RP F107 PDF

RECOMMENDED PRACTICE
No. DNV-RP-F107
RISK ASSESSMENT OF PIPELINE PROTECTION

MARCH 2001
Since issued in print (March 2001), this booklet has been amended, latest in October 2002.
See the reference to “Amendments and Corrections” on the next page.
DET NORSKE VERITAS

Veritasveien 1, N-1322 Høvik, Norway Tel.: +47 67 57 99 00 Fax: +47 67 57 99 11
FOREWORD
DET NORSKE VERITAS (DNV) is an autonomous and independent Foundation with the objectives of
safeguarding life, property and the environment, at sea and onshore. DNV undertakes classification,
certification, and other verification and consultancy services relating to quality of ships, offshore units
and installations, and onshore industries world-wide, and carries out research in relation to these
functions.
DNV publishes various documents related to the offshore industry, aimed at promoting quality and
safety on offshore units and installations.
The Recommended Practice publications (RP-series) cover proven technology and solutions which have
been found by DNV to represent good practice, and which represent one alternative for satisfying the
requirements stipulated in the DNV Offshore Standards or other codes and standards cited by DNV. The
DNV RP-series is divided into 6 parts, as follows.
A. Quality and Safety Methodology
B. Materials Technology
C. Structures
D. Systems
E. Special Facilities
F. Pipelines & Risers
As well as forming the technical basis for DNV verification services, the Offshore Standards and
Recommended Practices are offered as DNV’s interpretation of safe engineering practice for general use
by the offshore industry.
ACKNOWLEDGEMENTS
This Recommended Practice is based upon a project guideline developed by DNV for Statoil.
DNV would like to take this opportunity to thank Statoil for their financial and technical contributions.
DNV is further grateful for valuable co-operation and discussion with the individual personnel in Statoil
participating in the project.
This Recommended Practice has been distributed for both internal and external hearing. DNV would
like to thank all companies giving valuable feed-back and comments to this document.
Comments may be sent by e-mail to rules@dnv.com
For subscription orders or information about subscription terms, please use distribution@dnv.com
Comprehensive information regarding DNV services, research and publications can be found at http://www.dnv.com, or can be obtained from DNV,
Veritasveien 1, N-1322 Høvik, Norway; Tel +47 67 57 99 00, Fax +47 67 57 99 11.
© 2001 DET NORSKE VERITAS. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means,
including photocopying and recording, without the prior written consent of DET NORSKE VERITAS.
Printed in Norway by GCS AS.
If any person suffers loss or damage which is proved to have been caused by any negligent act or omission of Det Norske Veritas, then Det Norske Veritas
shall pay compensation to such person for his proved direct loss or damage. However, the compensation shall not exceed an amount equal to ten times the
fee charged for the service in question, provided that the maximum compensation shall never exceed USD 2 million. In this provision “Det Norske Veritas”
shall mean the Foundation Det Norske Veritas as well as all its subsidiaries, directors, officers, employees, agents and any other acting on behalf of Det
Norske Veritas.
Amended October 2002 Recommended Practice DNV-RP-F107, March 2001
see note on front cover Page 3
CONTENTS
1 General .....................................................................5 5.4 Ship traffic........................................................... 24
1.1 Introduction............................................................5 5.5 Simultaneous operations ..................................... 30
1.2 Objectives ..............................................................5 5.6 Trawling .............................................................. 30
1.3 Scope and Application ...........................................5 5.7 Anchor handling.................................................. 30
1.4 General considerations...........................................5 5.8 Frequency ranking............................................... 31
1.5 Limitations.............................................................5 6 Consequence .......................................................... 32
1.6 Definitions .............................................................6 6.1 Introduction ......................................................... 32
2 Methodology.............................................................7 6.2 Human safety ...................................................... 32
2.1 Introduction............................................................7 6.3 Release to the environment ................................. 33
2.2 Safety objectives ....................................................7 6.4 Economic loss ..................................................... 33
2.3 Acceptance criteria ................................................7 7 Risk assessment ..................................................... 35
2.4 System description.................................................7 7.1 General ................................................................ 35
2.5 Hazard identification..............................................8 7.2 Uncertainty assessment ....................................... 35
2.6 Risk Assessment ....................................................8 8 References .............................................................. 36
2.7 Risk reducing measures .........................................9 Appendix A. Example of risk assessment procedure
3 Activity description................................................11 for dropped objects ........................................................... 37
3.1 Platform/Rig ........................................................11 A.1 Introduction ......................................................... 37
3.2 Subsea operations ................................................11 A.2 Design Basis........................................................ 37
3.3 Fishing .................................................................11 A.3 Classification of objects (Section 3.1.1)............. 38
3.4 Ship......................................................................12 A.4 Drop frequency (Section 5.2.1) ........................... 38
4 Pipeline and protection capacity...........................13 A.5 Excursion of objects (Section 5.2.2) ................... 38
4.1 General.................................................................13 A.6 Hit probability (Section 5.2.2)............................. 40
4.2 Damage classification ..........................................13 A.7 Hit frequency versus energy (section 5.3)........... 41
4.3 Steel pipeline .......................................................14 A.8 Damage capacity versus energy (section 4) ........ 42
4.4 Flexible pipeline ..................................................15 A.9 Damage versus frequency ................................... 43
4.5 Umbilical .............................................................16 Appendix B. Impact capacity testing procedure....... 44
4.6 Different protection methods ...............................16 B.1 Introduction ......................................................... 44
5 Failure frequency...................................................19 B.2 Test energy.......................................................... 44
5.1 Introduction..........................................................19 B.3 Test Equipment ................................................... 44
5.2 Crane activity.......................................................19 B.4 Procedure ............................................................ 44
5.3 Energy calculation ...............................................21
DET NORSKE VERITAS

Recommended Practice DNV-RP-F107, March 2001 Amended October 2002
Page 4 see note on front cover
DET NORSKE VERITAS

1.4 General considerations

1 General When using this recommended practice, note that the
following points are applicable:
1.1 Introduction
This recommended practice presents a risk-based approach 1) Risk estimation should normally be conservative.
for assessing pipeline protection against accidental external 2) Repeated assessments for alternative protection
loads. Recommendations are given for the damage capacity measures may be required.
of pipelines and alternative protection measures and for 3) Economic criteria will often be decisive.
assessment of damage frequency and consequence. 4) In each project, the risk should be kept as low as
Alternative pipeline protection measures are also presented. reasonably practicable.
5) It is important to pay attention to the total risk picture.
The pipelines/risers/umbilicals under consideration will
1.2 Objectives
give a contribution to the risk of a installation and the
The objective of this recommended practice is to provide a total risk-picture of the installation has to be considered.
basis for risk assessment of accidental events which lead to
external interference with risers, pipelines and umbilicals and It is important to realise that a safe and economic pipeline
to give guidance on protection requirements. and umbilical design should be considered as part of a
complex system, which includes other areas such as:
The recommended practice gives guidance for pipeline and
riser protection design in accordance with the requirements – template design and field lay-out;
and safety levels stated in DNV Offshore Standard DNV- – subsea operations (drilling, completion, intervention,
OS-F101 Submarine Pipeline Systems (DNV 2000) and maintenance);
DNV Offshore Standard DNV-OS-F201 Metallic Risers − platform activities.
(DNV 2000b).
In order to achieve an optimum pipeline/umbilical protection
design, the whole life-cycle system efficiency should be
1.3 Scope and Application
evaluated. This implies that relevant interfaces and
This recommended practice focuses on providing a interactions with other designs, activities and operational
methodology for assessing the risks and required protection procedures shall be identified and described in details as
from dropped crane loads and ship impact to risers and early as possible. The whole system can then be optimised
pipeline systems within the safety zone of installations. with respect to safe operations and economy, and a sub-
Accidental scenarios with other relevant activities such as optimisation of the pipeline/umbilical design will be avoided.
anchor handling, subsea operations and trawling are also
discussed. Where applicable information exists, specific Among the areas, or aspects, of particular importance are
values or calculation procedures are recommended. If no
such information is available, then a qualitative approach is – Subsea wells: Stop of production should be minimised,
given. and measures to achieve this objective are consequently
of high priority. A shutdown can also affect the pipeline
The recommended practice is applicable for the following system as hydrates may form or wax is deposited. The
two scenarios: expected scope and frequency of intervention work
should also be considered.
a) control that implemented control and protection – Field lay-out: Optimising the field layout with respect to
measures are acceptable. the pipeline length or cable length can, in reality, be sub-
b) optimisation of planned protection. optimisation. The layout of pipelines and cables near
subsea wells or templates should also be evaluated with
All the generic frequencies presented in this recommended respect to rig-operations. At a fixed platform, the
practice, e.g. the drop frequency, are based on operations of optimum pipeline or cable routing can be in areas where
North Sea installations. These frequencies are not generally the lifting activity is low or none-existent, thus reducing
applicable for other parts of the world. However, the general the protection requirements.
methodology is applicable throughout the world.
− Rig heading relative to tie-in corridor: The pipeline tie-
Acceptance of protection measures can be based on operator in corridor should take into account the dominant rig
supplied risk acceptance criteria covering human safety, heading and anchor pattern.
environment and economics, or the failure frequencies given For subsea wells, possible scenarios involving simultaneous
in DNV-OS-F101. operations shall be defined at the design stage of a project.
With respect to pipelines, the risk methodology used in this
recommended practice is applicable to pipelines within 1.5 Limitations
offshore petroleum field developments. It should be noted This recommended practice covers only risk assessment of
that this document does not include regular 3rd party risk accidental loading from external events/interference on
evaluations as found in onshore developments. offshore risers, pipelines and umbilicals. The limits for the
application of this document are (see also Figure 1):
DET NORSKE VERITAS

Recommended Practice DNV-RP-F107, March 2001 Amended October 2002
Page 6 see note on front cover
– on a fixed or floating platform, below cellar deck; Furthermore, this recommended practice covers the risk
− on a subsea installation, at the connection point to the assessment from accidental external events only and hence is
subsea manifold/piping. a contribution to the total risk of pipeline operations. Other
risks, which contribute to the total risk of pipeline operations
The above limits indicate that this document covers tie-in as corrosion, erosion, burst etc. are not included.
towards subsea installations up to the outboard hub.
Requirements to any nearby protection structures should This recommended practice describes risk assessments
comply with this recommended practice. related to accidental scenarios of the lifetime of the pipeline
during normal operation conditions and planned activities
It is important that all parts of the subsea production system (e.g. drilling and completion operations). Risks related to
are covered either by this recommended practice or by other single, major, critical operations, such as construction work,
standards. For protection requirements of subsea installations are not included. The risk of such operations should be
reference is made to other standards, e.g. NORSOK (1998). addressed separately.
For purposes other than risk assessment for risers, pipelines
or umbilicals as covered by this recommended practice, the
information and methodology given should not be used
without further documentation/clarification.
Below cellar deck

Connection point
to subsea piping
Tie-in area up to outboard hub
Topside Riser Pipeline Subsea installation
Not covered Not covered

Covered by this document
by this document by this document
Figure 1 Application of the recommended practice
Damage: Damage to pipelines is divided into three

1.6 Definitions
categories, minor, moderate and major. The damage
Acceptance criteria: criteria used to express an acceptable categories form the basis for both the frequency calculations
level of risk for the activities. and the consequence evaluations. The damage classification
is given in 4.
Consequence: describes the result of an accidental event.
The consequence is normally evaluated for human safety, Frequency: used to describe the likelihood per unit time of
environmental impact and economic loss. an event occurring.
Consequence ranking: used to describe the severity of a Frequency ranking: used to describe the frequency of an
consequence. The consequence is ranked from 1 (minor, event. The frequency is ranked from 1 (low) to 5 (high).
insignificant) to 5 (major, catastrophic).
Platform: (as used in this document) refers to a permanent
Conditional probability: probability of one event given a installation, e.g. a concrete gravity base structure (GBS), a
preceding event. steel jacket, a tension leg platform (TLP), a floating
production unit (FPU), etc.
DNV: Det Norske Veritas
Rig: (as used in this document) refers to a temporary
installation, e.g. mobile offshore drilling unit.
Risk: expression of the product of the frequency

(probability) and the consequence of an accidental event.
DET NORSKE VERITAS

2 Methodology The criteria for human safety and environmental impact shall
be established considering the risk as a contribution to the
2.1 Introduction total risk for the platform or rig or the whole field.
Prior to any risk assessment, the safety objectives for the
activities and the acceptance criteria for the risk shall be 2.4 System description
defined by the operator. Prior to risk assessment, a complete system description
should be prepared. The description shall cover the entire
The basis of any risk evaluation relies on a comprehensive pipeline/umbilical lifecycle and should as a minimum
system description. This system description is used to consider the following:
identify hazards with potential to affect the pipeline /
umbilical. The identified hazards are evaluated in a risk 1) Activities potentially affecting pipeline/umbilical
assessment. integrity (see section 3);
– crane handling on platform or rig,
This section describes the above aspects of the risk – fishing (bottom trawling),
evaluation procedure and an overview of the total procedure – supply vessels and general ship traffic in the area or
is shown in Figure 2. close to the area considered,
– subsea operations (e.g. simultaneous operations as
2.2 Safety objectives drilling, completion and intervention),
To safely manage the activity, the operator shall define − others (planned construction work, etc).
safety objectives for avoidance or survival of accidental 2) Physical characteristics of the pipeline/umbilical (see
events, as required in DNV-OS-F101 section 2B. section 4);
– type (steel pipeline, flexible or umbilical),
2.3 Acceptance criteria – diameter, wall thickness, coating thickness,
In order to evaluate whether the risk of an accidental event is – material (steel and coating),
acceptable or not, acceptance criteria are required. The – construction details (connectors, swan necks, etc),
acceptance criteria shall state the acceptable limits for the − content (gas, oil, condensate, water, etc).
risks to human safety, environment and economy. The
operator shall establish the acceptance criteria prior to 3) Mitigation measures (see section 4 and 2.7);
beginning the risk evaluations. When considering several – protection,
pipelines, the acceptance criteria should reflect the total risk – routing,
level for all pipelines. – procedures.
The acceptance criteria shall be in line with the defined Safety objectives
safety objectives of the activity. Alternatively, the structural for the actvities
failure probability requirements given in DNV-OS-F101

Section 2 may be used as acceptance criteria, in which case
Definitions of
no consequence assessment is required and only the accept criteria
frequency of failure needs to be established. Note also that
this criterion is given per pipeline and several pipelines
should be treated individually. System
description
Guidance note:
The acceptable structural failure probability given in Table 2-5 in
Section 2 of DNV-OS-F101 may be modified, i.e. transformed Hazard
into a failure probablity per km given that any dependacy of identification See figure 3
accidental loading between different locations is accounted for.
- end - of - Guidance - note -
Risk Risk reducing
Guidance note: Assessment measures
Operational plans
For dynamic metallic risers, the requirements given in the DNV- Protection design
OS-F201 shall apply. Field lay-out
- end - of - Guidance - note - Acceptable

risk?
No
The acceptance criteria reflect acceptance of the risk Yes
contribution during a certain period. For a platform, the
activities are assumed to be continuous throughout the year, Acceptable
hence a year normally forms the basis for the risk protection
assessment. For drilling activities and intervention works
with duration less than a year, an equivalent annual risk is to Figure 2 Process Description of the Pipeline Protection
be used. Assessment.
DET NORSKE VERITAS

Amended October 2002
Recommended Practice DNV-RP-F107, March 2001
see note on front cover
Page 8
2.5 Hazard identification It is not normally practicable to protect against accidental

events that could occur during installation of pipelines and
Possible hazards that can cause damage to pipelines and
umbilicals. Risk reduction should therefore be specially
umbilicals should be identified based on the available
considered when drawing up operational plans and
information regarding activities in the area, see section 2.4.
procedures for such activities.
Hazard identification should systematically identify all
external accidental scenarios and possible consequences. Pipelines routed across known fishing areas should be
Table 1 states some typical hazards that can cause damage to designed against trawl interaction. Pipeline design against
risers, pipelines and umbilicals. The initial cause of the trawl interaction should be according to the DNV Guideline
hazard and the consequences for human safety, 13 Interaction between trawl gear and pipelines (DNV,
environmental impact and economic loss are not included in 1997). If the pipeline is designed against trawling in all
the table. Additional events should be included as applicable. phases, i.e. temporary and permanent, the hazard from
trawling may be ignored.
Table 1 Possible external hazards.

Operation/activity Hazard Possible consequence to pipeline
Dropped and dragged anchor/anchor chain from
pipe lay vessel
Impact damage
Vessel collision during laying leading to
dropped object, etc.
Installation of pipeline Damage to pipe/umbilical being laid or other
Loss of tension, drop of pipe end, etc. pipes/umbilicals already installed
Damage during trenching, gravel dumping, Impact damage
installation of protection cover, etc.
Damage during crossing construction. Impact damage
Installation of risers, Dropped objects Impact damage
modules, etc. (i.e. heavy lifts) Dragged anchor chain Pull-over and abrasion damage
Anchor handling Dropped anchor, breakage of anchor chain, etc. Impact damage
(Rig and lay vessel Dragged anchor Hooking (and impact) damage
operations)
Dragged anchor chain Pull-over and abrasion damage
Lifting activities Drop of objects into the sea Impact damage

(Rig or Platform operations)
ROV impact Impact damage

Subsea operations
(simultaneous operations) Manoeuvring failure during equipment Impact damage
installation/removal Pull-over and abrasion damage
Trawling activities Trawl board impact, pull-over or hooking Impact and pull-over damage
Collision (either powered or drifting) Impact damage
Tanker, supply vessel and Emergency anchoring Impact and/or hooking damage
commercial ship traffic
Sunken ship (e.g. after collision with platform Impact damage
or other ships)
2.6 Risk Assessment − estimated based on engineering judgement, operator

experience, etc.
An initial, accidental event (e.g. dropped container) can
develop into an end-event (e.g. hit of pipeline). In general, The frequency of occurrence is then given a ranking from 1
risk assessments consist of an estimation of the frequency of (i.e. low frequency) to 5 (i.e. high frequency).
the end-events and an evaluation of the consequence of the
end-events. Similarly, the consequence is either calculated or estimated,
then ranked from 1 (i.e. low, non-critical consequence) to 5
The frequency of occurrence can be either: (i.e. high, severe consequence).
– calculated when detailed information exists (e.g.

dropped crane load scenario), or
DET NORSKE VERITAS

Consequence
evaluation
Human safety,
Environmental impact,
Economical loss
Consequence
vs.
damage
Consequence
ranking
1 2 3 4 5
Frequency
estimation
5 Not acceptable
Frequency
4 ALARP region
vs. 3
damage
Acceptable
2
1 Event
Frequency
ranking Risk matrix
Figure 3 Process description of a risk assessment

(Figure is only schematic, actual acceptable limits need to be given by operator)
In this recommended practice, the end-event is classified into and not the corresponding frequency, of the incidents is
different damage categories (i.e. minor (D1), moderate (D2) found by such worst-case evaluations.
and major (D3) damage, see definition of damage in section
4.2) which forms the basis for the consequence ranking into If any of the risk-related basic parameters in the risk
5 different categories. The frequency ranking and assessment changes, e.g. the activity level, design,
consequence ranking shall be established for each of the parameters, operating procedures, are changed, the risk
relevant damage categories, thus giving the risk for each assessment should be updated to reflect these changes.
damage category.
In Figure 3, the ALARP (As-low-as-reasonably-practicable)
The risk is then evaluated by plotting the established region identifies an area where the risk is acceptable,
frequency and consequence in a risk matrix. The risk however further reduction of the risk should be pursued with
assessment is briefly described in Figure 3. The process for a cost-benefit evaluation.
dropped object scenario is described in detail in Appendix A.
The frequency ranking and the consequence ranking are 2.7 Risk reducing measures
further described in section 5 and section 6 respectively.
If the estimated risk is above the relevant acceptance
The risk matrix method makes it possible to effectively criterion, then risk reduction can be achieved by:
compare the risk from different events, even when the level
− reducing the frequency of the event,
of detailed knowledge varies.
− reducing the consequence of the event, or
For some isolated operations, the risk assessment − a combination of the above.
methodology outlined in this document is not applicable.
These are isolated critical operations such as larger lifting Table 2 presents some risk reducing measures. For ship
operations, e.g. lifting of new modules. The risk collision scenarios, additional risk reducing measures are
methodology is not applicable as reasonable frequency given in section 5.4.5.
estimates for such scenarios are difficult to obtain due to the In each project, the risk should be kept as low as reasonably
relative limited experience. For such operations Hazardous
practicable. This means that some low cost risk reduction
and Operability (HAZOP) studies, Failure Mode Effect
measures should be introduced even if the risk is considered
Analysis (FMEA) or other relevant methods can be used to to be acceptable. Frequency reduction measures shall be
identify critical conditions during the operations and possible
prioritised before consequence reduction measures.
equipment failures that can cause or aggravate critical
conditions, and ensure that effective remedial measures are
taken. Note however, that normally only the consequence,
DET NORSKE VERITAS

Page 10
To evaluate the economic effects of any risk reduction CM = cost of risk reducing measure
measures, a cost-benefit calculation shall be performed. The ∆CR = reduction in repair cost
cost-benefit value (CBV) is an evaluation of the ratio
between the increased cost of any additional measures, ∆CP = reduction in production loss
∆Cost, and the reduced risk, ∆Risk. A cost-effective solution PoF = probability of failure/failure frequency
will give a ratio less than unity. R = interest rate
Y = number of years
∆Cost
CBV = (1)
∆Risk
This can be calculated according to
CM
CBV =
∆C R + ∆C P (2)
∑y (1 + r ) y ⋅ PoF
where
Table 2 Risk reducing measures

Measure Reduces Comments
Limit lifting to certain zones, Frequency This reduces/eliminates the frequency effectively.
sectors, areas Often used when lifting heavy objects as BOP on rigs. The rig is withdrawn from
the area when lowering the BOP.
For pipe loading onboard a lay-barge only the crane on the side furthest away
should be used when laying parallel to or crossing existing line.
Limit the type of objects lifted in Frequency For example, only the cranes furthest away from the vulnerable area may lift
certain zones heavy objects.
Or to not allow pipe loading onboard lay barge within platform safety zone.
Reduces the frequency of the most critical objects, however does not eliminate
the risk totally.
Introduce safety distance Frequency The activity is either planned performed in a safe distance away from the
pipeline or vice versa (e.g. anchor handling).
Reduces/eliminates the risk efficiently.
Introduce safe areas Frequency Activity of a certain kind is not allowed within a specified area (e.g. trawling
nearby platforms).
Reduces/eliminates the risk efficiently.
Change the field lay-out Frequency By careful routing the same effect as for safety distance may be obtained for
parts of the pipeline.
Introduce extra Chaser Tug or Frequency To ensure that no interference between the anchor chain and the installation take
anchor chain buoys place.
Tie-in corridor in-line with rig Frequency The tie-in corridor should be in-line with the rig heading, thus the rig cranes are
heading above installation oriented in favourable positions.
Weather restrictions for Frequency If a prevailing current direction have been included in a safe distance evaluation,
operations. the activity should not be performed if the current direction is other than that
considered, or
If the frequency have shown to increase with increasingly worse weather, the
activity should be postponed until the weather normalises.
Increase the protection Consequence Increased protection will reduce the damage to the pipeline. Increased protection
may be obtained by a variety of solutions. It should be noted that some solutions
(e.g. massive tunnel structures) might introduce a very high risk to the pipeline
during installation, in addition also introduce scouring problems during the
lifetime.
Stop production in pipeline Consequence This effectively reduces the consequence of release, however this solution may
during activity be very expensive. Further, it does not reduce the economic consequence of
damage.
DET NORSKE VERITAS

Guidance note:
3 Activity description The possibility of smaller objects, which are not normally
accounted for in a dropped object scenario, falling into the sea
should be identified and taken into account. Inspections have
3.1 Platform/Rig revealed that there are a significant number of smaller objects on
the sea bottom close to platforms. These objects are not reported
3.1.1 Lifting activity as dropped from cranes.
The following information on the lifting activity is required - end - of - Guidance - note -
for input to the dropped object calculations, see section 5.2.
3.1.1.2 Lifting frequency
3.1.1.1 Object classification The lifting frequency of the identified objects shall be
The lifting activity description should include objects lifted established. The lifting frequency should include all
(where applicable): activities over a relevant time-period.
– between supply vessel and platform/rig, 3.1.1.3 Crane information

– between platform/rig and subsea installation, and
A typical platform has between one and four cranes, whereas
− internally on the platform, but with potential for objects a typical drilling rig has two cranes. Crane information
to drop into the sea. should be established considering:
Lifting activity information shall be collected for all relevant − crane location, for both derrick and normal cranes (note
operations, e.g. normal operating conditions for platforms that drop from some of the cranes may not have the
and drilling, completion, etc. for subsea installations. potential to hit a riser/pipeline);
All lifting operations with a possibility for a dropped load − crane operational radius and capacity, including
into the sea over or near to exposed pipelines or umbilicals limitations in operational area;
should be included. For estimating object excursion and hit − dedicated supply vessel off-loading locations;
energy, the object inventory should be as detailed as possible − platform specific aspect (e.g. one crane is normally used
including size and weight, see section 5.2. All lifting for food containers only).
activities during a representative time-period should be
covered. In lieu of more detailed information, the object 3.1.2 Anchor handling
classification in Table 3 may be used to establish the load For input to the dragged anchor calculations in section 5.7,
data. the following detailed information on the anchor handling
activity of a rig should be collected:
Table 3 Object classification, typical load data.
− anchor handling procedures;
Weight in air − anchor landing area and final placement, etc.;
no Description Typical objects 1,2
(tonnes) − type of anchor (size of anchor, chain and wire);
Drill collar/casing, − anchor penetration depth and dragging distance to
1 <2 achieve required holding capacity.
Flat/long scaffolding
2 shaped 2–8 Drill collar/casing
3 >8 Drill riser, crane boom 3.2 Subsea operations
Container (food, spare For input on the subsea operation evaluations in section 5.5,
4 <2 parts), basket, crane the following information on subsea operations should be
block collected:
Box/round
5 shaped 2–8 Container (spare parts),
basket, crane test block − procedures (drilling, completion and intervention);
6 >8 Container (equipment), − simultaneous operations (e.g. one well producing while
basket intervention work is performed on another);
7 Box/round >> 8 Massive objects as − manoeuvring routes above pipelines and umbilicals;
shaped BOP, Pipe reel, etc. − tools and equipment size;
1 Objects lifted during normal operation and maintenance will − frequency of operations.
normally be of all categories ranging from 1 to 6. Platform
cranes have a lifting capacity around 50 tonnes, thus only
derricks are normally used for lifting massive objects as in 3.3 Fishing
category 7. For input on the trawling evaluations in section 5.6, the
2 The classification in the table is based on platform activities following information should be established:
to/from supply vessels. For other activities e.g. to/from subsea
installations, an alternative classification may be more relevant. − type of activity (e.g. bottom trawling, pelagic trawling,
etc.);
− frequency for bottom trawling (based on normal
activities covering a relevant time-period);
− type of trawl equipment.
DET NORSKE VERITAS

Page 12
3.4 Ship − fishing vessel density (per km2),

Ship traffic data is used as the basis for a ship collision study, − supply boat arrivals to the platform (per year),
see section 5.4. The following ship traffic data are the typical − internal field transportation (per year),
background data required for the ship collision study: − effective loading/unloading time at the platform(hours
per year).
− merchant vessels passing the installation (per year),
− supply boats to nearby platforms (per year), In addition, the ship traffic in the area should be established
as input for emergency anchoring evaluations (see section
− supply boats to distant installations (per year),
5.7) for which information regarding the number and
− shuttle tanker to the platform (per year),
size/class of the different vessels should be obtained.
DET NORSKE VERITAS

4 Pipeline and protection capacity For dropped object scenarios, it should be noted that the
results of the risk assessment are not normally very sensitive
4.1 General to an absolutely “correct” capacity assessment. When the
loading is a complex compound of type of objects giving a
There are two typical accidental loading scenarios that can
variety of impact energies, a capacity estimate within ± 20%
lead to damage to riser, pipelines and umbilicals. These are
will normally give acceptable variations in the resulting risk
either impact (e.g. due to dropped objects) or pull-
level. However, the final risk estimate sensitivity to
over/hooking (e.g. due to dragged trawl board or anchor).
variations in capacity estimates should be checked if there is
The impact scenario is a complex dynamic, non-linear reason to believe that the final result is sensitive to the
mechanism that involves numerous parameters. In short, the capacity. For thin-walled, small diameter pipelines, flexibles
response (i.e. damage) of the riser, pipeline or umbilical is of and umbilicals without extra protection, the capacity is
a local nature, where the wall thickness and coating thickness normally negligible and may conservatively be set equal to
are important parameters. zero.
In this recommended practice, the given damage capacities The given capacity models given are focused on impact
of the pipeline and coating are conservatively assumed to loading and are given as energy absorption for different
absorb all of the available kinetic energy of the impacting levels of indentation, displacement or damage. The capacity
objects. However, energy absorption of the impacting object for buckling due to pull-over/hooking loading is only
itself, or into the soil, etc, may be accounted for, if discussed and is covered by the criteria for steel pipelines
documented. and risers given in DNV-OS-F101 and DNV-OS-F201
respectively. For umbilicals and flexible pipelines, the
Guidance note: capacity should be separately documented.
This is conservative as it is found that for small diameter
pipelines and soft soil conditions the absorption in the wall may The capacity of nearby fittings, connectors, flanges, etc.
be down to 50-60 % of the total kinetic energy. Further, for “non- should be individually determined. Such items may become a
rigid” objects such as containers, a considerable amount of weak link, especially when considering leakage.
energy will be absorbed by the object itself and not transferred to
the pipeline.
Using this recommended practice, the capacities for the
- end - of - Guidance - note - different protection methods shall be added to the capacity of
the pipeline/umbilical. Further, the protection is assumed to
The pull-over and hooking scenarios are of a global bending be completely damaged before the pipeline/umbilical is
behaviour and the bending stiffness of the pipeline or damaged. For concrete or polymer coatings on pipelines
umbilical is of importance. some interaction with the pipeline may be expected before
the ultimate capacity of the coating is reached. Protection
The impact capacities of pipelines, umbilicals and typical
failures are normally classified as minor damage (i.e. D1).
protection measures are given individually in this section.
Typical pipeline failure modes are indentation or puncturing The impact capacity can be determined by testing if the
of the pipe/umbilical wall (for impacting loads) and given formulations are not applicable. A testing procedure is
excessive bending (for pull-over loads). The failure modes given in Appendix B.
will be further classified according to the damage (i.e. D1 to
D3) and release (i.e. R0, R1 and R2) categories, see the
following section for damage class descriptions. 4.2 Damage classification
Material damage to the pipelines is classified by the
The capacity of the pipelines to withstand impact, pull-over following categories:
and hooking loads is dependent on both local pipeline
geometry (e.g. size and stiffness) and behaviour of load (e.g. − Minor damage (D1): Damage neither requiring repair,
impact energy, energy absorption by object). Until the event nor resulting in any release of hydrocarbons.
occurs, this information of the loading is not readily Smaller dents in the steel pipe wall, e.g. up to 5% of the
available and hence estimates of the capacity should be diameter, will not normally have any immediate
conservative. The capacity models given below describe an influence of the operation of the lines. This limit will
average capacity and should be used in risk assessments vary and must be evaluated for each pipe. Note however,
only. The capacity models should not be used for design if damage occurs then inspections and technical
purposes unless a characteristic lower bound model including evaluations should be performed in order to confirm the
safety factors is used and the applicability is further structural integrity.
documented. Minor damage to flexibles and umbilicals that do not
require repair action.
Guidance note: Any local damage to protective coatings or anodes will
For design of protection against trawling, the capacity not normally require repair action.
formulation given in the DNV Guideline no 13 (1997)
Interference between trawl gear and pipelines , which takes
account of the shape of typical trawl boards, should be used.
DET NORSKE VERITAS

Page 14
− Moderate damage (D2): Damage requiring repair, but 4.3 Steel pipeline
not leading to release of hydrocarbons. Dent sizes
restricting internal inspection (e.g. over 5% of the 4.3.1 Impact scenario
diameter for steel pipelines) will usually require repair. Most impacts are expected to result in a relatively “smooth”
Ingress of seawater into flexibles and umbilicals can dent shape. The dent - absorbed energy relationship for steel
lead to corrosion failures. However, the repair may be pipelines are given in equation (3), (Wierzbicki and Suh,
deferred for some time and the pipeline or umbilical 1988).
may be operated provided that the structural integrity is
confirmed. Equation (3) is based on a knife-edge load perpendicular to
Special consideration should be given to pipelines where the pipeline, and the indenting object covers the whole cross
frequent pigging is an operational requirement. For such section, see Figure 4. For conservatism, the effect of internal
pipelines, large dents will restrict pigging and lead to pressure is not included.
stop in production, and this damage should then be
considered as being major (D3) rather than moderate Detailed capacity evaluations, by e.g. FE analysis, may be
(D2) even though no release is expected. individually performed. Note however, that this requires
− Major damage (D3): Damage leading to release of detailed knowledge of the geometry of the impacting object.
hydrocarbons or water, etc. If the pipe wall is punctured
The additional failure of punching through the wall, leading
or the pipeline ruptures, pipeline operation must be
to leakage, can occur for higher velocity impacts or locally
stopped immediately and the line repaired. The damaged
small and sharp impact geometry. The possibility of leakage
section must be removed and replaced.
and total rupture is included as a progressive conditional
In case of a damage leading to release (D3), the following probability, where probability increases with increasing
classification of releases are used: impact energy.
− No release (R0): No release. Table 4 gives the proposed damage classification used for
− Small release (R1): Release from small to medium holes bare steel pipes.
in the pipe wall (<80 mm diameter). The pipeline may
release small amounts of content until detected either by 4.3.2 Pull-over/hooking scenario
a pressure drop or visually. Typical damage due to pull-over/hooking loads is local
− Major release (R2): Release from ruptured pipelines. buckling (i.e. buckling of the cross-section as a result of
Full rupture will lead to a total release of the volume of excessive bending). Buckling and other relevant failure
the pipeline and will continue until the pipeline is modes are covered in the criteria given in the DNV-OS-
isolated. F101. If these criteria are exceeded then the pipeline will
experience either increased ovalisation leading to a collapse
The damage categories are used for economic evaluations, of the cross-section or rupture due to excessive yielding in
whereas the release categories in addition are used for the longitudinal direction, the latter being most relevant for
estimating the risk for human safety and leakage to the small diameter pipelines (i.e. less than 6” – 8”).
environment. The release categories are of concern for the
human safety and for the environmental risk evaluations. The
classification of different failures into these categories will
depend on the type of line, e.g. steel or flexible, and the
protection.
1 1 3
 2π  2  D 2  δ 2 (3)
E = 16 ⋅   ⋅ mp ⋅   ⋅ D ⋅   mass velocity
 9   t  D
where:
= plastic moment capacity of the wall (= ¼ σy t )

2 dent
mp
δ = pipe deformation, dent depth
t = wall thickness (nominal)
σy = yield stress Figure 4 Dent prediction model (schematic).
D = steel outer diameter
DET NORSKE VERITAS

Table 4 Impact capacity and damage classification of steel pipelines and risers
Dent/ Impact Damage description Conditional probability2

Diameter energy
(%)1
D1 D2 D3 R0 R1 R2
<5 Eq. (3) Minor damage. 1.0 0 0 1.0 0 0

Major damage.
5 – 10 Eq. (3) 0.1 0.8 0.1 0.9 0.1 0
Leakage anticipated
Major damage.
10 – 15 Eq. (3) 0 0.75 0.25 0.75 0.2 0.05
Leakage and rupture anticipated.
Major damage.
15 – 20 Eq. (3) 0 0.25 0.75 0.25 0.5 0.25
Leakage and rupture anticipated.
> 20 Eq. (3) Rupture. 0 0.1 0.9 0.1 0.2 0.7
1 The energy limits for larger damage (i.e. 15 - 20%) should be carefully assessed as the energy levels might get unrealistic high.
2 For definition on damage categories (i.e. D1, D2, etc), see section 1.6.
4.4 Flexible pipeline Guidance note:
Note that neither calculations nor tests verify these levels, as tests
4.4.1 Impact scenario are normally performed up to minor damage only. Flexibles may
be conservatively assumed to have no capacity.
Unbonded flexible pipelines are typically built up of several
layers of reinforcement within layers of polymer. The actual - end - of - Guidance - note -
capacity will vary for similar pipes, which have only smaller
individual differences in design. No easy way of establishing 4.4.2 Pull-over/hooking scenario
the capacity exists, and the capacity should be determined for In general the pull-over/hooking scenario for a flexible
each individual pipe design. However, the impact capacity of pipeline is similar to that for steel pipelines. However, the
a flexible pipeline (or riser) is usually significantly less than flexible pipelines will then have a much larger final lateral
for a steel pipeline. If no other information exists the displacement and a smaller bending radius. The capacity
capacities given in Table 5 may be used as indicative values must be specifically determined or given by the
for impact capacity of 8”-10” flexible. manufacturer.
Table 5 Impact capacity and damage classification of flexible pipelines and risers
Impact energy2 Damage description Conditional probability1
D1 D2 D3 R0 R1 R2
Minor damage not leading to ingress
< 2.5 kJ 1.0 0 0 1.0 0 0
of seawater.
Damage needing repair.
2.5 – 10 kJ 0 0.50 0.50 0.50 0.50 0
Possible leakage.
10 – 20 kJ 0 0.25 0.75 0.25 0.25 0.5
Leakage or rupture.
> 20 kJ Rupture. 0 0 1.0 0.1 0.2 0.7
2 The capacities are given for 8-10 inch flexibles and should be adjusted for other dimensions. It is proposed to reduce the capacity by 25%
for 4-6 inch and increase the capacity by 25% for 12-14 inch lines.
DET NORSKE VERITAS

Page 16
For pull-over/hooking loads acting on umbilicals, capacities

4.5 Umbilical
as for flexibles may be applied.
Umbilicals are typically a complex compound of tubing,
electrical wires, reinforcement and protective layer. The most Normally, the only significant consequence of an umbilical
vulnerable parts of the umbilical are normally electrical breakage will be of an economic nature. It is assumed that
wires, and not the steel tubing. The weakest link in the loss of umbilical functions results in production stop (i.e.
umbilical should represent the capacity for the whole fail-safe principle). If this is not the case, then the
umbilical. The actual capacity should be determined for the environmental and human safety consequences of umbilical
specific design. However, if no other information is damage should also be evaluated.
available, the capacities given in Table 6 may be used.
Table 6 Impact capacity and damage classification of umbilicals

Impact energy3 Damage description Conditional probability1
D1 D2 D3 R0, R1 & R2
Minor damage not leading to ingress
< 2.5 kJ 1.0 0 0
of seawater.
2.5 – 5 kJ 0 0.50 0.50
Possible loss of function Note 2
5 – 10 kJ 0 0.25 0.75
Possible loss of function
> 10 kJ Loss of function 0 0 1.0
2 Not normally applicable, see section 6.1.
3 The given capacities are given for a reinforced umbilical. For umbilicals without reinforcement and for power cables, etc. the capacities
should be reduced.
The kinetic energy absorbed for two different cases may be

4.6 Different protection methods
expressed as given in equation (4) and (5) (Jensen, 1978).
4.6.1 Concrete coating Here, x0 denotes the penetration, b is the breadth of the
impacting object, h is the depth and D is the pipeline
Concrete coating may be used to shield pipelines from
diameter.
potential impact damage. The energy absorption in the
concrete coating is a function of the product of the penetrated For larger pipe diameters, equation (5) may give non-
volume and the crushing strength, Y, of the concrete. The conservative estimates and a denting shape more like
crushing strength is from 3 to 5 times the cube strength for equation (4) should be considered.
normal concrete density, and from 5 to 7 times the cube
strength for lightweight concrete (Jensen, 1978, 1983). The If no other information exists, energy absorption of 40 kJ
cube strength varies typical from 35 to 45 MPa. may be used for 45 mm normal density concrete coating
subject to a 30 mm wide indenting object.
Ek
E K = Y ⋅ b ⋅ h ⋅ x0 (4)
xo
bxh
4 D
EK = Y ⋅ b ⋅ D ⋅ x 03 (5)
3
xo
Figure 5 Impact in concrete coating.
DET NORSKE VERITAS

4.6.2 Polymer coating 2

Ep = ⋅ γ '⋅L ⋅ N γ ⋅ z 3 (7)
Polymer coating may be used to protect from potential 3
damage. Polymer coatings normally consist of a combination
of several layers of different thickness and material 2
Ep = ⋅ γ '⋅sγ ⋅ N γ ⋅ z 4 (8)
properties. Experimental results are necessary in order to 4
determine the potential absorption of energy for a given where sγ is a shape factor equal to 0.6, and L is the length of
coating. the impacting side.
If no other information exists the energy absorption The energy absorption of different objects is given in
capacities given in Table 7 may be used. Figure 6.
Table 7 Energy absorption in polymer coating
Type of coating Energy Energy Absorbed (kJ)
absorption 0 50 100 150 200 250
0.00
Corrosion coating with a thickness of 0 kJ 8"PIPE
maximum 3 – 6 mm. 0.25
Penetration (m)
12"PIPE
Thicker multi-layer coating 6-15 mm ~5 kJ 0.50

15"PIPE
(typical insulation coating 15-40 mm ~10 kJ 0.75 Conductor Pipe 30"
with varying thickness)
>40 mm ~15 kJ (not plugged)
Container - corner
1.00
Mechanical protection systems (e.g. Uraduct) 5 – 10 kJ Container - Side
1.25
If polymer coating is to be used as protection against specific 1.50

design loads, (i.e. trawl board impact loads) the protection
effect should be documented separately.
Figure 6 Absorbed energy in gravel
4.6.3 Gravel dump and natural backfill Energy absorption in natural back-filled sand is considerably
Gravel cover is the most common protection method for lower than for gravel. Natural back-filled sand is very loose,
pipelines. Based on full-scale tests the energy absorbed in the and pipes will not be plugged in sand. The impact resistance
gravel, when a falling pipe penetrates, can be described as: in back-filled sand can be assumed to be 2 - 10 % of the
gravel resistance.
E p =0.5 ⋅ γ '⋅D ⋅ N γ ⋅ A p ⋅ z + γ '⋅ z 2 ⋅ N q ⋅ A p (6)
Effective protection against dragged commercial ship
where: anchors can be obtained by burying the pipeline. The
required depth will depend on the size of the anchors of the
γ’ = effective unit weight of the fill material passing ships and the local soil conditions, i.e. how deep
D = diameter of a falling pipe anchors will penetrate.
Ap = plugged area of the falling pipe
z = penetration depth 4.6.4 Other protection methods
Nq , Nγ = bearing capacity coefficients Table 8 gives a short description of other protection methods
and the assumed lower bound impact capacity.
Guidance note:
The use of the plugged area of the pipe Ap must be seen in
relation to the size of the stones in the gravel. For small diameter
pipes compared with the diameter of the stones, the full cross
section of the pipe can be used. In cases where the stone/grain
size is small compared to the internal diameter of a penetrating
pipe, a equivalent area of the circumference multiplied with the
stone diameter may be used.
The bearing capacity coefficients can be chosen as Nq = 99

and Nγ = 137. Τhe effective unit weight is assumed to be 11
kN/m3. For other than non-tubular objects, like containers,
the energy absorption can become higher. The following two
equations are proposed for penetration with one of the side
edges and with one of the corners:
DET NORSKE VERITAS

Page 18
Table 8 Other protection methods

Method Description Impact
resistance
Concrete blankets are well
suited for low energy impacts
(e.g. trawl board impacts). In
general, individual cones of
Concrete concrete have only limited
impact capacity (in the order of 5 – 20 kJ
blankets 3 kJ), however several cones
may be activated during an
impact. Note that the stability of
such blankets need yo be
confirmed.
Sand bags are normally used to 5 – 10 kJ
Sand bags build artificial supports. Can be
used for protection. (assumed)
The bundle will act as an
effective protection against
impact loads. The energy
absorption can be calculated as
for a bare steel pipe, however Acc. to
Bundles the damage classification will equation (3)
be changed. The only critical
failure will normally be
leakage. Special attention
should be made to towheads
and to intermittent bulkheads.
Similar to bundles. Special Acc. to
Pipe-in-pipe attention should be made to equation (3)
intermittent bulkheads.
Tunnel structures are normally
introduced in order not to
Tunnel restrain pipeline movements.
structures, Varies,
Tunnel structures can be made
nearby normally at
up with a variety of geometry
protection least 50 kJ
and material. Thus almost any
structures required capacity level can be
obtained.
Trenching without backfilling
will have a positive but limited
effect against dropped objects,
ships sinking, etc, as these will
reduce the possibility to hit the
Trenching N.A.
pipeline/umbilical depending on
the width of the trench and the
size of the impacting object.
(i.e. only direct hits will be
accounted for)
DET NORSKE VERITAS

5 Failure frequency
Table 9 Frequencies for dropped objects into the sea
5.1 Introduction
Type of lift Frequency of dropped
In order to assess the pipeline/umbilical risk from accidental object into the sea
loading, it is necessary to establish the frequency of such (per lift)
event. The assessment can be approached deterministically
Ordinary lift to/from supply vessel 1.2·10-5
(quantitative) by considering frequency of exposure, drop with platform crane < 20 tonnes
frequency and probability of impact, or heuristically
Heavy lift to/from supply vessel 1.6·10-5
(qualitative) through the approach of generic data based on with the platform crane > 20 tonnes
operator experience.
Handling of load < 100 tonnes with
the lifting system in the drilling 2.2·10-5
The quantitative approach requires a significant amount of derrick
information regarding the field specific activities and the
Handling of BOP/load > 100 tonnes
system. This method is applicable to activities which are with the lifting system in the 1.5·10-3
regularly performed, e.g. crane activities, and where drilling derrick
operational experience exists.
For irregular activities, such as emergency anchoring, a more 5.2.2 Object excursion and hit probability
general evaluation may be the only means to assess the
frequency. The object excursion in water is extremely dependent on the
shape and weight of the object. Long slender objects, e.g.
The various input parameters are given in the following pipes, may experience an oscillating behaviour, see
sections. The procedure to establish the failure frequency for Aanesland (1987) and Figure 7, whereas massive, box-like
dropped objects from cranes is detailed described in objects will tend to fall more or less vertical.
Appendix A.
5.2 Crane activity

5.2.1 Drop probability
The drop frequency is based on the accident data issued by
the UK Department of Energy covering the period 1980-861
(DNV 1996b). During this period, 81 incidents with dropped
objects and 825 crane years are reported. The number of lifts
in the period was estimated to 3.7 million, which corresponds
to 4.500 lifts to/from vessel per crane per year. This gives a
dropped object probability of 2.2·10-5 per lift. For lifts above
20 tonnes the drop probability has been estimated to 3.0·10-5
per lift. The frequency is further split between fall onto deck
(~70%) or into the sea (~30%). Figure 7 Observed fall-patterns for dropped pipe joints
in water (Aanesland, 1987)
Lifts performed using the drilling derrick are assumed to fall
only in the sea, and with a dropped loads frequency as for The actual fall-pattern for a pipe is dependent on the entry
ordinary lifts with the platform cranes, i.e. angle into the sea, however patterns a), d) and e) in Figure 7,
2.2·10-5 per lift. are dominant and found for most entry angles.
The data show that the frequency of losing a BOP during The following values are recommended for use in
lowering to or lifting from a well is higher than for other calculations of the object excursion on the seabed. The object
typical crane lifts. A frequency of 1.5·10-3 per lowering or excursions on the seabed are assumed to be normal
lifting operation is proposed used (SikTec, 1992). For the last distributed with angular deviations given in Table 10.
part of the lift, when the BOP is directly above any vulnerable
parts, a significantly lower probability of a drop is assumed.
The proposed dropped object frequency is given Table 9. It is

possible to refine these estimates for given operations
considering the experience with individual crane types and
specific operating conditions. The annual frequency of a crane
or crane boom falling into the sea is from 4.4·10-7 to 6.7·10-7.
1
Detailed dropped object data are available for this period. No
more recent data are yet available in sufficient detail to be used
in this methodology.
DET NORSKE VERITAS

Page 20
The normal distribution is defined as: The breadth of each ring can be taken at 10 metre intervals.
The hit probabilities within each of these rings may then be
1 x 
2 calculated for different deviation angles and the actual sea
1 −  
2 δ  depth.
p( x) = e (9)
2π δ
Guidance note:
where: Special attention should be given to risers and in particular
vertical sections of risers. For risers, any vertical sections will
p(x) = Probability of a sinking object hitting the complicate the hit calculations. A way of calculating the
probability of hit to a riser is to:
sea bottom at a distance x from the
vertical line through the drop point. 1) Split the riser into different sections (i.e. normally into vertical
section(s) and horizontal section(s)), and
x = Horizontal distance at the sea bottom
(metres) 2) Calculate the hit probability of these sections. The final
probability is then found as the sum of all the probabilities for the
δ = Lateral deviation (metres), see Table 10 different sections.
and Figure 8.
Table 10 Angular deviation of object category.

Lsl Ar
no Description Weight Angular B+D
(tonnes) deviation (α)
(deg)
1 <2 15
2 Flat/long shaped 2–8 9
3 >8 5
4 <2 10 Drop-
point ri
1
5 Box/round shaped 2–8 5
ro
6 >8 3
7 Box/round shaped >> 8 2
1 A spread on the surface before the objects sinks is included.
Figure 9 Probability of hit within a ring, defined by

inner radius, ri, and outer radius, ro, from the drop point.
d α
Within a certain ring, the probability of hit to a pipeline or
umbilical with an object, Phit,sl,r, can be described as the
exposed area which gives a hit within a ring divided on the
total area of the ring, multiplied with the probability of hit
δ
within the ring, see equation (12).
Figure 8 Symbols used in eq. (9).
L sl ⋅ ( D + B / 2 + B / 2)
The probability that a sinking object will hit the seabed Phit , sl , r = Phit , r ⋅ (12)
Ar
within a distance r from the vertical line through the drop
point is then where:
r
Phit,sl,r = Probability of hit on subsea line (sl) within a
certain ring, r.
P( x ≤ r ) = ∫ p( x)dx (10)
−r Phit,r = Probability of hit within the ring, eq. (11).
The actual extent of the vulnerable items on the seabed, e.g. Lsl = Length of subsea line within the ring (m)
pipeline, within each ring can easily be incorporated by D = Diameter of subsea line (m), see Figure 10.
dividing the probability in several “rings”, see Figure 9. The B = Breadth of falling object (m), see Figure 10.
probability of hit within two circles around the drop point, Ar = Area within the ring (m2), see Figure 9.
Phit,r with inner radius ri and outer radius, ro, can be found by
Phit , r = P(ri < x ≤ ro ) = P( x ≤ ro ) − P( x ≤ ri ) (11)
DET NORSKE VERITAS

The effect of currents may be included if one dominant

current direction can be identified. This can be applicable for
rig operations over shorter periods, such as during drilling,
completion and intervention above subsea wells. However,
for a dropped object assessment on a fixed platform, seasonal
changes in current directions can be difficult to incorporate.
B Note also that the current mat change direction through the
water column for large water depths. If applicable, this
D should be accounted for.
The effect of currents should be considered when

B/2 D B/2
establishing a “safe distance” away from lifting activities.
Furthermore, a conservative object excursion should be
Figure 10 Definition of Hit Area determined, including also consideration of the drift of the
objects before sinking, uncertainties in the navigation of
For containers and massive objects, B can be set to the anchor handling vessel, etc.
average of the two shortest sides, and for tubular objects, B
can be set equal to the diameter for front impact and equal to
5.3 Energy calculation
the length for side impacts.
5.3.1 Kinetic energy
Guidance note:
By including the inclination of the tubulars the hit area will The kinetic energy of a dropped object depends on the mass
increase. However, including impacts from horizontal oriented and the velocity of the object. Furthermore, the velocity
tubulars, the capacity evaluations given in section 4 may be through the water depends on the shape of the object and the
conservative as they initially only consider knife edge loading. mass in water.
The terminal velocity is found when the object is in balance
Initially, one drop point per crane can be chosen. This is with respect to gravitation forces, displaced volume and flow
normally taken to be located between the loading zone for resistance. After approximately 50-100 metres, a sinking
the supply vessels and the lay-down area(s) on the platform. object will usually have reached its terminal velocity. When
Alternatively, several drop points may be used to describe the object has reached this balance, it falls with a constant
the crane activity in details. velocity, i.e. its terminal velocity. This can be expressed by
the following equation:
Pipes stacked and lifted together should be considered as one
lift, however the hit probability should be multiplied by the
number of pipes in the stack. (m −V ⋅ ρ water )⋅g = 1 ⋅ρ water ⋅C D ⋅ A⋅vT 2 (13)
2
5.2.3 Deep water applications where:
When considering object excursion in deep water, the
spreading of long/flat objects will increase down to m = mass of the object (kg)
approximately 180 metres depth. From 180 metres and g = Gravitation acceleration (9.81 m/s2)
further down the spreading does not increase significantly V = volume of the object (the volume of the
and may conservatively be set constant (Katteland and displaced water) (m3)
Øygarden, 1995). Note also that for deep waters, the ρwater = density of water (i.e. 1025 kg/m3)
spreading of objects on the seabed does not necessarily
CD = drag-coefficient of the object
follow the normal distribution, see Katteland and Øygarden,
(1995). A = projected area of the object in the flow-
direction (m2)
5.2.4 Effect of currents vT = terminal velocity through the water (m/s)
The effect of currents also becomes more pronounced in
deep water. The time for an object to reach the seabed will Guidance note:
increase as the depth increases. This means that any current For riser calculations, it should be noted that the terminal velocity
can increase the excursion (in one direction). At 1000 metres of objects hitting the riser close to the surface is hard to predict
depth, the excursion has been found to increase 10-25 metres The velocity could either be higher or lower than the terminal
for an average current velocity of 0.25 m/s and up to 200 velocity depending on the velocity the objects has as it hits the
surface and how the objects penetrate the surface, thus giving
metres for a current of 1.0 m/s (Katteland and Øygarden, higher or lower kinetic energy. In lieu of more detailed
1995). information, the objects can be assumed to have a velocity equal
to the terminal velocity at all depths below 50 metres and equal to
the velocity in a 30-metre drop in air for depths less than 50
metres.
DET NORSKE VERITAS

Page 22
The kinetic energy of the object, ET, at the terminal velocity Table 11 Drag coefficients
is:
Cat. no. Description Cd Ca
1 2
1,2,3 Slender shape 0.7 – 1.5 0.1 – 1.0
ET = ⋅m⋅vT (14) 4,5,6,7 Box shaped 1.2 – 1.3 0.6 – 1.5
2
All Misc. shapes 0.6 – 2.0 1.0 – 2.0
Combining these to equations gives the following expression (spherical to complex)
for the terminal energy:
It is recommended that a value of 1.0 initially be used for Cd,
m⋅g  m  after which the effect of a revised drag coefficient should be
ET = ⋅ − V  (15) evaluated.
C D ⋅ A  ρ water 
In addition to the terminal energy, the kinetic energy that is 5.3.3 Projected area
effective in an impact, EE, includes the energy of added For long-shaped objects, the projected area in the flow
hydrodynamic mass, EA. The added mass may become direction is assumed to equal the projected area of the objects
significant for large volume objects as containers. The when tilted at a certain angle. This means that the projected
effective impact energy becomes: area of a pipe is:
1 Apipe = L ⋅ D ⋅ sin x° (where xo ∈[0, 90] deg, measured from

E E = ET + E A = (m + m a ) ⋅ vT2 (16) the vertical)
2
where ma is the added mass (kg) found by ma = ρw· Ca ·V. As shown in Figure 7, a pipe will constantly change direction
when falling, and so the projected area will also change. A
Tubulars shall be assumed to be waterfilled unless it is uniform distribution of the angle should be used, or
documented that the closure is sufficiently effective during alternatively the angle may be taken as 45° for object
the initial impact with the surface, and that it will continue to categories 1, 2, and 3, respectively. Other objects are
stay closed in the sea. assumed to sink in such a way that the projected area equals
the smallest area of the object.
It should be noted that tubular objects experiencing a
oscillating behaviour will have constantly changing velocity, 5.3.4 Energy vs. conditional probabilities
and it has been observed that for 50% of the fall-time the
object have a velocity close to zero (Katteland and In lieu of accurate information, Table 12 may be used for
Øygarden, 1995). energy estimates. Table 12 gives a suggested split of the
object’s energy into energy bands with a conservative
conditional probability of occurrence. The division for the
5.3.2 Drag and added mass coefficients conditional probabilities is proposed for a pipeline with
The drag and added mass coefficients are dependent of the normal protection requirement, and a normal distribution of
geometry of the object. The drag coefficients will affect the the impact energies. For pipelines that are required to resist
objects terminal velocity of the object, whereas the added high impact energies and for which the share of objects that
mass has influence only as the object hits something and is give high impact energies is significant, a refinement of the
brought to a stop. Typical values are given in Table 11. energy groups in the upper range should be considered.
DET NORSKE VERITAS

Table 12 Conditional probabilities of impact energies (see notes)

Energy band (kJ)8
Description
< 50 50 - 100 100-200 200-400 400 - 800 > 800
1
< 2 tonnes 30% 18% 14% 12% 11% 15%
Flat/long
shaped 9 2 – 8 tonnes 2 5% 8% 15% 19% 25% 28%
> 8 tonnes 3 - - 10% 15% 30% 45%
< 2 tonnes 4 50% 30% 20% - - -
Box/round 5
shaped 2 – 8 tonnes - 20% 30% 40% 10% -
> 8 tonnes 6 - - - - 70% 30%
Box/round >> 8 tonnes 7 - - - - 30% 70%
shaped
1 The distribution is made based on the following assumptions:
Only (open) pipes included.
The objects weigh 0.5, 1.0 and 1.5 tonnes, with 1/3 of all objects within each weight.
The angle at the surface is assumed equally distributed from 0 – 90 degrees.
The terminal velocity is assumed linear from minimum to maximum for 0 and 90 degrees respectively.
The length of the pipes is approximately 12 m.
Only pipes included.
The object weight is assumed equally distributed from 2 to 8 tonnes.
The length of the pipes is approximately 12 m.
The object weights are assumed to be within 9 to 10 tonnes.
Only pipes included.
50% of the pipes have length of approximately 6 m, 50% have length ~12 m.
Objects considered:
The object weigh 0.5, 1.0 and 1.5 tonnes, with 1/3 of all objects within each weight.
Container, baskets (large volume, low density) (30%), velocity ~ 5 m/s
Equipment, e.g. (small volume, massive, high density) (70%), velocity ~10 m/s
The object weight is assumed equally distributed from 2 to 8 tonnes.
Objects considered:
container, baskets (large volume, low density) (70%), velocity ~5 m/s
equipment, e.g. (small volume, massive, high density) (30%), velocity ~10 m/s
The object weigh 10 to 12 tonnes.
Objects considered:
container, baskets (large volume, high density) (70%), velocity ~5 m/s
equipment, e.g. (medium volume, massive, high density) (30%), velocity ~10 m/s
The object weigh above 8 tonnes
equipment, e.g. (massive, high density), velocity ~5 to 10 m/s
8 Added mass is included.
9 For objects dropped from the derrick more objects will have a surface entry angle closer to 90 degrees.
5.3.5 Hit frequency vs. energy Fhit , sl , r = N lift ⋅ f lift ⋅ Phit , sl , r (17)
The frequency of hit can be estimated based on the number
of lifts, the drop frequency per lift and the probability of hit where:
to the exposed sections of the subsea lines. For a certain ring
around the drop point, the hit frequency is estimated by the
following:
DET NORSKE VERITAS

Page 24
Fhit,sl,r = frequency of hit to the subsea line within a N = Number of ships involved in a specific
certain ring (per year) activity potentially threatening the
Nlift = number of lifts installation/riser, i.e. passing ships in the
flift = frequency of drop per lift lane per year, arrivals to the platform per
year etc.
Phit,sl,r = probability of hit to a subsea line within a
certain ring, see equation (12) P1 = Probability of being on collision course, i.e.
probability of being on collision per pass for
passing ships in the lane, geometric
The total frequency of hit to a subsea line is assessed by probability of hitting the platform for ships
summarising the hit frequencies to the pipeline within each during waiting in the safety zone (normally
ring around the drop point. downwind of the installation) etc.
P2 = Probability of loss of control or faulty
Finally, within each of the capacity energy regions, see navigation onboard the ship.
section 4, the frequency is added up and given a ranking as P3 = Probability of failure to warn or divert a ship
proposed in section 5.8. on collision course, or ship “recovery” from
its errant state. The cause for this may be
5.4 Ship traffic absence from the bridge, absorbed in other
activity, accident, asleep, alcohol or radar
5.4.1 Introduction failure.
Risers may be subject to potential interference with ships and Priser = Probability of hitting the riser given a hit
ship collisions with riser should be determined to decide; with the platform. This probability may be
found by geometrical evaluations of the
− whether to locate riser inside or outside a jacket, platform and the riser.
− whether a J-tube or caisson protection is needed, or
− the location of the riser versus loading operations.
If the last probability (riser collision frequency given a hit
Damage to riser from ship collisions that do not impair the with the platform, Priser) in equation (18) is omitted, then the
platform integrity but may be of consequence to the riser result will be the probability of hitting just the platform. Priser
should be evaluated to ensure that the riser is adequately is further explained in sections 5.4.2.1 to 5.4.2.4. Different
protected. scenarios are also described in these sections.
Different methods are used to calculate the collision Ship collision damage to the riser can be due to collision
frequency for different vessel types. It is not the type of between the riser and:
vessel, but the way the vessels traffic the area around the
installation that influences the selection of the calculation 1) passing vessels; merchant vessel or a supply vessel to
method. other fields;
2) shuttle tanker approaching the platform field;
An assessment of the frequency and the associated kinetic 3) fishing vessel;
energy of ship collisions damaging the riser must be based 4) standby vessel;
on ship traffic data, type of vessels and geometric 5) a supply vessel to the current field.
evaluations.
Any of these scenarios can occur while the vessel is:
The procedure for estimating the frequency of collision,
FColl_Riser, between a riser at the installation and a vessel is − powered, or
described by the equation: − drifting.
The last scenario (i.e. supply vessel) can also occur while the
FColl _ Riser = N ⋅ P1 ⋅ P2 ⋅P 3 ⋅Priser (18) supply vessel is:
where:
− waiting to load/unload in the vicinity of the platform
− loading or unloading
DET NORSKE VERITAS

5.4.2 Calculation of the different collision probabilities N = Number of ships passing in the ship lane per
The different probabilities presented in equation (18) must be year.
calculated with regard to the specific scenarios 1-5 listed in P1 = Probability of being on collision course per
section 5.4.1. The basic principles for these calculations are pass.
described in the subsequent sections. As the riser will P2 = Probability of loss of control onboard the
represent only a fraction of the platform, the probability for ship, when on collision course per pass,
hitting the riser will be smaller than hitting the platform. The typically specified by a minimum time
probability of hitting the riser given a hit on the platform, period of 20 min.
Priser, must be based on geometrical evaluations of the P3 = Probability of failure to warn or divert a ship
installed riser. on collision course, or ship “recovery” from
its errant state.
Guidance note:
Priser = Probability of hitting the riser given a hit
Geometrical evaluations include the riser location, size and
configuration. For instance, flexible risers will normally have a with the platform.
steeper path down to the seabed compared with metallic catenary
risers. This means that a larger section of the metallic catenary
riser is exposed to vessel impact (from a specific direction). P1 is often called “geometric collision probability”. Merchant
Further, the effect of shielding should be accounted for and
effects like the vessel may hit another installation or a bridge vessels will usually sail in dedicated lanes during passage
between two installations and thus the hit energy may be reduced from one destination to another. The location of the ships
and the course may be changed. within these lanes is assumed to be normal distributed. This
- end - of - Guidance - note - is illustrated in Figure 11. P1 is given by:
Given a hit on the riser, the result may be a leak or full bore 1 x 
2
rupture, but the extent of the damage to the riser is also 1 −  

2 δ 
P1 = D ⋅ ⋅e (19)
dependent on the type of protection, if any. 2πδ
The method described in the following sections is based on a where:
collision example with a riser running eastwards from the
underside of a platform, see Figure 11. It must be stressed D = Collision diameter = Wa + Bvessel, where Wa is
that the calculation for this set-up will be valid only for this apparent platform width and Bvessel is ship beam.
particular configuration and adaptations to other studies and δ = Standard deviation (normally given together
configurations should be done only after careful evaluation. with the ship lanes)
The overall frequency of collision with the riser is found by x = Distance from centre of lane to the installation
adding together the frequencies for the different scenarios as
described in the following sections.
With respect to P2 , there are normally six different reasons
why a vessel will continue on a course towards an
5.4.2.1 Collision calculations for passing vessels installation. These are:
Merchant vessel routes will pass in dedicated lanes
depending on the destination. This will also apply to shuttle − absence of crew on bridge
tankers to other installations. Vessel routes outside 10 nm − crew absorbed in other tasks
will normally give negligible contribution to the collision − crew asleep
risk. − accident
− alcohol/drug abuse
Calculations must be performed for each vessel route and − radar failure/poor visibility
then the results are summed to find the total frequency of hits
from the passing vessels. Collisions between offshore P2 is normally set to 2⋅10-4, which is confirmed by Fujii et.
installations and ships under power, running in a distinct al. (1974, 1984) and Solem (1980).
direction, are described by the equation (18) where the
different variables will be: P3 is dependent on contingency measures on the installation.
Aspects that will decide this value are:
– standby vessel always stationed near the installation

– fog horns and navigational aid systems installed at the
platform
− RACON (RAdar beaCON), see section 5.4.5, installed
Fog horns and strobelights are mounted on nearly all

offshore installations in the North Sea, and do not influence
the initial probability P3.
DET NORSKE VERITAS

Page 26
In case of an errant vessel on collision course, a standby P3 is normally set to 1.0 without a standby vessel present and
vessel, if present, will go towards the errant vessel and use 0.14 with a standby vessel present. If RACON is installed P3
light and sound to alert the vessel. A standby vessel will also will be 0.9 without a standby vessel present. If both RACON
be able to identify the errant ship and therefore the effect of and a standby vessel are present the probability will be 0.13,
radio calls will be significant. (Fujii, et. al., 1984).
W
N g
i p pin
Sh e
S
E lan
Probability distribution of
ships across the
shipping lane
Platform
Riser
on
l l isi ter
Co a m e
di
Fd=
probability of Cross section where a
finding a ship in vessel could hit the
this cross section platform if it does not
keep an effective watch
Figure 11 Normal distributed geometric collision probability
A geometrical evaluation of the probability of impact with

Priser, the probability of hitting the riser given a hit on the the riser, given collision with the platform, is given by:
platform, is calculated by assuming that there is a
relationship between the probability of hitting the installation ( L + Bvessel ) ⋅ α
and the probability of hitting the riser. Priser _ i = (20)
Wa + Bvessel
If, for instance, a riser is connected to the east side of the
where:
platform and running eastwards, the probability of hitting the
riser may be equal from north and south side. The probability i = North, south, east, or west.
of hitting the riser will be lower from the west side of the
L = Exposed width of riser.
installation, because the exposed area of the riser will be
smaller and the platform structure will hinder the vessels Wa = Platform width of the current side at sea
from reaching the riser. From the east side, the supporting level.
structure of the platform will not have any influence on the α = Reduction factor depending on support
probability of impact. For all cases (north, south, east and structure interference
west) the exposed area will be small compared to the Bvessel = Width of vessel
platform.
DET NORSKE VERITAS

The width of the vessel, Bvessel, is added to the diameter of the The basis for equation (23) is:
installation as the vessel must pass a minimum of half of the
vessel width on either side of the installation to avoid a – The term 365 ⋅ 24 ⋅ V gives the total distance covered by
collision. a vessel travelling at its transit speed normalised to 1
year.
The water depth where the riser is vulnerable to a ship – Multiplying by vessel density gives the total distance
depends on the ship type, but a water depth of at least 5 covered by all vessels in the vicinity of the platform.
metres should be considered as a vulnerable section. – Multiplying by the platform diameter gives the fraction
of those vessels heading towards the platform.
Equation (20) describes the geometrical relationship between – The terms P2 and P3 are equivalent to those given in
riser and platform for a vessel coming from a particular equation (18).
direction. To account for the four directions, north, east, − Priser is calculated according to equation (20) in section
south and west, it is necessary to summarise the geometrical 5.4.2.1.
relationship for all the directions before this is multiplied
with the frequency of hitting the platform. The frequency of 5.4.2.3 Collision risk of standby vessels
hitting only the platform is described by the equation:
For standby vessel, only drifting collision is normally
included. The vessel does not move as a vessel that passes or
FHit _ Platform = N ⋅ P1 ⋅ P2 ⋅ P3 (21) visits the installation. If the vessel is loading/unloading from
an installation, it will act as a supply vessel, and the risk
If the probability of hitting the platform is assumed to be
should be included in supply vessel collisions.
equal for each side (this may not always be the case), the
total frequency of hitting the riser will be: A standby supply vessel has redundant machinery. The
frequency of machinery breakdowns should thus be
  somewhat lower than the frequency for vessels with one
FColl _ Riser = FHit _ Platform ×  14

∑ P  j (22) engine. On most supply vessels, the two redundant engines
j = Riser _ i normally have several minor machinery systems that are
common for both engines. The risk reducing effect is thus
Riser_i is the geometrical relationship between the platform
assessed to be 30 % by DNV (1998). For single engine
and the riser in each direction, e.g. north, east, south and
tankers operating in the North Sea, the machinery breakdown
west, ref. equation (20).
frequency is 2.0⋅10-5 per hour (DNV, 1998). A typical
machinery breakdown frequency for supply vessels is thus
5.4.2.2 Collision calculations for random distributed
1.4⋅10-5 per hour. This frequency corresponds to a machinery
vessels
breakdown of certain duration. For most breakdowns, the
For ships that are distributed randomly near the installation machinery will be started after only few minutes and hence
and running in random directions, as is typical for fishing these breakdowns are not included in the frequencies given
activities, the frequency per year of collision with the riser above.
may be calculated as (Technica, 1987):
A standby-vessel will normally be situated close to the
FColl _ Riser = (365 ⋅ 24 ⋅ V ⋅ D ⋅ ρ ) ⋅ P2 ⋅ P3 ⋅ Priser installation. It is conservatively assumed that the vessel
(23) moves independently of the weather conditions, and thus has
where: equal probability for drifting in all directions. This is a
conservative assumption, as a standby-vessel without a
V = ship speed [km/h]. special duty normally will be downstream of the installation.
D = collision diameter of installation [km].
The annual frequency for a standby vessel collision with the
ρ = density of ships [per square km] riser may be expressed by the following equation:
P2 = Probability of loss of control onboard the
ship for a specific minimum of time period Fcoll , wait = N ⋅ P1 ⋅ ( P2 ⋅ t ) ⋅P 3 ⋅Priser (24)
(20 minutes) will normally have the same
value as in section 5.4.2.1, (Fujii et. al., 1974 where:
and 1984 and Solem, 1980)
P3 = Probability of failure of warning or diverting N = Number of standby vessels (per year),
a ship on collision course, either by normally one.
contingency measures effected on the P1 = Geometric probability of hitting the
platform, or on the approaching vessel. Will platform, D/(2πR).
normally have the same value as in section D = Typical diameter of installation, plus the
5.4.2.1. average of the width and length of a typical
Priser = Probability of hitting the riser given a hit ship [m], Wa + ½(Bvessel+Lvessel).
with the platform. R = Radius of stand by zone ( normally 1 km).
P2 = Frequency of machinery breakdown per
DET NORSKE VERITAS

Page 28
hour (typically 1.4⋅10-5 per hour). The probability of loss of control onboard the supply vessel
T = hours per year for vessel to be in the vicinity given collision course is found to be P2 = 2.7⋅10-6 per
of the platform (8760 hrs for a whole year) approach, based on data from Technica (1987). The
probability is significantly lower than for merchant vessels,
P3 = Probability of failure to correct the situation. as the crew onboard a supply vessel approaching an
(Normally taken as 1, as machinery installation is aware that the installation exists. The crew on a
breakdowns included in P2 need longer supply vessel is thus likely to be more observant than the
repair time than available) crew on a passing merchant vessel.
Priser = Probability of hitting the riser given a hit
with the platform, given by equation (20). For P3 , the contingency arrangements described in section
A typical standby vessel is a supply vessel with length of 80 5.4.2.1 are also valid for the supply vessels sailing to/from
metres and displacement of 5000 tons. The energy of such a other installations (scenario B)
vessel is dependent on the drifting speed. The drifting speed
Assessments of supply vessels sailing to other installations
is normally about 3-5 % of the wind speed. The maximum
should also be taken into account.
velocity when a vessel is situated upwind for the installation
is assumed to be hurricane, 32.6 m/s. The maximum kinetic Low energy impacts (scenario C)
energy for a drifting supply vessel is thus 10 MJ.
Collision can occur during loading and unloading of the
5.4.2.4 Collision risk of supply vessel supply vessel. Low energy collision during loading or
unloading will follow the same methodology as described in
Collision calculations between supply vessel and installation
the previous section 5.4.2.3. Supply vessels are designed for
normally include the following scenarios:
several different operations, and have large power compared
A) collision with supply vessel that approaches the to size. During poor manoeuvring or in bad weather
conditions, the vessel can hit the installation during loading
installation;
and unloading.
B) collision with passing supply vessel that is sailing
to/from other installations; A technical failure will lead to only a relatively slow drifting
C) collision with drifting supply vessel that hits the of the vessel into the installation and hence this scenario will
installation during loading/unloading or similar not cause impact energies large enough to threaten the
operations. integrity of the platform structure, but can cause damage to
the riser.
A) and B) are high-energy collisions, and C) is a low energy
collision. Shuttle tankers near to the installation will also be P1_i may be expressed as the probability of hitting a specific
included in these categories, but the tonnage will be side, i, of the platform with regard to the wind directions
significantly larger. given a technical failure or faulty manoeuvring. P1 will
therefore be the sum of the probabilities of hitting each side
The supply vessel activity will depend on the activity at the of the platform. Operations in winds exceeding a certain
platform, i.e. start-up, normal operation etc. wind force will normally be cancelled.
The total frequency is calculated according to equation (18), P2 will represent technical failure and faulty manoeuvring.
for which the input is described below. The normal failure rate of a single ship engine failure is
2⋅10-5 per hour (Technica, 1987). Modern supply vessels will
High energy impacts (scenario A and B) have a lower probability of engine failure.
Modern navigational systems and procedures will ensure that During loading/unloading, the close location and short time
the installation is not used as the final navigational target, from incident to a possible impact means that prevention of a
and the probability for a collision course is limited (scenario failure situation cannot be expected (P3=1).
A). Based on experience from similar studies of fixed
installations, it is estimated that 10 % of the vessel In lieu of accurate data, the probability for low impact
approaching the installation is on collision course, which collisions may be found with generic data. A generic
collision probability of 6.0⋅10-4 per visit can be used for
gives P1 = 0.1. This value is somewhat high as the process
impacts with steel jackets (J.P. Kenny, 1998). Note that this
for selection of final navigational target outside the
probability constitutes the product of P1, P2 and P3. It is
installation is relatively new. If the supply vessels used the assumed that the frequency for collision with other platform
installation as final navigational target, the probability to be types is about the same.
on collision course would be 1.0.
Maximum manoeuvring speed for supply vessels is normally
given as 2.8 m/s. Since the collisions are most likely in the
longitudinal direction, an added-mass coefficient of 1.1
(10%) (DNV, 1988) is chosen. For supply vessels with 5000
tons displacement, the maximum collision energy during
loading and unloading is 22 MJ.
DET NORSKE VERITAS

The probability, Priser, of hitting the riser, given a hit on the Guidance note:
platform while loading/unloading, is calculated following a The corresponding kinetic energy for a vessel of 2500 tonnes
different procedure than that used for passing vessels. Supply with a velocity of 4 knots will then be:
vessels are located stationary close to the platform while Bow and stern impacts:
loading/unloading and are normally positioned upwind of the E = 1/2⋅(1.1⋅2.5⋅106) ⋅(4⋅0.514)2 = 5.8 MJ
platform with the bow in the wind direction. Side impacts:
E = 1/2⋅(1.4⋅2.5⋅106) ⋅(4⋅0.514)2 = 7.4 MJ
Guidance note:
Figure 12 shows a typical situation when loading/unloading. The For collisions with the platform, the vessel itself may absorb
probability of hitting the platform will be in a 180° sector some of the impact energy. For riser collisions this will normally
dependent on the wind direction. The exposed area with a not be the case.
probability of hitting the catenary riser will be smaller, i.e. 20° in
this example, ref. Figure 12 (In the figure the centre of the vessel - end - of - Guidance - note -
have been used giving an additional width of half the vessel
breadth so that the vessel can pass). The probability of hitting the
riser will then be a fraction of : 20/180 = 0.11 of the probability
of hitting the platform. (It is here assumed that the exposed riesr 5.4.4 Total collision frequencies
area is close to the surface and therfore may be hit by a ship)
To find the total frequency of collision between the riser and
ship traffic, all the frequencies from different type of vessel
activity, as described in the previous sections, can be
presented in a tabular form according to impact energy. As
Su sse
pp l
ve
an unprotected riser probably will experience a rupture when

ly
Loading Exposed hit by a vessel, the necessity of dividing the probabilities into
area riser area
different energy classes may be discussed.
1/
5.4.5 Risk reducing measures
2
Bvessel Riser
The most important overall risk reduction measure is to
Platform shaft
avoid a collision with the platform. Further, for impacts with
relatively low kinetic energies, i.e. 0 – 15MJ, installation of a
collision net will reduce the probabilities of hitting a riser.
Impacts during loading/unloading can therefore be reduced.
However, as this activity will only contribute to some of the
Figure 12 Catenary riser area exposed to supply vessel total probability of an impact with a riser, the cost benefit
collision during loading/unloading. effect should be considered. Overall, the design of a riser and
- end - of - Guidance - note - the location relative to the platform will be an important
consideration.
5.4.3 Impact calculation methodology
The vessel types that can hit the platform represent different Measures that will decrease the failure risk of hitting the
weight categories and velocities, giving different hit platform, P3, are:
energies. A division into different kinetic energies and vessel
types is therefore relevant. Risers are normally very fragile, − RACON (RAdar beaCON): A device emitting a strong
and a collision with a vessel will most likely result in severe pulse when triggered by a nearby ship radar. This makes
damage or rupture. the installation easy to identify on the ship radars.
RACON is assumed to reduce the P3 for all vessel traffic
For a direct hit or impact, the kinetic energy is given by the except supply vessels at low speed in the vicinity of the
following equation: platform.
− RADAR (ARPA): A radar with a competent operator
and 24 hours watch where all ships are plotted and
1
E= ⋅ (M + a ) ⋅ V 2 (25) monitored when closer than a predetermined distance,
2 typically 12 nm.
where: − Assignment of standby vessel: A dedicated standby
vessel is assumed to reduce P3 for all vessels except the
M = Displacement (kg) supply vessels of low speed in the vicinity of the
a = Hydrodynamic added mass (kg), for bow platform. The standby vessel will take action in
and stern impact it is 10% of the situations in which a vessel on collision course is for
displacement and for sideways impact it is instance 5 nm from the complex, and will give
40% of the displacement with drifting information on course, speed and size of the errant ship.
vessels (DNV, 1988)
V = Ship speed (m/s)
DET NORSKE VERITAS

Page 30
5.5 Simultaneous operations Further, a possible manoeuvring failure of the service

Simultaneous operations are defined as work activities vessels, which are handling the anchors during the anchoring
performed on a well or a subsea installation while production operations, may cause an anchor to be dropped. If the service
continues through the pipeline. The failure frequency should vessel is located above one pipeline, this can be hit. Safe
be established based on the whole operation and not isolated distances to pipelines should be ensured during anchor
sub-operations. Previous operator experience and generic handling
failure data will be the basis for frequency estimation.
The typical weight of a rig anchor is 12 tonnes. If an anchor
The methodology applied in this recommended practice is is dropped during the lowering operation, the anchor may
not suitable for estimating the risk for an accident during have a kinetic energy in the range of above 800 kJ. The
critical, isolated operations such as BOP installation. The kinetic energy of a dropped anchor chain will be in the order
risk of such operations should be controlled by other of 1-5% of the kinetic energy of the anchor.
methods such as HAZOP, although it should be noted that
such worst-case evaluations normally establish only the 5.7.2 Dragged rig anchor
consequence of an event and not the frequency.
If more than one of the anchor chains breaks, the rig may
drift off and there is a risk of impact to the flowlines by
5.6 Trawling dragged anchor chains. According to Worldwide Offshore
Trawling activity is usually concentrated in certain areas. If Accident Databank (DNV, 1996b), the statistical frequency
pipelines and umbilicals are routed in such areas the annual of drifting rig is 6.4·10-3 per rig year.
frequency of a trawl board hit will normally be very high,
e.g. from 10-2 up to 100 per km and year. The failure Guidance note:
frequency of the same order as the hit frequency unless the This corresponds to a frequency of drifting rig of 1.2·10-3 during
pipelines and umbilicals are protected against trawling. a drilling and completion operation (total duration 70 days). The
frequency of a failure in the pipeline or umbilical due to a
If a pipeline is designed to withstand trawling, then the dragged anchor will be less than drifting rig frequency depending
on the anchor area relative to the pipeline or umbilical route.
failure frequency is negligible (i.e. only minor damage to the
protection). If not already designed, larger diameter pipelines - end - of - Guidance - note -
(i.e. larger than 12”-14”) may be protected by coating to
reduce the failure frequency. Smaller diameter pipelines, 5.7.3 General shipping
flexibles and umbilicals should be trenched, gravel dumped,
etc. Emergency anchoring due to drifting ship can represent a
risk to subsea installations, where potential hazards are
Reference is made to the DNV Guideline 13 (1997) for related to dropped anchors and dragged anchor/anchor chain.
pipeline design against trawl interaction. Both shuttle tankers, supply vessels and commercial ships
may come into a drifting situation. A stand-by vessel can
5.7 Anchor handling usually change the drifting course of a ship.
5.7.1 Rig operations The mass of an anchor is typically 10 tonnes for a shuttle
tanker and 2 tonnes for a supply ship. Typical reasons for
A rig entering a new location and performing rig anchor dropped anchor during an emergency situation are human
handling poses a risk of external impact to pipelines and
error during the anchoring operation, failure of the chain
umbilicals. There is a risk related to a anchor chain falling
braking system or loss of the power supply to the chain
onto a pipeline/umbilical or a drifting rig dragging an anchor
over a pipeline/umbilical. braking system.
A rig is normally be moored with eight anchors. Pipelines Dependent on the mass of the chain and the dragging length,
and umbilicals may cross below the anchor chain. An anchor a dragged anchor chain can endanger pipelines and
chain that breaks may hit one pipeline or umbilical umbilicals (i.e. abrasion of protection and pipe wall) in
depending on the breaking point and on pipeline/umbilical addition to the more dramatic hooking scenario.
route relative to the anchor chains.
The risk of emergency anchoring from shuttle tankers is
It is proposed to assume a frequency of 0.01 breakage per generally low. Shuttle tankers are provided with a dynamic
year per anchor chain (DNV, 1997b). This is based on positioning system and the redundancy of the machinery is
known anchor breakage events up to 1993 for offshore rigs high. The likelihood of machinery failure is consequently
and production vessels. lower for shuttle tankers than for other ships. Furthermore, it
should be noted that loading of shuttle tankers is weather
Guidance note: restricted, i.e. the tankers will usually stay at a safe distance
The total duration of a drilling and completion operation is about from the installations during bad weather conditions.
70 days, giving a frequency of 0.002 for breakage of one of the
anchor chains during drilling and completion. The frequency for
permanently moored platforms should be set individually,
however it is assumed to be lower than the above.
DET NORSKE VERITAS

Commercial shipping routes should also be evaluated to 5.8 Frequency ranking

establish a relevant frequency of emergency anchoring
Both a quantitative and qualitative evaluation may be used
hitting the pipeline. For the relevant shipping lane(s) the
for a total evaluation of the pipeline protection effectiveness.
pipeline is crossing the vessel size/class distribution should
In order to compare the frequency and risk of any of the
be established. Given the vessel class typical anchor size is
relevant hazards, an individual ranking from 1 (low
given and thereby anchor size distribution may be
frequency) to 5 (high frequency) is proposed, see Table 13.
established based on the vessel distribution. For the different
Note, however, that the limits given in Table 13 may be
anchor sizes the seabed penetration may be established for
adjusted to comply with case specific requirements.
the local soil condition. By combining a generic frequency of
emergency anchoring in the area of interest and the The loading frequency is combined with the damage
conditional frequency of anchor penetration, the required evaluation to derive at the failure frequency.
trenching depth of the pipeline can be established to satisfy
the acceptance criteria. Note that the failure frequencies are given for the whole
pipeline and as such the length of the pipeline shall not be
Guidance note: decisive for the total failure frequency of the pipeline.
Commercial ships normally uses stockless anchors and the anchor
size is determined based on the ships equipment number. The
equipment number is a function of the ship displacement, the
breadth, the freeboard and the profile area.
Table 13 Annual failure frequency ranking for one pipeline/umbilical

Category Description Annual
frequency
1 So low frequency that event considered negligible. <10-5
(low)
2 Event rarely expected to occur. 10-4 > 10-5
3 Event individually not expected to happen, but when summarised over a large 10-3 > 10-4
(medium) number of pipelines have the credibility to happen once a year.
4 Event individually may be expected to occur during the lifetime of the pipeline. 10-2 > 10-3
(Typically a 100 year storm)
5 Event individually may be expected to occur more than once during lifetime. >10-2
(high)
DET NORSKE VERITAS

Page 32
6 Consequence – personnel involved in work on the company’s facilities

(1st party),
6.1 Introduction − personnel outside the company’s facilities who could be
affected by the company’s activities (3rd party).
Potential consequences of accidental events to pipelines and
umbilicals must be established with consideration human There is usually very little human activity in the vicinity of
safety, economic loss and environment impacts. Table 14 pipelines. Pipeline releases at the platform approach or near
presents a matrix for identifying of potential consequences subsea structures may have consequences for 1st party
for damage to pipelines and umbilicals. personnel on a platform or rig. In the pipeline mid-line zone,
releases can endanger 3rd party personnel.
Table 14 Identifying potential consequences for pipeline
and umbilical damage Only major release scenarios (i.e. category R2) from
pipelines conveying gas can endanger personnel. A gas cloud
Pipeline Human safety Environmental Material
contents impact damage nearby the platform or the rig can be ignited resulting in a
ball of fire or an explosion. Ignition will only occur if the gas
Gas Relevant Normally not Relevant above the sea surface is of flammable concentration and
relevant4 possible ignition sources are present within this cloud.
Condensate Relevant Relevant1 Relevant
Oil Relevant Relevant Relevant The size and distribution of a gas cloud from a subsea
Water Normally not Relevant 5
Relevant pipeline release will be influenced by the depth, currents and
relevant prevailing winds. In addition, the composition of the gas will
Umbilical Normally not Normally not Relevant influence the cloud formation, as rich gas may form a cloud
relevant2 relevant2,3 that does not rise but extend over a large area, whereas dry
1 Condensate normally disperses / evaporates quicker than oil. gas will rise rapidly. It is often difficult to accurately predict
During storm conditions the condensate can be gone within the outcome of such events, although it is possible to
hours. This means that leakage from a condensate pipeline is establish critical zones with major potential for harm to life.
less likely to give significant environmental consequence to the
environment. In major release events, it may be assumed in 1-10 % of
2 Damage to an umbilical will normally not cause any these events the gas release will ignite and a large number of
consequence for humans or the environment. However, safety persons onboard the rig or the platform will be exposed.
and environment should be considered if damage to an
umbilical leads to failure in the subsea installation which in turn The following scenarios have potential for endangering 3rd
leads to a release. party personnel:
3 Release of fluids from an umbilical will normally be a small
amount and can normally be neglected. − emergency anchoring,
4 Gas release can result in pollution if the gas contains injected
chemicals or releases H2S dissolving into the water. − pipe laying (when laying parallel pipes, damage to
5 The water may be processed water which contains substances installed and producing pipelines can have potential
dangerous to the environment. impact on barge personnel).
The consequences for human safety may be classified as

shown in Table 15. Note that for the proposed ranking
6.2 Human safety category 2 and 4 are not used for human safety consequence
The human safety consequence of pipeline or umbilical ranking.
failure should be established with regard to:
Table 15 Safety consequence ranking

Category Description
1 (low) No person(s) are injured.
2 (not used)
3 (medium) Serious injury, one fatality (working accident)
4 (not used)
5 (high) More than one fatality (gas cloud ignition)
DET NORSKE VERITAS

6.3 Release to the environment – the amount and type of spillage;

– the weather conditions, including wave heights, wind
Environmental consequences should be established both for
and current speed;
minor and for major release scenarios (i.e. R1 and R2). The
environmental consequence of any leakage from damaged − time to reach and amount to arrive at sensitive areas.
pipelines should consider polluting impacts on: Environmental consequences are normally expressed as
estimated time to achieve full recovery of the affected
− eco-system in the water, including seabed vegetation,
populations/areas. This will include evaluation of the
plankton, fish and sea mammals such as whales and
different species’ vulnerability to oil spillage, the
seals;
effectiveness of the oil spillage preparedness measures in the
− coastal environment, including beaches and coastal
area, etc.
regions that either have great value as refuge for birds or
contain extraordinary vegetation; An environmental consequence assessment of spillage as
− sea birds, comprising birds living, mating or on passage outlined above is both complex and time consuming. A much
in the area; more general evaluation may be made by considering only
− fish in fish farms and related industries in the area. the amount of release and relating this to the annual
allowable spillage amounts in the acceptance criteria. This
The environmental impact on the above are dependent on will implicitly account for the impacts on the environment.
the: The amount categorisation given in Table 16 may be used as
guidance.
Table 16 Spillage ranking

Category Description Amount of release
1 Non, small or insignificant on the environment. Either due to no release of internal ~0
(low) medium or only insignificant release.
2 Minor release of polluting media. The released media will decompose or be <1000 tonnes
neutralised rapidly by air or seawater.
3 Moderate release of polluting medium. The released media will use some time to <10000 tonnes
(medium) decompose or neutralise by air or seawater, or can easily be removed.
4 Large release of polluting medium which can be removed, or will after some time <100000 tonnes
decompose or be neutralised by air or seawater.
5 Large release of high polluting medium which can not be removed and will use > 100000 tonnes
(high) long time to decompose or be neutralised by air or seawater.
In general, repairing offshore pipelines is a time consuming

6.4 Economic loss
affair. The work will normally take approximately one to
The economic consequence of any damage to pipelines can three months to complete, as all work is performed subsea.
be classified with respect to the delay in production from a The actual duration is however strongly dependent on time to
pipeline. The cost of production delay normally exceeds the mobilise, the efficiency of repair systems and the weather
actual cost of repairing the damage. However, both the cost conditions. Typical repair operations that are planned prior to
of repairing and the cost of any delay in production delivery failure occurring are expected to take shorter time than the
from affected fields must be included in the evaluation. above estimate, whereas complex repair operations, e.g.
bundle repair, are anticipated to take longer time.
The economic consequences may be classified as stated in
Table 17. It should be noted that variations between different Any potentially critical elements with respect to upholding
projects can change the limits stated. Alternatively, the actual the platform production (e.g. water injection lines,
cost for production delay and repair may be used in the cost- umbilicals) should be identified.
benefit evaluations of the proposed protection design, and
would affect the expression in favour of additional risk For umbilicals, only economic damage classification is
reduction measures, see also section 2.7. normally relevant, as the tubing typically contains only a
small amount of toxic liquids and will not normally endanger
human safety.
DET NORSKE VERITAS

Page 34
Table 17 Economic consequence ranking

Category Description Production
delay/
Downtime
1 Insignificant effect on operation, small or insignificant cost of repair 0 days
(low)
2 Repair can be deferred until scheduled shutdown, some repair costs will occur. <1 month
3 Failure causes extended unscheduled loss of facility or system and significant

repair costs. Rectification requires unscheduled underwater operation with pre- 1-3 months
(medium)
qualified repair system before further production.
Failure causes indefinite shutdown and significant facility or system failure costs.
Rectification requires unscheduled underwater operation without pre-qualified
repair system before further production.
4 3-12 months
Or
Failures resulting in shorter periods of shut down of major parts of (or all of) the
hydrocarbon production for the field.
Total loss of pipeline and possible also loss of other structural parts of the
platform. Large cost of repair including long time of shut down of production.
5
Or 1-3 years
(high)
Failures resulting in shut down of the total hydrocarbon production for a longer
period.
DET NORSKE VERITAS

7 Risk assessment Consequence
1 2 3 4 5
7.1 General
Frequency of occurrence
The final risk assessment consists of coupling the relevant
5 Not acceptable
frequency rankings with the consequence rankings and then
comparing the result against the acceptance criteria. 4 ALARP region
Figure 13 gives an example, where the dark shaded areas 3
indicate the defined total acceptance criteria where additional Acceptable
2
protection is required, see also section 2.6.
1 Event
If the risk level is not acceptable, then mitigation measures Figure 13 Example of risk matrix with acceptable risk
should be taken to reduce the risk, see section 2.7. The length level indicated.
of pipeline to be protected should be so that the overall risk
of both the protected and the unprotected parts are
acceptable.
Risk matrices should be established for 7.2 Uncertainty assessment

A risk assessment as outlined in this recommended practice
− each identified hazardous situation (i.e. dropped objects, is normally be based on several assumptions. The main
trawling, etc.); assumptions should be clearly stated and the effect on risk
− each relevant location (i.e. mid-line zone, near platform should be discussed or evaluated with sensitivity studies.
or near subsea installations);
– each consequence (human safety, environmental impact Sensitivity studies/evaluations should include:
and economic loss).
− variations in load data,
− variations in drop point,
Note that normally only one of the hazardous situations will − variations in pipeline and umbilical capacity,
dictate the protection requirements. If several hazards give – variations in consequences.
high risk then any cumulative effects, i.e. dependency
between events should be accounted for so that the total risk
level is acceptable. Alternatively, the acceptance criterion
may be adjusted to account for such effects. Note that the
same may be observed when splitting hazardous situations
into numerous underlying specific events. In such cases the
results could indicate acceptable risk levels for all specific
events, however the correct cumulative risk could be
unacceptable.
DET NORSKE VERITAS

Page 36
8 References Jensen, J., J. (1978) “Impact Strength of Concrete Coating on

Pipelines”, SINTEF
Jensen, J.J and Høiseth, K. (1983) “Impact of dropped

Det Norske Veritas (1988) “Design Guidance for Offshore objects on lightweight concrete”, Publication unknown.
Steel Structures Exposed to Accidental Loads”, DNV Report
no. 88-3172 Katteland, L.H. and Øygarden, B. (1995), “Risk analysis of
dropped objects for deep water development”, Proc. of the
Det Norske Veritas (1996b), “Worldwide Offshore Accident 14th OMAE
Databank (WOAD)”, version 4.11, December 1996
Moan, T., Karsan, D. and Wilson, T. (1993), “Analytical
Det Norske Veritas (1997), “Interference between Trawl Risk Assessment and Risk Control of Floating Platforms
Gear and Pipelines”, Guideline no. 13 Subjected to Ship Collision and Dropped Objects”,
Proceedings to the 25th OTC in Huston, OTC no.: 7123
Det Norske Veritas (1997b), “Protection study GFSAT –
Risk assessment for pipelines and umbilicals”, DNV report Norsok (1998), “Subsea structures and Piping Systems”, U-
no.: 97-3373, revision 03 002, revision 2
Det Norske Veritas (1998), “Risikobilde Tankskip”, DNV Wiezbicki, T. and Suh, M.S. (1988), “Indentation of tubes
Report no. 98-3222 under combined loading”, Int. Journal of Mechanical
Science, 1988, vol. 30, no.3-4, p229-248.
Det Norske Veritas (2000), “Offshore Standard Submarine
Pipeline Systems”, DNV-OS-F201 SikTec A/S (1992), ”Lasthåndtering på flyterigg”, report no.:
ST-92-CR-001-02
Det Norske Veritas (2000b), “Offshore Standard Metallic
Risers”, DNV-OS-F201 Solem, Richard M. (1980), “Probability Models and
Grounding and Collisions”, Proceedings form Automation
Fujii, Y. and Yamanouchi, H. (1974), “The probability of for Safety in Shipping and Offshore Petroleum Operations,
Stranding”, Inst. Of Navigation Journal 27, 2 A.B. Aune and J. Viletstra (edt.), North-Holland Publication
Company
Fujii, Y., Yamanouchi, H. and Matui T. (1984), “Survey of
Traffic Management Systems and brief Introduction to Technica (1987), “Ship-Modu Collision Frequency”, Report
Marine Traffic Studies”, Electronic Navigation Research no.3, RABL Project, London, July 1987
Institute Paper no. 45, Japan
Aanesland, V. (1987), ”Numerical and experimental
J.P. Kenny (1998), “Protection of offshore Installations investigation of accidental falling drilling pipes”, Proc. of the
Against Boat Impact. Background Report”. OTI 88 535, 19th OTC, no. 5497
(HMSO)
DET NORSKE VERITAS

Appendix A. Example of risk assessment procedure for dropped objects

A.1 Introduction
This appendix gives an example of a detailed risk assessment of dropped objects on a 20-inch pipeline coming into a small
platform. References to the recommended practice are stated where applicable.
The field layout with the pipeline approach and crane location is given in Figure A1. Note that the crane can only work on the
platform west side and the vessel approach is from the north. The pipeline exit on the west side and after 40 metres the heading
is north.
Cr
Crane an
ee
x
zo clu
ne sio
Su esse
n
pp l
v
ly
Drop point
10 m
N Loading
Pipeline
area
40 m
Platform Leg
20 m
60 m
Figure A1 Field layout.
A.2 Design Basis

The following main data are chosen to demonstrate the use of this Recommended Practice:
Pipeline data:
Outer diameter (D) ......................................................................................... : 508 mm
Wall thickness (t)............................................................................................ : 18 mm
Yield stress (σy).............................................................................................. : 450 N/mm2
Concrete thickness.......................................................................................... : 60 mm
Environmental data:
Water depth : ............................................................................................. : 100 m
Acceptance criteria:
The acceptance criteria as given in the DNV-OS-F101 applies, i.e. the annual failure frequency shall be less than 10-5, i.e.
safety class high.
DET NORSKE VERITAS

Page 38
A.3 Classification of objects (Section 3.1.1)

The platform has only one crane with a limited operational radius. The items lifted on an annual basis are given in Table A1.
For the simplicity of this example internal lifts are assumed to result in hit onto the platform and not into the sea.
Table A1 Object classification of annual crane load data lifted to and from supply vessels.
Weight in air Number lifted
no Description Typical objects
(tonnes) per year
1 <2 Drill collar/casing, scaffolding 700
Flat/long
2 shaped 2–8 Drill collar/casing 50
3 >8 Drill riser, crane boom 5
4 <2 Container (food, spare parts), basket, crane block 500
Box/round
5 shaped 2–8 Container (spare parts), basket, crane test block 2500
6 >8 Container (equipment), basket 250
7 Box/round >> 8 Massive objects as BOP, Pipe reel, etc. 0

shaped
Total 4005
A.4 Drop frequency (Section 5.2.1)

The generic drop frequency for crane activities can be determined according to Table 9. For this example all lifts are below 20
tonnes and the frequency of dropped load into the sea is then 1.2·10-5 per lift.
A.5 Excursion of objects (Section 5.2.2)

Based on the crane location, the vessel approach area and the land area on the platform a most likely drop point is chosen. The
drop point is found 10 metres off the platform north edge and 20 metres from the platform west side, as indicated on Figure
A1. Some shielding effect from the platform legs are anticipated.
The excursion of different objects is a stochastic event. A normal distribution as given in equation (9) is used to describe the
fall pattern for each of the object categories. Due to the limited water depth, any currents will have limited effect on the
excursion of the objects and is therefor not accounted for.
From the drop point concentric rings of increasing 10 metres radius are drawn up, see figure A2. The conditional probabilities
for objects from each of the object categories to fall within these rings are given in Table A2. As an example, the probability
of an object in category one hitting within the first 10-metre ring is calculated in the following.
The lateral deviation, δ, in 100 metre water depth (d) is for objects in category 1 with an angular deviation of α equal to 15
deg, found by
δ = d ⋅ tan α = 100 ⋅ tan 15 = 26.8 m
The probability of one object in category 1 falling within the first 10 metres then becomes
2
10 10 1 x 
1 −  
Phit ,10 ( x ≤ 10 m) = ∫ p( x)dx = ∫
−10 −10 2π δ
e 2  δ  dx = 0.2910
The probability of hit per seabed area (m2) is found by dividing the hit probability within the first 10-metre radius by the area
of this radius: (Note that Phit,Ar is not given explicitly in the Recommended Practice, but used to ease the calculations. The only
difference to Phit,sl is that the exposed area of the pipeline is not included. This exposed area is accounted for later.)
Phit ,10 0.2910

Phit , Ar ,10 = = = 0.000926 m −2
Ar π ⋅ (10m) 2
This number can be found in Table A2 as the first item for the category 1 objects.
DET NORSKE VERITAS

Table A2 Conditional probability of hit for each of the objects to fall within 10-metre intervals on the seabed.
Object Probablity per m2
Deviaion
0 - 10 - 20 - 30 - 40 - 50 - 60 - 70 - 80 - 90 - 100- 110- 120-
no Desc. Angu- Late- 10 20 30 40 50 60 70 80 90 100 110 120 130
lar ral1
(deg) (m)
0.00 0.00 0.00 5.79 2.6 1.07 3.95 1.31 3.83 9.93 2.27 4.55 8.02
1 15 26.8 0926 0269 0123 E-05 E-05 E-05 E-06 E-06 E-07 E-08 E-08 E-09 E-10
Flat/
2 long 9 15.8 0.00 0.00 9.45 2.12 3.52 4.18 3.47 2.01 7.99 2.19 4.09 5.22 4.52
shape 1503 0341 E-05 E-05 E-06 E-07 E-08 E-09 E-11 E-12 E-14 E-16 E-18
3 5 8.8 0.00 0.00 1.38 2.73 1.71 3.18 1.72 2.83 0 0 0 0 0

2378 0245 E-05 E-07 E-09 E-12 E-15 E-19
4 10 17.6 0.00 0.00 0.00 2.98 6.62 1.13 1.46 1.41 1.01 5.34 2.09 6.0 1.27
1367 0333 0107 E-05 E-06 E-06 E-07 E-08 E-09 E-11 E-12 E-14 E-15
Box/
5 round 5 8.8 0.00 0.00 1.38 2.73 1.71 3.18 1.72 2.83 0 0 0 0 0
shape 2378 0245 E-05 E-07 E-09 E-12 E-15 E-19
6 3 5.2 0.00 5.97 8.63 4.74 8.25 0 0 0 0 0 0 0 0

3004 E-05 E-08 E-12 E-18
No items of category 7 are to be lifted, thus this category is excluded.
1 The lateral deviation is for 100 metre waterdepth.
120 m
110 m 14 m
100 m
15 m
90 m
80 m
17 m
70 m
60 m
21 m
50 m
40 m
30 m
29 m
20 m
10 m
40 m
Drop point
10 m
10 m
12 m
Pipeline
12 m 11 m 11 m
70 m 80 m 90 m 100 m 110 m 120 m
Platform Leg
20 m
60 m
Figure A2 Field Layout with indication of 10-metre interval rings for calculating
the object excursion and hit probability.
DET NORSKE VERITAS

Page 40
A.6 Hit probability (Section 5.2.2)

The hit probability depends on the excursion of the objects as calculated in Table A2 and the length of pipeline within each
ring and the pipeline diameter and object size.
The length of pipeline within each section is given in Figure A2 and Table A3. The pipeline diameter is 0.63 metres including
coating and the object size is assumed to be 12 metres long for the slender objects and 5 metres long for the box shaped.
Table A3 Length of pipeline within each of 10-metre interval rings on the seabed.
Pipeline length within each ring
0 - 10 – 20 - 30 - 40 - 50 - 60 – 70 - 80 - 90 - 100- 110- 120-

10 20 30 40 50 60 70 80 90 100 110 120 130
Length(m) 0 0 0 0 0* 0* 11 51 41 21 17 15 14
* Assumed shielded by the platform legs and bracing.
The resulting conditional probability of hitting the pipeline is given in Table A4. As an example the conditional probability for
the 60-70 metre radius ring for object category 1 is calculated. The conditional probability of hitting the seabed within this ring
is found in Table A2, being (Phit, 70 / Ar ) = 3.95E-06 per m2. The length of the exposed pipeline is 11 metres as given in Table
A3 and the breadth of the object is conservatively taken as the whole length of a pipe string, i.e. 12 metres. The conditional
probability of hitting the pipeline then becomes
Phit , 70
Phit ,sl , 70 = ⋅ Lsl ⋅ (D + B ) = 3.95 ⋅ 10 −6 m −2 ⋅ 11m ⋅ (0.63m + 12m ) = 0.00055
Ar
Table A4 Conditional probability for each of the objects to hit the pipeline within 10-metre intervals on the seabed.
Object Probablity
Breadth 0 - 10 - 20 - 30 - 40 - 50 - 60 - 70 - 80 - 90 - 100- 110- 120-

no Descrip. Sum
(m) 10 20 30 40 50 60 70 80 90 100 110 120 130
0 0 0 0 0 0 0.00 0.00 0.00 2.63 4.87 8.62 1.42 0.00
1 12 055 0842 020 E-05 E-06 E-07 E-07 162
Flat/
2 long 12 0 0 0 0 0 0 4.83 1.29 4.14 5.8E 8.78 9.88 8E- 6.2
shaped E-06 E-06 E-08 -10 E-12 E-14 16 E-06
3 12 0 0 0 0 0 0 2.39 1.82 0 0 0 0 0 2.4

E-13 E-16 E-13
4 5 0 0 0 0 0 0 9.02 4.03 2.32 6.31 2E- 5.07 9.99 1.3

Box/ E-06 E-06 E-07 E-09 10 E-12 E-14 E-05
round 0 0 0 0 0 0 1.06 8.11 0 0 0 0 0 1.1
5 shaped 5
E-13 E-17 E-13
6 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0
The final hit frequency is found by multiplying the number of lifts given in Table A1 with the drop frequency of 1.2·10-5 per
lift and the conditional hit probabilities given in Table A4. The results are given in Table A5.
DET NORSKE VERITAS

Table A5 Resulting hit frequency.

Objects Drop
Number lifted Conditional hit
frequency per Hit frequency
Weight in per year probability
no Description lift
air (tonnes)
1 <2 700 1.2E-5 0.00162 1.36E-5
Flat/long
2 shaped 2–8 50 1.2E-5 6.2 E-06 3.72E-9
3 >8 5 1.2E-5 2.4 E-13 ~0
4 <2 500 1.2E-5 1.3 E-05 7.80E-8
Box/round
5 shaped 2–8 2500 1.2E-5 1.1 E-13 3.3E-15
6 >8 250 1.2E-5 0 ~0
Sum 1.368E-5
The annual hit frequency is found to be 1.37·10-5. In order to find the failure frequency the energy of the objects and the
capacity of the pipeline need to be considered.
A.7 Hit frequency versus energy (section 5.3)

The impact energy of each object can be determined as described in section 5.2.3. For the example the conditional impact
energy distribution as given in Table 12 is used.
Combining Table 12 and the results of hit frequency given in Table A5 above, the hit frequency can be established for different
energy levels. Table A6 gives the resulting frequency for each object category and Table A7 gives the resulting accumulated
hit frequency, see also figure A3.
Table A6 Hit frequency for different impact energy levels.

Objects Energy level (kJ)
Weight in
no Description <50 50-100 100-200 200-400 400-800 >800
air (tonnes)
1 <2 4.09E-06 2.45E-06 1.91E-06 1.63E-06 1.50E-06 2.04E-06
Flat/long
2 shaped 2–8 1.85E-10 2.96E-10 5.54E-10 7.02E-10 9.24E-10 1.03E-09
3 >8 0 0 1.53E-18 2.29E-18 4.59E-18 6.88E-18
4 <2 3.99E-08 2.39E-08 1.60E-08 0 0 0
Box/round
5 shaped 2–8 0 6.39E-16 9.59E-16 1.28E-15 3.20E-16 0
6 >8 0 0 0 0 0 0
Table A7 Accumulated hit frequency for different impact energy levels

Energy level (kJ)
>0 >50 >100 >200 >400 >800
Annual hit frequency 1.37E-05 9.58E-06 7.10E-06 5.18E-06 3.54E-06 2.04E-06
DET NORSKE VERITAS

Page 42
1.0E-04
Frequency of Hit
1.0E-05
1.0E-06
1.0E-07
>0 kJ >50 kJ >100 kJ >200 kJ >400 kJ >800 kJ
Energy
Figure A3 Accumulated annual hit frequency for different impact energy levels.
A.8 Damage capacity versus energy (section 4)

For each of the damage classes defined in section 4.2 (D1, D2, D3, R0, R1and R2), conditional probabilities for damage to the
pipeline can be determined as proposed in Table 4. The impact energy required to create a dent of 5% is found by:
1 1 3 1
 2π  2  D 2  δ 2 6 N 2  0.508m  2
3
E = 16 ⋅   ⋅ m p ⋅   ⋅ D ⋅   = 13.37 ⋅ 0.25 ⋅ 450 ⋅ 10 2
⋅ (0.018m )   0.508m ⋅ (0.05) 2 = 14.7 kJ
 9   t   D m  0.018m 
The results for larger dents are given in Table A8. In addition the 60 mm concrete coating has impact resistance. According to
section 4.6.1 the impact capacity of the coating is taken as (both expressions calculated):
 4   N N 4 
E k =  Y ⋅ b ⋅ h ⋅ x0 ; Y ⋅ b D ⋅ x03  =  3 ⋅ 35 ⋅ 10 6 2 ⋅ 0.03m ⋅ 0.3m ⋅ 0.06m;3 ⋅ 35 ⋅ 106 2 ⋅ 0.03m 0.63m ⋅ 0.06m 3 
 3   m m 3 
= (56.7 kJ;48.9 kJ ) ≈ 50 kJ
Here the breadth, b, and height, h, of the impacting object is assumed to be 30 mm and 300 mm respectively. The concrete
coating thus has an impact capacity of approximately 50 kJ. The total capacity of the pipeline and coating is given in Table A8.
Table A8 Conditional impact capacity of pipeline and coating.
Dent/ Impact energy Damage description Conditional probability

Diameter
Steel pipe Total
(%) only (Coating D1 D2 D3 R0 R1 R2
included)
<5 < 15 kJ < 65 kJ Minor damage. 1.0 0 0 1.0 0 0
Major damage.
5 – 10 15 – 40 kJ 65 – 90 kJ 0.1 0.8 0.1 0.9 0.1 0
Leakage anticipated
Major damage.
10 – 15 40 – 75 kJ 90 – 125 kJ Leakage and rupture 0 0.75 0.25 0.75 0.2 0.05
anticipated.
Major damage.
15 – 20 75 – 115 kJ 125 – 165 kJ Leakage and rupture 0 0.25 0.75 0.25 0.5 0.25
anticipated.
> 20 > 115 kJ > 165 kJ Rupture. 0 0.1 0.9 0.1 0.2 0.7
DET NORSKE VERITAS

A.9 Damage versus frequency

Damage versus frequency can be determined by combining the “hit frequency versus energy” and “damage capacity versus
energy” as found in section A7 and A8 respectively.
Table A8 Failure frequency versus damage category.
Dent/ Impact energy Damage description Frequency

Diameter
Steel pipe Total
(%) only (Coating D1 D2 D3
included)
<5 < 15 kJ < 65 kJ Minor damage. 4.87E-06 0 0
Major damage.
5 – 10 15 – 40 kJ 65 – 90 kJ 1.24E-07 9.91E-07 1.24E-07
Leakage anticipated
Major damage.
10 – 15 40 – 75 kJ 90 – 125 kJ Leakage and rupture 0 7.32E-07 2.44E-07
anticipated.
Major damage.
15 – 20 75 – 115 kJ 125 – 165 kJ Leakage and rupture 0 1.92E-07 5.77E-07
anticipated.
> 20 > 115 kJ > 165 kJ Rupture. 0 5.85E-07 5.27E-06
Totals 4.99E-06 2.50E-06 6.21E-06
Damage class D1 is not considered to give damage leading to failure. The failure frequency is obtained by adding the results
for damage class D2 and D3. From Table A8, it can be seen that the annual frequency of failure is 8.7·10-6 which is within the
acceptance criteria of 1·10-5.
As the failure frequency is within the allowable and for this example it is assumed that any other hazards do not represent risks
for the pipeline of the same order of magnitude as dropped objects, it is concluded that the protection proposed for this pipeline
is adequate.
DET NORSKE VERITAS

Page 44
Appendix B. Impact capacity testing procedure

B.1 Introduction
For some components, the stated capacity formulations may not be applicable, or may result in estimates with large
uncertainty, etc. If it is necessary to establish the exact capacity, impact testing may be performed. A procedure for destructive
testing of components to establish impact capacity to be used in risk assessments is presented below. This procedure is focused
on determination of the impact capacity of steel pipes with diameter up to 10”-12”, flexibles and umbilicals.
The testing should reflect the accidental situations under consideration, and should aim to determine the capacity limits for the
different damage categories given in the methodology, e.g. D1 to D3.
B.2 Test energy

The test energy shall be based on the kinetic energy that is representative for the objects that are most likely to hit the
component, as calculated according to section 5.2, or if possible, the energy should be increased until a damage equal to
category D3 is obtained.
B.3 Test Equipment

B.3.1 General
The test rig should simulate a realistic situation. Such tests are not normally instrumented to record the material behaviour
during impact, only the final damage are measured. As the impact calculations for the risk assessment are not detailed, no
instrumentation is necessary.
In the simplest form, the test rig could be a crane with a remotely controlled release hook. It shall be ensured that the test
hammer will not rotate during the testing.
B.3.2 Hammer
The test hammer should normally have a mass of 1 tonnes, see Table B1. The front of the hammer should be made up with a
rectangular plate of 300 mm height/length and 50 mm width with a conical shape and an edge radius of 7 mm.
If the shape of the falling objects is known, e.g. an anchor chain, the actual shape can be used as the hammer front.
B.3.3 Support conditions

The support conditions should represent the most onerous case for the actual configuration, e.g. soil conditions similar to the
actual location, swan neck configuration, etc.
However, if the test is performed on stiff supports, then the test will reflect the true capacity of the component, i.e. all energy
will be absorbed by the component and none transferred to supports. In this way, the results will not be project specific and
may then be used for other projects.
B.4 Procedure
The testing should be repeated to ensure that the results are consistent. For design applications, the lowest reported value
should be used.
For risk assessment, the capacity will normally be the (mean) value found. However, for components where capacity is
sensitive to the shape of the hammer front, the capacity should be taken as 0.9 of the reported (mean) value. Examples of the
latter are multi-layer coatings for pipes, flexible pipes and umbilicals. In Table B1, the profile of the impacting object is given
along with directions to deciding the impact capacity.
DET NORSKE VERITAS

Table B1 Impact testing – applicable profile, mass and capacity

Description Test profile Test mass Applicable capacity
Simulating impact of any object
Steel pipes, protected or not R = 7mm 1 tonnes x
Steel pipes with coating (total capacity) R = 7mm 1 tonnes x or x = 0.9xR=7mm1
Flexibles and/or umbilicals protected R = 7mm 1 tonnes x = 0.9xR=7mm
Any additional protection (not coating) R = 7mm 1 tonnes x or x = 0.9xR=7mm1
Simulating impact of a 7” pipe (equal to tubing/liner) falling horizontally
Coating for steel pipes Simulate 7” pipe falling horizontally 0.6 tonnes x = 0.9x7” pipe
Flexibles and/or umbilicals Simulate 7” pipe falling horizontally 0.6 tonnes x = 0.9x7” pipe
1 If protection is sensitive to the test profile, R, the capacity should be reduced to 0.9 the observed capacity
Definitions:
x : observed impact capacity
xR = 7mm : observed impact capacity for test profile with R=7mm
x7” pipe : observed impact capacity for test profile that simulates a 7” pipe falling horizontally
R : profile as shown in Figure B1
Where nothing else is indicated, pipelines/umbilicals are considered not protected.
Use of Table B1 v
This table applies for activities in the vicinity of subsea
templates. The table is to be used as follows:
For the pipeline/umbilical/protection in question, the testing 90o

requirements and applicable capacity can be read in the R
relevant row. For example, for a flexible pipe to be tested for
any object hitting the pipe, the following data apply:
– Test profile: R = 7 mm
– Test mass: 1 tonne
– Applicable capacity: x = 0.9·xR=7mm (i.e. the applicable
capacity is 0.9 of the tested value) Figure B1 Profile for deciding impact capacity.
- o0o -
DET NORSKE VERITAS

RP F107 PDF

Uploaded by

Copyright:

Available Formats

RP F107 PDF

Uploaded by

Document Information

Original Title

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

RP F107 PDF

Uploaded by

Copyright:

Available Formats

RECOMMENDED PRACTICE

RISK ASSESSMENT OF PIPELINE PROTECTION

DET NORSKE VERITAS

Comments may be sent by e-mail to rules@dnv.com

Printed in Norway by GCS AS.

DET NORSKE VERITAS

DET NORSKE VERITAS

1.4 General considerations

DET NORSKE VERITAS

Below cellar deck

Tie-in area up to outboard hub

Topside Riser Pipeline Subsea installation

Not covered Not covered

Figure 1 Application of the recommended practice

Damage: Damage to pipelines is divided into three

Risk: expression of the product of the frequency

DET NORSKE VERITAS

failure probability requirements given in DNV-OS-F101

- end - of - Guidance - note - Acceptable

DET NORSKE VERITAS

2.5 Hazard identification It is not normally practicable to protect against accidental

Table 1 Possible external hazards.

Lifting activities Drop of objects into the sea Impact damage

ROV impact Impact damage

2.6 Risk Assessment − estimated based on engineering judgement, operator

– calculated when detailed information exists (e.g.

DET NORSKE VERITAS

Figure 3 Process description of a risk assessment

DET NORSKE VERITAS

Table 2 Risk reducing measures

DET NORSKE VERITAS

– between supply vessel and platform/rig, 3.1.1.3 Crane information

DET NORSKE VERITAS

3.4 Ship − fishing vessel density (per km2),

DET NORSKE VERITAS

DET NORSKE VERITAS

= plastic moment capacity of the wall (= ¼ σy t )

D = steel outer diameter

DET NORSKE VERITAS

Dent/ Impact Damage description Conditional probability2

<5 Eq. (3) Minor damage. 1.0 0 0 1.0 0 0

DET NORSKE VERITAS

For pull-over/hooking loads acting on umbilicals, capacities

Table 6 Impact capacity and damage classification of umbilicals

The kinetic energy absorbed for two different cases may be

Figure 5 Impact in concrete coating.

DET NORSKE VERITAS

4.6.2 Polymer coating 2

Thicker multi-layer coating 6-15 mm ~5 kJ 0.50

If polymer coating is to be used as protection against specific 1.50

The bearing capacity coefficients can be chosen as Nq = 99

DET NORSKE VERITAS

Table 8 Other protection methods

DET NORSKE VERITAS

5.2 Crane activity

The proposed dropped object frequency is given Table 9. It is

DET NORSKE VERITAS

Table 10 Angular deviation of object category.

Figure 9 Probability of hit within a ring, defined by