follows:
–– In the MISOF model the immediate ignition probability is independent of leak
size and the phase of the HC fluid. A higher probability of immediate ignition is
assigned for leaks from pumps than for other equipment types.
–– In the OLF and TDIIM model the immediate ignition probability is assumed to
be dependent on release rate but is independent of the equipment on which the
leak occurs as well as the phase of the HC fluid that is released.
• Delayed ignition: All ignited leaks that are not ignited immediately are defined as
delayed ignitions. During the formation of the flammable gas cloud within a process
module (given that the leak has not ignited immediately) the ignition intensity is (with
some exceptions) assumed to be uniformly distributed throughout the module rather
than assigning ignition probability to point sources that may cause ignition upon
exposure to flammable fluid. Two types of ignition source mechanisms are contributing
to the delayed ignition within the module: continuous and discrete sources.
• Continuous ignition: When a continuous ignition source is first exposed to a mixture
above LFL there is an assigned probability that exposure to the ignition source
causes ignition, but prolonged ignition of the specific source does not increase
this probability. A simplified way of visualising a continuous ignition source is, for
example, the probability of faulty isolation of an object that will cause ignition if
exposed. Ignition intensity will then be similar to the probability of faulty isolation.
Since the ignition intensity is assumed uniform in the module, the ignition
contribution is proportional to flammable cloud size (as long as ignition source
isolation is not accounted for as discussed below).
–– In the MISOF model the ignition intensity from continuous sources in a
classified area such as a process or wellhead area is 6.3 x 10-6 per m3
exposed to flammable fluid. This is made up of contributions from sources of
different types. If the number of rotating machinery items such as pumps and
compressors are known, the generic contribution per m3 of flammable gas
cloud from rotating machinery (3.7 x 10-6) may be replaced by a conditional
ignition probability of 3.7 x 10-3 per rotating item exposed to flammable fluid.
Furthermore, specific failure rates (per hr) are provided for the various types of
electrical equipment in terms of EX protection.
–– In the TDIIM model the continuous ignition probability per item of equipment
exposed to flammable fluid for pumps and compressors. The continuous ignition
intensity for other sources are expressed per m2 exposed to the flammable fluid.
36
• Discrete ignition: When a discrete ignition source is exposed to a mixture above LFL
the ignition probability per unit time is constant for as long as the ignition source is
exposed (when ignition source control is not accounted for). A discrete ignition source
is effective at distinct points in time, and can be visualised as an object generating
sparks at random intervals (or a source of ignition being introduced at the scene
of the leak after start of the leak). Ignition intensity will then be proportional to
frequency of sparks:
–– In the MISOF model the ignition intensity from discrete sources is 1.5 x 10-8
per m3 per second exposed to flammable fluid. Again, this is made up of
contributions from sources of different types. The generic contribution per m3
per second from rotating machinery (1.5 x 10-9 /m3/sec) may be replaced by
a conditional ignition probability of 1.6 x 10-6 per unit rotating machinery per
second exposure of flammable fluid. For electrical equipment, failure rates (per
hr) dependent on EX protection is given.
–– In the TDIIM model a conditional discrete ignition probability per item per
second exposed to flammable fluid is given for pumps for compressors. The
discrete ignition intensity for other sources are expressed per m2 per second
exposed to the flammable fluid.
The probabilities assigned to the ignition sources for the MISOF and TDIIM models are
given in Table 5-1 and Table 5-2.
Table 5-1: Ignition Intensity Values Used in the MISOF Model
Immediate Ignition
Pumps - All Release Rates 0.072
All Other leaks – All release rates 0.0007
Generic - All releases if pumps not modelled explicitly 0.0023
Delayed Ignition
Item/Area Continuous Discrete
Rotating Machinery (general)* 3.7 x 10-6 /m3 1.5 x 10-9 /m3/sec
Rotating Machinery (per item exposed)* 3.7 x 10-3 1.6 x 10-6 /sec
Electrical Components (general)** 1.8 x 10-6 /m3 1.5 x 10-9 /m3/sec
Other Sources (general) 6.0 x 10-7 /m3 1.2 x 10-8 /m3/sec
Total for Classified Areas 6.1 x 10-6 /m3 1.5 x 10-8 /m3/sec
* These values are alternatives and should not be used in combination.

** Specific failures are given for electrical types according to EX protection, but not be used in combination with general value
for electrical components.
37
Table 5-2: Ignition Intensity Values Used in the TDIIM Model
Immediate Ignition
Release rates of 0.1 kg/s – 1 kg/s 0.0001
Release rates of 1 kg/s – 10 kg/s 0.001
Release rates greater than 10 kg/s 0.01
Delayed Ignition
Item/Area Continuous Discrete
Pumps (per item exposed)* 9.6 x 10-5

2.1 x 10-7 /sec
Compressors (per item exposed) 2.3 x 10-3 5.1 x 10-6 /sec
Other equipment 2.6 x 10-6/m2 2.1 x 10-9/m2/sec
Electrical Equipment 2.6 x 10-6/m2 2.7 x 10-8/m2/sec
Personnel 3.0 x 10-6/m2 4.0 x 10-8/m2/sec
Other sources 1.3 x 10-6/m2 1.7 x 10-8/m2/sec
*For oil leaks from pumps a value of 1.5 x 10-2 per item is used for continuous sources and 8.9 x 10-6 per item per second for
discrete sources.
Coarse Example
A 5 kg/s release in a classified area of an offshore installation creates a flammable gas
cloud of 400 m2 and a volume of 1500 m3 which remains for 300 seconds before the
concentration falls below the lower flammable limit. The area has 2 compressors both of
which are exposed to the cloud prior to isolation and shut down. There are no hot work
hours or external sources, and ignition source control is not accounted for.4
MISOF Model
Immediate Ignition 7.0 x 10-4
Delayed (continuous sources)

• Compressors: 2 x 3.7 x 10-3 = 7.4 x 10-3
• Electrical Components: 1500 m3 x 1.8 x 10-6 /m3 = 2.7 x 10-3
• Other Sources: 1500 m3 x 6.0 x 10-7 /m3 = 9.0 x 10-4
Delayed (discrete sources)

• Compressors: 2 x 300 secs x 1.6 x 10-6 /sec = 9.60 x 10-4
• Electrical Components: 1500 m3 x 300 secs x 1.5 x 10-9 /m3/sec = 6.75 x 10-4
• Other Sources: 1500 m3 x 300 secs x 1.2 x 10-8 /m3/sec = 5.40 x 10-3
Total 1.87 x 10-2
4 The approach illustrated here uses a mathematical simplification which makes only a small difference to the results where the number
of ignition sources and the ignition strengths are both small.
38
TDIIM Model
Immediate Ignition 1 x 10-3
Delayed (continuous sources)

• Compressors: 2 x 2.3 x 10-3 = 4.60 x 10-3
• Other Equipment: 400 m2 x 2.6 x 10-6/m2 = 4.60 x 10-3
• Electrical Equipment: 400 m2 x 2.6 x 10-6/m2 = 1.04 x 10-3
• Personnel: 400 m2 x 3.0 x 10-6/m2 = 1.20 x 10-3
• Other Sources: 400 m2 x 1.3 x 10-6/m2 = 5.20 x 10-3
Delayed (discrete sources)

• Compressors: 2 x 300 secs x 5.1 x 10-6 = 3.06 x 10-3
• Other Equipment: 400 m2 x 300 secs x 2.1 x 10-9/m2 = 2.52 x 10-4
• Electrical Equipment: 400 m2 x 300 secs x 2.7 x 10-8/m2 = 3.24 x 10-4
• Personnel: 400 m2 x 300 secs x 4.0 x 10-8/m2 = 4.80 x 10-4
• Other Sources: 400 m2 x 300 secs x 1.7 x 10-8/m2 = 2.04 x 10-4
Total 1.37 x 10-2
It can be seen from the above example that although the models are similar in formulation
the values calculated for individual pieces of equipment may vary significantly. The main
reason for the difference is related to the statistical basis for the MISOF model relative to
the TDIIM model. The latter model covered incident data prior to 1998. Also, the quality
of the data base forming the basis for MISOF are higher than the data used to set the
parameters in TDIIM.
• Ignition due to hot work or exposure of external ignition sources: In addition to
the ignition probability associated with ignition mechanisms described above,
contributions to the delayed ignition probability may occur as a result of hot work
activity as well as ignition sources outside the module:
–– It is recommended that ignition probability due to hot work is equated to the
number of class A5 hot work hours per year divided by the total number of hours
per year. For example, the additional ignition probability for a module with 40
hours of hot work per year is 40/8760 = 0.0046. However, the use of a habitat
should be accounted for. This is dealt with in the MISOF model by assuming the
probability of human failure of closing habitat door (30%) and multiplying by the
probability of gas ingress upon exposure of habitat with open door (ranging from
17% in case of small leak to 100% in case of leak size exceeding 30 kg/s).
–– If the number of hot work hours assumed in deriving the correlations are
considered unrepresentative, a more appropriate assessment can be made
using the full UKOOA model.
5 Class A hot work includes welding, burning and grinding. Class B hot work includes activities such as sandblasting, use of non Ex-
electrical equipment and photographing.
39
–– Conditional ignition probabilities upon exposure of flammable fluid to strong

ignition sources should be accounted for, and assigned ignition probabilities
based upon the exposure of flammable fluid. Such ignition sources include gas
turbine air intakes, combustion engine air intakes, gas ingress to enclosures
protected by ventilation systems, non-EX equipment in unclassified areas,
supply vessels and flares.
–– T
he MISOF model suggests an ignition probability of 50% for exposure of
flammable fluid to a gas turbine air intake at any point within 5 minutes of
shut down of the turbine.
–– F
or combustion engine air intakes exposed to stoichiometric gas the
MISOF model suggests an ignition probability of 90% if no flame arrestor
is installed and 1% if flame arrestor is installed. An adjustment factor for
non-stoichiometric gas exposure is also presented.
–– F
or ignition probabilities for the other sources listed, such as supply
vessels see [8].
• Ignition control: The ignition models include specific guidelines to account for
reduction of delayed ignition probability upon flammable gas exposure due to
isolation and shut down of ignition sources upon confirmed gas detection.
–– For discrete sources, each of the three models propose reduction factors of the
discrete ignition intensities upon confirmed gas detection. The reduction factor
per ignition mechanism category reflect the fraction of ignition sources within
each category that are successfully shut down or isolated upon gas detection.
–– Likewise, a fraction of continuous ignition sources that are isolated or shut
down per ignition mechanism category are suggested for each of the three
ignition models. However, ignition probability due to continuous ignition sources
is not necessarily eliminated upon isolation. Continuous ignition sources that
are shut down will gradually cool down until their temperature is too low to
ignite the flammable fluid. Hence, upon shutdown of a continuous ignition
source the models suggest a “half time” indicating the interval for the ignition
intensity of sources that are shut down to be reduced by 50% due to cooling.
• The TDIIM and OLF models suggested correction factors for general platform specific
properties such as age of the installation, technology (such as progress in safety
system and barrier design), maintenance level and manning level. These types of
suggested correction factors are not included in the MISOF model as they cannot
conclusively be extracted from the statistical basis.
There are significant uncertainties associated with estimating ignition probabilities since
ignition mechanisms are not fully understood and the nature of ignition probability is too
complex to capture in a mathematical model. This is noted in the MISOF report which points
out that the ignition model is based on a statistical framework which, in turn, is based on
observed historical incidents reported in the UK offshore oil and gas industry between 1992-
2016 and Norwegian offshore industry between 1992-2016. In fact, the base line probability
of ignition within the process module and the main contribution distribution between
immediate, delayed and external ignition sources are based on only three ignited leaks; this
represents a challenge with regards to distribution of the conditional ignition probability from
the various types of potential sources of ignition. These three registered events were the only
ones found representative for a typical major accident hazard release scenario addressed
40
in quantitative risk analysis of fires and explosions, when scrutinising the registered leak
events in the UK sector between 1992 and 2016 and Norwegian sector between 1992 and
2016. A few observed ignited events such as ignition due to exposure of combustion air intake
and ignition due to hot work activity are observed in addition to these three, but not counted
in this context since they are covered by other parts of the ignition model rather than in the
calculation of ignition probability within the hazardous area within the process module.
The ignition model parameters are coupled to 1,0936 observed leaks meeting relevant
criteria, which is the decisive number in terms of statistical variability. The high number of
leaks per ignited leak implies that the overall ignition probability can be set with reasonable
confidence. In the MISOF model, it is shown that the uncertainty with regards to the overall
ignition probability is within a factor of two if the guidelines are adhered to. Since the
delayed ignition probability is estimated per exposed m3 there are additional uncertainties
relating to predicting the total number (and duration) of cubic meters exposed to flammable
fluid from the 1,093 leaks (ignited and unignited). In addition, there is uncertainty relating to
the validity of the basic assumption of the model; that the ignition probability observed for
the average installation in the Norwegian and British offshore sector for the previous 20-25
years is representative of the future ignition probability for a specific offshore installation in
the similar geographical region.
5.2 BEVI Model

The Reference Manual “Bevi Risk Assessments” (BEVI manual) [12], describes the regulatory
requirements for performing Quantitative Risk Assessment (QRA) for onshore facilities
handling hazardous substances. The model is used mainly in the Netherland but could
be applied elsewhere. The manual describes assumptions, models and basic information
intended for use in QRAs. The current version is version 3-3 and is available in Dutch.
Version 3-2 is available in English [13]. The methodology is the same in both versions.
When estimating ignition probability in line with the BEVI manual the ignition sources are
identified and ignition probability calculated based on contribution from
• Direct (immediate)7 ignition
• Delayed ignition
5.2.1 Immediate Ignition

Specific probabilities for immediate ignition are tabulated in the BEVI manual. The
probabilities for immediate ignition are dependent on type of installation, substance
category (reactivity of flammable fluid) and release quantity or rate. For example, for a
substance classed as “Category 0”8, the probability of immediate ignition varies with the
release rate (continuous releases) or quantity released (instantaneous releases). These
probabilities are given in Table 5-3. The manual [12,13] provides a fuller listing.

6 653 leaks fulfilling certain criteria in the HSE HCRD between October 1992 and March2015 and 200 leaks in RNNP (Norwegian sector)
between 2001-2015.
7 The term “direct ignition” is used in the manual but has the same meaning as “immediate ignition”. The term “immediate” is used in
this section for consistency.
8 See [11, 13] for definition.
41
Table 5-3: Immediate Ignition Probabilities for Flammable Gases
Probability of Immediate Ignition

Continuous Instantaneous
Releases Releases
Average /High Reactivity9 Low Reactivity
< 10 kg/s < 1,000 kg 0.2 0.02
10 – 100 kg/s 1000 – 10,000 kg 0.5 0.04
> 100 kg/s > 10,000 kg 0.7 0.09
5.2.2 Delayed Ignition

The delayed ignition mechanism is similar to the description of continuous and discrete
ignition sources given in section 5.1: i.e., ignition probability increases upon prolonged
exposure of an ignition source to flammable fluid. Ignition sources are represented either
as point sources or line sources or area sources (for populations only).
In general, the probability of delayed ignition during the time window 0 to t, denoted P(t),
from a given ignition source is expressed as
P(t) = Ppresent (1 – e-ωt)
Where Ppresent is the probability that the ignition source is present in the flammable cloud
ω (s-1) is the effectiveness of the ignition source
and t (s) is duration of exposure.
The probability of ignition of a flammable cloud during a time window of one minute for a
number of sources listed in the BEVI manual is tabulated in Table 5-4 for comparison.
Table 5-4: Examples of ignition probabilities in 1 minute given in the BEVI manual
Effectiveness of ignition Probability of ignition

Source type Ignition source
source, ω (s-1) in one minute
Point Adjacent process installation 0.01155 0.5
Flare 1.00000 1.0
Oven (outside) 0.03838 0.9
Oven (inside) 0.00996 0.45
Boiler (outside) 0.00996 0.45
Boiler (inside) 0.00436 0.23
Line High-voltage cable (per 100m) 0.00372 0.2
Motor vehicles (See Section 5.2.3) 0.00851 0.4
Trains (See Section 5.2.3) 0.02682 0.8
Ships 0.01155 0.5
Population Households (per person) 0.00017 0.01
Offices (per person) 0.00017 0.01
9 Highly reactive substances include hydrogen, acetylene, ethylene oxide and carbon disulphide
42
5.2.3 Motor Vehicle and Train Ignition Sources

The probability of an ignition due to a motorway in the vicinity of an installation is
determined by the average traffic density and the probability of ignition per vehicle. The
default value assumes an average speed of 80 km/h and 1500 motor vehicles per hour.
Similarly, the probability of an ignition due to a railway line in the vicinity of an installation
is determined by the average number of trains. The default is an average speed of 80 km/h
and 8 trains per hour.
The number of ignition sources in the cloud is given by:
d=NE/v
Where N the number of passing vehicles/trains per hour
E length of road or railway in the flammable cloud (km)
and v average vehicle/train speed (km/hour)
If d <= 1, then d is equal to the probability that the source is present when the flammable
cloud passes by, i.e., Ppresent, the probability of an ignition is then;
P(t) = d (1 - e-ωt)
If d > 1, then d represents the average number of sources that is present when the
flammable cloud passes by; the probability of an ignition is then:
P(t) = (1 - e-dωt)
Example
If the traffic density is 1500 vehicles per hour and vehicles are travelling at 80 km/hour
through a flammable cloud for 200 metres the average number of sources present is:
d = 1500 x 0.2 /80 = 3.75
The effectiveness of the ignition for one vehicle, ω, is equal to 8.51 × 10-3 s-1 giving the
probability of 0.4 in 60 seconds.
If the cloud is present for 30 seconds, the probability of ignition is:
P(30) = (1 – e-3.75 x 0.00851 x 30) = 0.616
5.2.4 Combined Probability

The total ignition probability, when applied to the calculation of societal risk, can be
calculated by combining the probabilities from the immediate ignition probability, Pimm, and
individual remote ignition sources using the equation:
[ ]
n
P(t)total = Pimm + (1 - Pimm ) 1- ∏ (1 - P

present (1 - e-ωt))i
i=1
43
Subject to a maximum value of 1. A different formulation is used for the calculation of

individual risk (referred to as the “free field method”). Details are provided in [11 and 13].
5.3 The CCPS Model

A model aimed primarily at evaluating the probability of ignitions in onshore installation is
described in a publication by the CCPS [14]. The guidelines define 3 levels of estimation:
• L
evel 1 (Basic) Analysis – suitable for Process Hazard Analysis (PHA) risk matrix
applications and some Layers of Protection Analyses (LOPAs) and Failure Mode,
Effect and Criticality Analyses (FMECAs).
• L
evel 2 (Intermediate) Analysis – suitable for LOPAs, FMECAs and screening level
QRAs.
• L
evel 3 (Advanced) Analysis – suitable for QRAs, cost-benefit analysis and
consequence modelling when frequency is also estimated.
The level 3 analysis evaluates the probability of immediate ignition as a combination of

contributions from auto-ignition (based on the ratio of the discharge temperature to the
material’s auto ignition temperature) and static discharges, which are a function of the
discharge pressure and the minimum ignition energy of the released material.
For delayed ignitions, the level 3 analysis estimates the probability based on the following:
• Ignition source strength
• Duration of exposure
• Modification factors to account for
–– The magnitude of the release
–– The minimum ignition energy for the material
–– The temperature of the release
–– Whether the release is indoors or outdoors
–– The effectiveness of ignition control
–– The influence of ventilation.
Equations, suggested data values and guidance is available in [14].
44
6. Recommended data sources for

further information
This document gives sufficient detail of the use of the UKOOA ignition probability
correlations for it to be used for standard scenarios. The model guidance [1] should be
consulted if use of the full model is required.
Details of all other ignition models referred to in this datasheet are given in summary form
to provide an overview of the main features. The source documents referenced in each
case should be consulted prior to use of these models to gain a full understanding of the
methodology and the full list of data values.
45
7. References
[1] Guidance on assigning ignition probabilities in onshore and offshore quantitative risk assessments,
2nd Edition Energy Institute, May 2019
[2] B. Bain, M. Celnik & G. Korneliussen, Practical Implementation of the UKOOA Ignition Model, Hazards XXIII
Symposium, Southport, Nov 2012.
[3] Lloyd’s Register Consulting, Blowout and Well Release Frequencies – Based on SINTEF Offshore Blowout
Database, 2016, Report No. 19101001-8/2017/R3, April 2017.
[4] SINTEF, Annual reports entitled “Blowout and Well Release Characteristics and Frequencies”.
[5] Acton, M.R., Acton, O.J. and Robinson. C., A Review of Natural Gas Transmission Pipeline Incidents to Derive
Ignition Probabilities for Risk Assessment, Hazards 26 (Edinburgh), Symposium Series No. 161, IChemE,
May 2016.
[6] IGEM/TD/2 Edition 2, Assessing the Risks from High Pressure Natural Gas Pipelines, Institution of Gas
Engineers and Managers, Communication 1764, 2013.
[7] Cox, Lees and Ang, 1991. Classification of Hazardous Locations, Rugby: Institution of Chemical Engineers,
ISBN 0 85295 258 9.
[8] Modelling of ignition sources on offshore oil and gas facilities – MISOF(2), Report for Norwegian Oil and Gas
Association, Report no: 107566/R2, 2018-11-15.
[9] Ignition modelling in Risk Analysis, report for OLF, Scandpower report no. 89.390.008/R1, Revision 01, 19
March 2007.
[10] JIP Ignition modelling - Time Dependent Ignition Probability Model, DNV report no. 96-3629, revision 4,
18.02.1998, ref Guidelines for use of JIP ignition model, DNV report no 99-3193 / Scandpower report no
27.29.03, revision 01, 23.04.99.
[11] Lloyds Register. Process leak for offshore installations frequency assessment model – PLOFAM, Report no:
105586/R1. 2016
[12] Reference Manual Bevi Risk Assessment, Version 3.3, 01.07.15, National Institute of Public Health and the
Environment (RIVM) (In Dutch).
[13] Reference Manual Bevi Risk Assessments, Version 3.2, 01-07-2009, National Institute of Public Health and
the Environment (RIVM) (in English).
[14] CCPS, Determining the Probability of Ignition of a Released Flammable Mass, Wiley, July 2014.
46
47
www.iogp.org
Registered Office Brussels Office Houston Office
City Tower Avenue de Tervuren 188A 19219 Katy Freeway
Level 14 B-1150 Brussels Suite 175
40 Basinghall Street Belgium Houston, TX 77094
London EC2V 5DE USA
T +32 (0)2 790 7762
United Kingdom
[email protected] T +1 (713) 261 0411
T +44 (0)20 3763 9700 [email protected]
[email protected]
The data presented in section 2

provide estimates of the probabilities
of hydrocarbon releases igniting
to result in an explosion and/or a
sustained fire. This data may be
applied to any of the leak types
described in the Process Release
Frequencies datasheet.

IOGP-Report 434-06-1 - Ignition Probabilities

Uploaded by

Copyright:

Available Formats

IOGP-Report 434-06-1 - Ignition Probabilities

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

IOGP-Report 434-06-1 - Ignition Probabilities

Uploaded by

Copyright:

Available Formats

REPORT SEPTEMBER

RISK ASSESSMENT DATA DIRECTORY

Photography used with permission courtesy of

IOGP welcomes feedback on our reports: [email protected]

These Terms and Conditions shall be governed by and construed in accordance

RISK ASSESSMENT DATA DIRECTORY

VERSION DATE AMENDMENTS

1.0 September 2019 First release

2. Summary of recommended data 7

3. Guidance on use of the UKOOA model 27

4. Review of primary data source 33

5. Alternative ignition models 35

6. Recommended data sources for further information 45

FPSO Floating Production Storage and Offloading (Installation)

LPG Liquefied Petroleum Gas

NAP Normal Atmospheric Pressure

NUI Normally Unmanned Installation

OLF Oljeindustriens landsforening

QRA Quantitative Risk Assessment

SIMOPS Simultaneous Operations

TDIIM Time Dependant Internal Ignition Model

UKOOA United Kingdom Offshore Operators Association

The data presented in section 2 provide estimates of the probabilities of hydrocarbon

2. Summary of recommended data

2.1 Ignition Probability Curves

Although a distinction is made between “immediate” and “delayed” ignitions in this

Table 2-1: Onshore Ignition Scenarios

Source: Energy Institute [1]

Table 2-2: Offshore Ignition Scenarios

Source: Energy Institute [1]

Table 2-3: Special (Derived) Ignition Scenarios

Source: Energy Institute [1]

Data Sheet 1: Scenarios 1 – 4

Data Sheet 2: Scenarios 5 – 7

Data Sheet 3: Scenarios 8 – 11

Data Sheet 4: Scenarios 12, 13 & 30

Data Sheet 5: Scenarios 14 – 16

Data Sheet 6: Scenarios 17 & 18

17 - Offshore Process Liquid 18 - Offshore Process Liquid

Offshore Process Liquid (17) Offshore Process Liquid NUI (18)

Data Sheet 7: Scenarios 19 – 22

Data Sheet 8: Scenarios 24 – 26

24 - Offshore FPSO Gas 25 - Offshore FPSO Gas 26 - Offshore FPSO Gas

Data Sheet 9: Scenarios 23 & 27

23 - Offshore Riser 27 - Offshore Riser

Offshore Riser (23) Offshore Engulf_blowout_riser (27)

The two probabilities can be combined using the following equation:

Pign = 1 - [(1 - Pigngas/LPG )(1 - Pignliquid )]

Further options for determining intermediate correlations are given in [2]

2.2 Ignition timing

Table 2-4: Suggested Ignition Timing Distribution Given Ignition Occurs

Relative Cumulative Probability

100 0.63 0.6

1000 0.94 0.86

Figure 2-2: Ignition Timing Distribution Given Ignition

2.3 LNG Releases

2.4 Blowout Ignition Probabilities

Early ignition Delayed ignition Very Delayed

Shallow Gas Blowout Historic 0.07 0.11 0.07

Drilling and Heavy