JPE Dec 2012 - Sample Issue

December, 2012 Vol.12, No.
4
Great Southern Press Clarion Technical Publishers
Journal of
Pipeline Engineering
incorporating
The Journal of Pipeline Integrity
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Journal of Pipeline Engineering
Editorial Board - 2012
Dr Husain Al-Muslim, Pipeline Engineer, Consulting Services Department, Saudi Aramco, Dhahran,
Saudi Arabia
Mohd Nazmi Ali Napiah, Pipeline Engineer, Petronas Gas, Segamat, Malaysia
Dr-Ing Michael Beller, Landolt Steuer & Unternehmensberatung AG, Luzern, Switzerland
Jorge Bonnetto, Operations Director TGS (retired), TGS, Buenos Aires, Argentina
Dr Andrew Cosham, Atkins Boreas, Newcastle upon Tyne, UK
Dr Sreekanta Das, Associate Professor, Department of Civil and Environmental Engineering, University
of Windsor, ON, Canada
Leigh Fletcher, Welding and Pipeline Integrity, Bright, Australia
Daniel Hamburger, Pipeline Maintenance Manager, Kinder Morgan, Birmingham, AL, USA
Dr Stijn Hertele, Universiteit Gent Laboratory Soete, Gent, Belgium
Prof. Phil Hopkins, Executive Director, Penspen Ltd, Newcastle upon Tyne, UK
Michael Istre, Chief Engineer, Project Consulting Services,
Houston, TX, USA
Dr Shawn Kenny, Memorial University of Newfoundland Faculty of Engineering and Applied
Science, St Johns, Canada
Dr Gerhard Knauf, Salzgitter Mannesmann Forschung GmbH, Duisburg, Germany
Prof. Andrew Palmer, Dept of Civil Engineering National University of Singapore, Singapore
Prof. Dimitri Pavlou, Professor of Mechanical Engineering,
Technological Institute of Halkida , Halkida, Greece
Dr Julia Race, School of Marine Sciences University of Newcastle,
Newcastle upon Tyne, UK
Dr John Smart, John Smart & Associates, Houston, TX, USA
Jan Spiekhout, Kema Gas Consulting & Services, Groningen, Netherlands
Prof. Sviatoslav Timashev, Russian Academy of Sciences Science
& Engineering Centre, Ekaterinburg, Russia
Patrick Vieth, President, Dynamic Risk, The Woodlands, TX, USA
Dr Joe Zhou, Technology Leader, TransCanada PipeLines Ltd, Calgary, Canada
Dr Xian-Kui Zhu, Senior Research Scientist, Battelle Pipeline Technology Center, Columbus, OH,
USA

S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
4th Quarter, 2012 249
The Journal of
Pipeline Engineering
incorporating
The Journal of Pipeline Integrity
Volume 12, No 4 Fourth Quarter, 2012
Contents
Stephen J Wuori ...........................................................................................................................................251
Pipelines for the 21st Century: safety, innovation, and technology
Dr Mo Mohitpour ........................................................................................................................................255
Obituary
Willard A Maxey ..........................................................................................................................................257
Obituary
Eric Jas, Dermot OBrien, Roland Fricke, Alan Gillen, Prof. Liang Cheng, Prof. David White,
and Prof. Andrew Palmer .............................................................................................................................259
Pipeline stability revisited
Prof. Andrew Palmer ....................................................................................................................................269
10
-6
and all that: what do failure probabilities mean?
Dr Filip Van den Abeele and Raphael Denis .................................................................................................273
Numerical modelling and analysis for offshore pipeline design, installation, and operation
Rob Bos, Suzanne Mooij, Leen Pronk, and Wessel Bergsma ..........................................................................287
Risk control at lower cost
Pipeline Pigging Conference in Houston: 25 years ........................................................................................305
OUR COVER PICTURE shows a graphic of typical loading pattern for a subsea
pipeline. The figure is taken from the paper on pages 273-286 which examines the
issue of numerical modelling and analysis for the design, installation, and operation of
subsea pipelines.
The Journal of Pipeline Engineering
has been accepted by the Scopus
Content Selection & Advisory Board
(CSAB) to be part of the SciVerse
Scopus database and index.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
The Journal of Pipeline Engineering 250
1. Disclaimer: While every effort is made to check the
accuracy of the contributions published in The Journal of
Pipeline Engineering, Great Southern Press Ltd and Clarion
Technical Publishers do not accept responsibility for the
views expressed which, although made in good faith, are
those of the authors alone.
2. Copyright and photocopying: 2012 Great Southern
Press Ltd and Clarion Technical Publishers. All rights
reserved. No part of this publication may be reproduced,
stored or transmitted in any form or by any means without
the prior permission in writing from the copyright holder.
Authorization to photocopy items for internal and personal
use is granted by the copyright holder for libraries and
other users registered with their local reproduction rights
organization. This consent does not extend to other kinds
of copying such as copying for general distribution, for
advertising and promotional purposes, for creating new
collective works, or for resale. Special requests should
be addressed to Great Southern Press Ltd, PO Box 21,
Beaconsfield HP9 1NS, UK, or to the editor.
3. Information for subscribers: The Journal of Pipeline
Engineering (incorporating the Journal of Pipeline Integrity)
is published four times each year. The subscription price for
2012 is US$350 per year (inc. airmail postage). Members of
the Professional Institute of Pipeline Engineers can subscribe
for the special rate of US$175/year (inc. airmail postage).
Subscribers receive free on-line access to all issues of the
Journal during the period of their subscription.
4. Back issues: Single issues from current and past volumes
are available for US$87.50 per copy.
5. Publisher: The Journal of Pipeline Engineering is
published by Great Southern Press Ltd (UK and Australia)
and Clarion Technical Publishers (USA):
Great Southern Press, PO Box 21, Beaconsfield
HP9 1NS, UK
tel: +44 (0)1494 675139
fax: +44 (0)1494 670155
email: jtiratsoo@gs-press.com
web: www.j-pipe-eng.com
www.pipelinesinternational.com
Editor: John Tiratsoo
email: jtiratsoo@gs-press.com
Clarion Technical Publishers, 3401 Louisiana,
Suite 255, Houston TX 77002, USA
tel: +1 713 521 5929
fax: +1 713 521 9255
web: www.clarion.org
Associate publisher: BJ Lowe
email: bjlowe@clarion.org
6. ISSN 1753 2116
T
HE Journal of Pipeline Engineering (incorporating the Journal of Pipeline Integrity) is an independent, international,
quarterly journal, devoted to the subject of promoting the science of pipeline engineering and maintaining and
improving pipeline integrity for oil, gas, and products pipelines. The editorial content is original papers on all aspects
of the subject. Papers sent to the Journal should not be submitted elsewhere while under editorial consideration.
Authors wishing to submit papers should do so online at www.j-pipeng.com. The Journal of Pipeline Engineering now
uses the ScholarOne manuscript management system for accepting and processing manuscripts, peer-reviewing, and
informing authors of comments and manuscript acceptance. Please follow the link shown on the Journals site to submit
your paper into this system: the necessary instructions can be found on the User Tutorials page where there is an Author's
Quick Start Guide. Manuscript files can be uploaded in text or PDF format, with graphics either embedded or separate.

Please contact the editor (see below) if you require any assistance.
The Journal of Pipeline Engineering aims to publish papers of quality within six months of manuscript acceptance.
Notes
v v v
www.j-pipe-eng.com
is available for subscribers
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
T
his is a transcript of the Petrobras Luncheon keynote address given at the International Pipeline
Conference held in Calgary on 24-28 September, and organized by the Pipeline Systems Division of
the ASME and the Canadian Energy Pipelines Association.
by Stephen J Wuori
President, Liquids Pipelines and Major Projects, Enbridge Inc, Edmonton, AB, Canada
Pipelines for the 21st Century:
safety, innovation, and technology
I
WOULD LIKE to address four topics today: the
role of pipelines, opposition to pipelines, technology
developments, and regulatory environment. But first I want
to lay out a challenge to you one that I will repeat at the
end of my speech: we must have unwavering dedication to
zero leaks and zero incidents. It must be our goal. Pipeline
companies in North America are being held to a standard
of perfection, and it is technology that will help us reach
that goal.
Role of pipelines
I see us focusing on two key frontiers regarding the
expectations of the public:
to find ever-smaller features in the pipelines; and
to find ever-smaller leaks in the pipelines.
Our industry is not perfect, but we must continue to progress
and be proud of the vital function we provide to society.
Let me congratulate all of you here in the audience for your
relentless devotion to research and the development of newer
and better technologies to keep moving this industry forward
in continuing to deliver oil and natural gas in the safest, most
efficient and most economical way possible and to do it
with this one goal in mind: zero leaks. Without us providing
our services to tens of millions of people every day, the very
nature of peoples lives would be fundamentally altered.
People may read negative things about us these days in the
media and on line, and they may hear the negative stories on
the six-oclock news, but there is a more important element
of communication going on in nearly every aspect of their
regular lives that reinforces for them and for us that we
take our jobs very seriously and we do that job well.
Whenever they turn the heat on in their homes, or are
able to put gas in their cars or enjoy the wide variety of
fresh goods at their local grocery store, that is us doing
our job well.
Some things we need to keep in mind:
90% of all crude shipped in North America is
shipped by pipeline.
Canadians obtain 70% of our energy from
hydrocarbons and these hydrocarbons are delivered
by pipelines.
The oil and natural gas industry is the backbone
of the American economy supplying more than
60% of US energy.
Every year since 2007, Canadian pipelines have
delivered more than $100 billion worth of energy
to Canadian energy users and to export customers.
Canadas transmission pipeline companies, operate
approximately 109,000 km of pipeline in Canada
and the United States. These energy highways
moved approximately 1.2 billion barrels of liquid
petroleum products and 5.3 trillion cubic feet of
natural gas last year.
AOPL members carry nearly 85% of the crude oil
and refined petroleum products moved by pipelines
in the United States.
Although we are in the heart of the upstream producing
region for Canada, this is truly an international conference
with dozens of energy producing nations represented, so
it is important to note that these statistics represent only
Canada and the US. However, I think we can all agree
they demonstrate the crucial role our industry plays in
the global economy.
Guest Editorial
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
represented, even in this very room, on the advancement
of technology in support of pipeline inspection and leak
detection. Without your research and the application of
new technologies, we would not be able to continue to work
towards our goal of zero leaks.
The safety of the communities and environment along our
rights-of-way and the integrity of our pipelines and facilities
are of the utmost importance for this industry and also
for the citizens of the countries in which we operate. As
an example, Enbridge itself, during the last decade alone,
transported nearly 12 billion barrels of crude oil with a safe
delivery record better than 99.999%. Our safety record as
a company is strong, but our goal is zero incidents, a goal
many other companies in North America and around the
world share, and we are relentlessly committed to continuing
to improve our processes, technology and vigilance so that
we can deliver on this goal.
To reflect on improvements made in recent years:
Spills along rights-of-way in the US have fallen from
2.0 incidents per thousand miles in the earliest
three-year period (1999-2001) to 0.8 incidents
per thousand miles in the latest three-year period
(2007-2009), a decline of 60%.

The volume released along rights-of-way has fallen
from just over 600 barrels per thousand miles in
the earliest period to less than 400 barrels in the
most recent period, a decline of 35%.

The liquid spilled from pipelines in Canada over
the past ten years is equivalent to three teaspoons
dripped out of a gasoline nozzle over the course of
50 fill-ups of 50 li each.

5.5 litres is the amount of liquid spilled per million
litres transported by pipeline in Canada between
2002 and 2011.
We have also found that there has been a significant reduction
in types of spill (corrosion, features, equipment failures) and
this is directly attributable to the advancements in technology
as a result of your considerable and diligent work.
Akin to the improvements in medical technology that
have progressed from x-rays to CAT scans to MRI,
pipeline integrity has seen dramatic advancements. In
fact we have employed much of this medical technology
for pipeline monitoring and maintenance. Ten years ago
in our industry we were able to detect dents; throughout
the 1990s and 2000s, however, a variety of inspection
techniques enabled us to identify corrosion features both
internally and externally on a pipe; and today we have
the ability to detect smaller and smaller features. In fact,
Enbridge has used more crack-detection tools that the rest
of the world combined.
Opposition to pipelines
So whats the problem? Why all the scrutiny and opposition
and why now? To use a medical analogy I will compare
pipelines with the 100,000 or so miles of arteries and veins
in our bodies. As vitally important as they are to our health,
we typically only think about our arteries or veins if and
when something goes wrong with one of them. Similarly,
the public does not think about the thousands of miles
of pipelines as a vital function of society, but rather only
notices them when something goes wrong.
Environmental groups have made it clear that we are a
deliberate and strategic target for opposition to the oilsands
and, quite frankly, it makes sense. If you are opposed to the
oilsands, it doesnt make much sense to go after producers
who work in remote areas surrounded by industry workers
for a few YouTube hits.
They also know that they will only undermine their popular
support by attacking the end consumer. The mom and dad
with three kids putting gas in their SUV to drive around
the neighbourhood for groceries and to attend sporting
events are people you want as a grassroots advocates: they
are not a good target.
We, on the other hand, are the efficient target. We are
everywhere, in peoples back yards and communities. We
run across cultivated fields and along public and private
roads. Our terminals and pump stations can be in close
proximity to large populations and are clearly branded.
Get people upset about pipelines and you hit the oilsands
industry where it hurts. If we cant move the product to
market, we cant utilize or capitalize on it to its full potential
and that product cannot support growth, the economy or be
available for energy consumers growing needs. This frustrates
the market, our clients and the consumer.
Another exposure we manage daily in our industry is
the complexity of the terrain, geography, and diversity
of demographics along of a pipeline right of way. There
are endless opportunities to sow and cultivate opposition
on all kinds of issues and for all kinds of reasons, even
if many of those reasons are imagined or exaggerated.
Our focus has to be on what opportunities this scrutiny
provides for us, how it can make us better operators and
service providers. We all know that these opposition
strategies affect all of us who work in every part of this
industry. And one of the most important ways we can
address this is through our science, and our continually
advancing technology.
Technology development
As integral as our industry is to all developed and developing
economies worldwide, what is intriguing is the work
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Growth of construction features
Compressor and pump station
Measurement
Integrity management systems are being developed in
most companies throughout our industry that encompass
comprehensive analysis of pipelines and set prescriptive
performance-based regulations and standards intended to
meet the dynamic nature of pipeline operations.
Regulatory environment
We need to have and do have very strong regulators
in Canada and the United States. The public must have
confidence in them as it is the role of the regulators to be
the proxy of the people.
Conclusion: a call to action
We all agree that incidents, especially significant ones, are
unacceptable, but we are a good industry, we do an important
job well, and we need to continue to move forward. We have
too many important customers, both on the refining and
producer side, but also at the curbside, who are all relying
on us and help give us that motivation and momentum to
go forward every-day.
We still have the confidence and imperative to grow our
business and our infrastructure to meet our ever-growing
need for energy.
With your on-going efforts and research we will reach this
goal. Your work is being noticed and appreciated. And
it is having an impact you heard the statistics on the
improvements in spill records since 1999.
Again, I think it is worth repeating, especially in these days
of increased media coverage and public scrutiny focused
largely on the oilsands and our pipeline industry, that we
Let me cite a few examples of what is going on in the industry:
As many of you know, CEPA recently announced
its Integrity First Programme an industry-wide
initiative that will improve pipeline safety as well
as environmental and social performance. This
programme is based on sharing best practices
and applying advanced technology throughout
the industry, and will focus on inline inspection
and leak detection in four key areas: prevention,
emergency response, reclamation, and education.
AOPL and API leadership teams: with a focus
on public awareness and damage prevention,
eight leadership teams on pipeline performance
improvement have been established whereby
executive leaders of pipeline operators join
operational personnel to focus on the following
specific pipeline safety improvement areas:
Research and development/enhanced technology
Leak detection
Enhanced data integration
Sharing safety practices and lessons learned
Damage prevention
External communications
Strategic planning; and
Emergency response
An international organization with which you are no doubt
familiar, the Pipeline Research Council International
(PRCI), develops dynamic research programmes devoted to
identifying, prioritizing, and implementing the industrys
core research objectives. Examples of collaborative
technology developments providing a foundation for the
safe and reliable operation of the worldwide pipeline
infrastructure include such key areas as:
Corrosion: location and assessment
Mechanical damage: location and assessment
ROW monitoring
Steve Wuori giving his
speech at the IPC.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
So I would leave with you this thought: as you read the
paper, or watch the TV, or go online and see the stories
about pipelines, about our industry, I would urge you not
to be discouraged. Know that all of you here are part of
what makes these pipes flow every day and for that reason
we work in a good and vital industry and because of your
innovation and effort, it will continue to get better.
Lets take this time under the microscope to show the public
what we can do and what we know we can do better for the
future of this industry. Lets keep doing our jobs well and
continually look for ways to do our jobs better, and we will
get to that goal of zero spills.
We cant do it without the hard work and innovation of the
people in this room and for that I thank you.
do our job well. That every day we are delivering millions of
barrels of energy safely to many communities that wouldnt
be able to exist, function, or prosper without that pipeline
infrastructure and supply.
Your work is critical to the future of this industry and the safe
delivery of the products we transport. We will continually
work to advance pipeline safety and integrity with all the
players involved many of whom are with us today and
they include representatives from both corporate and
industry associations world-wide as well as regulators and
members of the community.
We are all working to make pipelines and pipeline
transportation of hydrocarbons safer today and are dedicated
to continuous future improvement.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Dr Mo Mohitpour
31 March, 1942 21 August, 2012
Obituaries
M
O WAS larger-than-life, a big man with a very kind
heart. Laurence Sterne, the Anglican cleric and author
of The Life and Opinions of Tristram Shandy, had a father
who was also larger-than-life: Sterne wanted to portray him
in his famous novel, but to do so he found that he had to
create not one but two characters, Mr Shandy and Uncle
Toby, because one would have been completely inadequate
I feel the same about Mo.
In the best possible way there were two Mo Mohitpours: the
first, the Mo whom we knew engineer, teacher, researcher,
author, the conscientious representative of his colleagues
was an ever-popular and gregarious companion. But there
was also another Mo, equally remarkable, but known only to
his family, to his confidantes, and to the many beneficiaries
of his kindnesses. This was the Mo of the warm heart and
numerous good turns; the Mo with the deep love for,
and pride in, his family; the wise counsellor and the Mo
of unswerving loyalty to people and to the engineering
institutions he so much loved, the IMechE and the ASME;
the Mo in whom you could place your absolute trust.
Mo was educated in his native Iran and came to England
to study, initially to brush up his English (he later added
Spanish and Portuguese to his repertoire), before going up
to the University of Surrey where, in 1968, he obtained
an Honours BSc in Mechanical Engineering. He went
on to complete his PhD at Imperial College, University
of London, in 1972. He spent the early years of his career
in Iran and the UK working for Bechtel and Fluor among
others, on pipeline and offshore projects, before emigrating
to Canada in 1982.
At Canuck Engineering he was responsible for supervising
the mechanical and hydraulic design of a variety of pipeline
projects ranging from the Trans-Quebec and Maritimes gas
pipeline system, through an early bitumen-transportation
line, to the use of the then novel strain-based method
for the Norman Wells pipeline which traversed areas of
discontinuous permafrost. One of Mos great passions
was working in pipeline hydraulics and he had few equals
in that realm. He got to practise his art extensively when
Canuck was bought by Nova Corporation, and he worked
initially in the Alberta Gas Trunk Line Division before
transferring to the new International Division. The latter
work took him to every continent except Antarctica though
he came close, working in southern Chile. He consulted,
gave training, did business development for NovaCorp
International throughout the world, and made friends
everywhere he went. He was responsible for setting-up the
pipeline courses initially given in the Continuing Education
Department of the University of Calgary the precursor
to the Universitys Pipeline Engineering Centre. Mo was
an energetic and enthusiastic instructor and a great mentor
to many. He was always eager to share his knowledge, and
the aforementioned training programmes became a great
resource for the six pipeline books he co-authored, the last
of which had just gone to press when he passed away.
Mo had an uncanny ability to recognise an industry need
and was the source of many initiatives benefitting the
pipeline community. He had also tremendous powers of
persuasion, convincing his peers in the Offshore Mechanics
and Arctic Engineering Division of ASME that they should
hold their 1992 Conference in land-locked Alberta. It
became, and has remained, the most successful conference
in that Divisions 32 year history! Shortly thereafter Mo
determine that Calgary should be the location of a world-
class speciality conference on energy pipelines and, in his
inimitable way, he soon had created a nucleus of helpers.
His conference timing could not have been better coming
as it did when stress-corrosion cracking was making its
presence felt in the Canadian transmission pipeline industry.
Gaining the considerable support of senior industry people
in his endeavours was second nature to Mo, so ever the
optimist, and there was never any doubt in his mind that
the inaugural International Pipeline Conference in 1996
would be successful. His fingerprints can be found all over
the seven IPC events that have followed.
He was also a driving force behind the creation of the
Pipeline Division of the ASME. Several Divisions of the
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
a smoking hot deal whether it was on Petaling Street in
KL, the flea markets of Beijing, or the stalls along
Copabacana. Quite how he intended transporting home
some of the huge objects he had purchased left him
characteristically unfazed.
Mo passed away after a short illness from congestive heart
failure leaving behind his beloved wife Carol, his son Bijan,
daughter Rachel, and the sunshine of his riper years four
grandchildren of whom he was immensely proud.
A moving and happy celebration of Mos life was held at
the International Pipeline Conference in Calgary on 27
September, attracting upwards of 200 of those attending
the conference to share stories and reminisce about this
remarkable man.
Dr Alan Murray
ASME conferred on Mo their Distinguished Service Award
and he was very proud of his election to the Fellowship grade
of the Institution of Mechanical Engineers, the ASME, and
the Engineering Institute of Canada.
When TransCanada merged with the Pipeline Division
of Nova in 1999, it shortly thereafter disengaged from
international work. Mo took this as an opportunity to branch
out on his own and he successfully set up Tempsys Pipeline
Solutions providing consultancy services and training to
many North American and international clients. In doing
so he added to an ever widening circle of friends: Mo had no
need for Facebook or Linkedin, he had long since perfected
the art of making friends and influencing people.
Travelling with Mo internationally was a never-to-be-
forgotten experience. His cultural background and natural
competitiveness meant he was always on the lookout for
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Willard A Maxey
1930 2012
A
T THE recent International Pipeline Conference,
held in Calgary in September, a special session the
Willard A (Bill) Maxey Distinguished Lecture Series was
inaugurated to celebrate the life and achievements of Bill
Maxey, who died last March.
Bill was truly a dominant figure in the technical development
of the pipeline industry. Throughout his 40-year career
serving the industry, he developed many concepts and
methods that are still at the core of linepipe specification
and integrity assessment practices. Many of the experimental
techniques that he developed for carrying out demanding
(and potentially dangerous) full-scale testing are in use today,
though digital techniques have greatly simplified the tasks
of data acquisition and reduction. Bill made continuous
technical contributions in a number of fields throughout his
more than 30 years with Battelles Columbus Laboratories
and subsequent time with Kiefner and Associates, but
is probably best remembered for his theoretical and
experimental work on fracture, which provided practical and
intuitive solutions that the industry could apply to fracture
control design and integrity management.
His massively-influential 1974 paper Fracture initiation,
propagation and arrest was re-presented, as part of the
Distinguished Lecture Series, by Dr Gery Wilkowski, a
former colleague at Battelle. Bill progressively updated the
work presented in this paper over subsequent years, to keep
pace with the developments taking place in the industry,
particularly in terms of the materials being used and the
fluids being transported. For example, he was one of the
earliest researchers to recognize, and show how to deal with,
the special problems presented by fracture control in carbon
dioxide pipelines. It is also noteworthy that Bills work on
the integrity of steel transmission pipelines was subsequently
extended to LNG piping; plastic gas distribution pipe; nuclear
pipe, pressure tubes, vessels, and steam generator tubing;
and chemical plant piping, involving a variety of materials
from aluminum to zircalloy. His work is the technical basis
of many codes and standards, not only in North America,
but around the world.
Apart from his technical and experimental skills, Bill Maxey
had an extraordinary ability to find solutions from outside
the main stream, and it was often this which allowed him
to by-pass the road-blocks encountered by more linear
thinking. Those who worked with him, whether as colleagues
or through organizations such as the Pipeline Research
Committee, benefited enormously from his guidance. He
is remembered by all who knew him as an extraordinary
researcher and a patient and effective communicator, whose
open and friendly manner invited dialogue. He will truly
be missed.
Dr Brian Rothwell
Dr Gery Wilkowski
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
T
HE STABILITY assessment of the 40-in North Rankin A trunkline, operated by Woodside Energy Ltd,
has provided better insight into fuid-soil-pipe interactions during extreme storm events. The resulting
conclusion of the work is that the trunkline, a major subsea natural gas artery in Australias Northwest Shelf
since its installation in 1982, can continue to be operated safely for the next 30 years from a hydrodynamic
stability point of view. This conclusion was reached after substantial study and physical model testing was
performed considering the tripartite interaction between fuid, seabed, and pipeline.
To provide vital information feeding into the stability analysis, a physical model testing programme was
developed, and a new world-class hydrodynamic testing facility designed, constructed, and commissioned
at the University of Western Australia. This facility allows the replication of near-seabed conditions during
tropical cyclones in controlled laboratory conditions, and observation of the interaction between ocean,
seabed, and pipeline. Tests were performed using a range of pipeline embedment profles, storm realizations,
and pipe fxity conditions simultaneously to model hydrodynamic loading onto the pipeline and seabed
scour. This data were then used in the three-dimensional numerical modelling of pipeline response using
fnite-element analyses, which included the effects of seabed instability.
*Corresponding authors contact details:
tel: +61 8 9322 7922
email: eric.jas@atteris.com.au
by Eric Jas*
1
, Dermot OBrien
1
, Roland Fricke
2
, Alan Gillen
2
, Prof. Liang
Cheng
3
, Prof. David White
3
, and Prof. Andrew Palmer
4
Pipeline stability revisited
T
HE 134-km long, 40-in diameter, and 23.8-mm wall
thickness North Rankin A (NRA) trunkline was
constructed by Woodside Energy Ltd in 1982 based on
a design life of 30 years. The gas pipeline links the NRA
Platform to the North West Shelf Venture gas plant on the
Burrup Peninsula (Fig.1) Primary stabilization is provided
in the form of a concrete weight coating. In the area of
interest, the concrete weight coating is 64 mm thick and
has a density of 3,043 kg/m
3
, and the corrosion coating
comprises a 6-mm thick layer of asphalt enamel with a
density of 1,281 kg/m
3
; the contents density for stability
design purposes is 90 kg/m
3
, and consequently the specific
gravity (SG) of the pipeline in this area is 1.23 relative to
seawater. The current practice, 30 years after the NRA
trunkline was installed, is for large-diameter pipelines to be
designed in similar water depths with a much thicker (and
sometimes much higher density) concrete weight coating,
with much higher SG values. This assists considerably in
achieving on-bottom stability without the need for applying
secondary-stabilization measures.
Along the first 22.8 km from shore, the trunkline is covered
with a minimum of 2.5 m of quarried rock to provide
protection from accidental external impacts. From KP 22.8 to
KP 123.8 the pipeline was post-trenched by ploughing in loose
and variably cemented carbonate marine sands and silts. The
plough formed an open V-shaped trench below the pipeline
with the intention that the depth of the trench would place
the top of the pipeline at or below the natural seabed level.

In April 1989, a severe Tropical Cyclone (TC) Orson
caused significant changes to the seabed bathymetry along
the trunkline, which resulted in the distinct V-shaped
ploughed trench shape disappearing. Consequently, the
required sheltering which the trench previously offered
was no longer present everywhere along the pipeline route.
Typical data from a survey undertaken after TC Orson in
1989 are shown in Fig.3, reconstructed to provide a three-
dimensional visualization of the degree of burial along a
typical length of the trunkline.
Upon discovering the changed bathymetry of the seabed in
relation to the trunkline, a remedial stabilization programme
was developed and implemented between 1990 and 1992.
This comprised rock dumping along selected sections of
the pipeline, both to improve pipeline stability and to also
stabilize the seabed either side of the pipeline.
1 Atteris Pty Ltd, Perth, WA, Australia
2 Woodside Energy Ltd, Perth, WA, Australia
3 University of Western Australia, Crawley, WA, Australia
4 National University of Singapore, Singapore
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
embedment levels, which are not compatible with the
existing codes and recommended practices.
At the time when the assessment commenced (2006) the
only available and reliable recommended practice was DNV
RP E305 [1], which:
does not provide guidance on pipe-seabed interaction
forces for pipelines on carbonate soils;
does not allow for the effect of pipeline embedment
on soil resistance and hydrodynamic loading; and
does not consider the effects of seabed instability
within the response of the pipeline during storm
loading.
The successor to DNV RP E305, published in 2007, is
DNV RP F109 [2]. This updated code does allow for
some effects of pipeline embedment; however it does not
consider asymmetrical embedment levels, and also does
not provide quantitative guidance for carbonate soils. Both
recommended practices focus on pipeline on-bottom stability
with relatively low embedment levels, and are not suited
to the assessment of highly embedded pipeline sections.
In addition, none of the existing codes and recommended
practices considers the effects of the changes in seabed
bathymetry and characteristics during a storm event
i.e. when subjected to the wave- and current-induced
hydrodynamic loads on pipeline stability. Such changes can
include sediment scour and deposition, excess pore pressure
build-up and dissipation and, sometimes, liquefaction.
The pipeline engineers who undertook the assessment
considered these effects of great importance. It has been
mentioned before that the seabed, if it comprises a fine- or
medium-sized uncemented material, will lose strength and
become mobile during the ramp-up period of a storm, long
before the pipeline becomes unstable [3], depending on
the SG of the pipeline. Whilst the recommended practices
and guidelines do not incorporate these effects within their
methodologies, many pipelines that are in operation have a
track record of losing contact with the seabed over sometimes
The desire to extend the lifespan of the trunkline beyond
2012 triggered the need to undertake a rigorous engineering
assessment of this asset. It included a study of the
hydrodynamic stability of the system for the next 30 years.
A screening process indicated that the critical area
that needed thorough review was the section along the
Continental Shelf, between the 26 m and 73 m water depth
contours, or between KP 22.8 and KP 116. The two adjacent
pipeline sections were either stabilized by quarry rock (the
shore approach) or stable under the pipelines own weight
(the platform approach).
The challenges faced by the pipeline engineers who carried
out the assessment were the following:
The trunkline comprises sections with highly variable
levels of embedment in sediments, ranging between
0% (of pipeline diameter) to 100% or more.
Along many of these areas, the level of embedment
either side of the trunkline is not the same; in some
areas there is as much as 100% embedment on one
side with little to no embedment on the other side.
At the commencement of the assessment there was
insufficient clarity as to the degree of instability of
the seabed in the immediate vicinity of the trunkline.
The SG of the trunkline (1.23 in seawater) along
the area of interest is relatively low.
A review of existing 3D pipeline-stability software
packages indicated that they would be inadequate
to undertake the assessment accurately in the given
seabed conditions, unless considerable modifications
were made to the software to account for the effects of
seabed instability onto the pipeline-response model.
The Northwest Shelf of Western Australia comprises
carbonate soils, and international pipeline codes and
recommended practices are written on the basis of
any seabed sand being siliceous.
In summary, the assessment required an unconventional
methodology in view of the nature of the sediments, the
potential effects of seabed instability, and the pipeline
Fig.1. The NRA trunkline.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
over more than two decades, including survey data from
annual and post-tropical cyclone inspections. Also, past
studies which had assessed the potential of the seabed
sediments along the pipeline route to liquefy and/or scour
were studied. Mechanical design properties of the trunkline
were also collated to create an overall picture of the asset
and its environment.
Table 1 presents the metocean data applicable to the section
of the pipeline route between KP 52 and KP 63. The seabed
along this section of the pipeline route comprises a 1 - 2
m thick layer of fine- to medium-carbonate sand overlying
a calcareous rock pavement. The sand has a D50 of 150 -
200 microns.
significant lengths in areas of loose sediments through the
forming of scour holes. The forming of scour holes along
a pipeline can sometimes be so extensive that the pipeline,
depending on its SG, experiences self-burial over time.
Methodology
As a consequence of the limitations in the design codes and
recommended practices, an unconventional methodology
was developed for this case (Fig.4). The aim of the stability
assessment was to develop an understanding of the processes
contributing to the stability (or instability) of the trunkline
and to assess whether satisfactory evidence can be gathered
to demonstrate that the risk of future trunkline instability
is sufficiently low.
The following main steps were identified when developing
the methodology of the stability assessment of the NRA
trunkline.
Overview of the assessment
Input data gathering
The NRA trunkline had previously been the subject of
much study, in particular in relation to its hydrodynamic
stability. A significant amount of data had been collected
Project Execution
Plan
Step 1
Input Data Gathering
Step 5
2D Physical Model
Step 6
3D Stability Assessment
Step 7
Remedial Stabilisation
Design
Step 8
Redefne Trunkline
Stability Criteria
Close-Out Report
Trunkline
Stable?
Step 2
Seabed Stability
Step 4
Rock Berm Stability
Assessment
Step 3
2D Pipeline On-Bottom
Stability
Fig.2. Typical as-built post-lay ploughed trench profle.
Fig.3. Various levels of pipeline embedment post-Tropical
Cyclone Orson.
Fig. 4. The assessment process.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Excess pore pressure build-up will result in soil softening and
reduced soil passive resistance, and could result in partial
pipeline flotation where significant excess pore pressures
are generated. However, it is difficult to precisely correlate
a decrease in lateral soil resistance to a value of excess pore
pressure. Although excess pore pressures generated in the
region where the pipeline stability has been analysed in detail
are not expected to have a significant effect, sensitivity load
cases have been performed in the 3D FEA analysis using
reduced soil passive-resistance values to assess the potential
effects of excess pore pressure build-up on pipeline stability.
A regional (free-field) scour analysis was also performed.
Two independent methods were used: the first used the
Soulsby method [5] to determine the volume of sediment
suspended in the water column, while the second assessed
the possibility of sheet flow, using the Flores and Sleath
method [6] to estimate the regional (free-field) scour depth.
The regional scour analysis indicated that over the long term,
the regional scour depth is likely to be limited to less than
0.1 m along the NRA trunkline route in the area of interest.
Local scour was assessed using the computational-fluid-
dynamic (CFD) package SCOUR-2D developed by the
Hydraulics Research Group, led by Professor Liang Cheng
at the University of Western Australia. It is believed that
local scour did occur during TC Orson in 1989, whereby
the spoil banks of the ploughed trench (and other
unconsolidated, cohesionless, fine-grained sediments) were
deposited on and around the pipeline inside the V-shaped
ploughed trench. Following the suspected liquefaction of
this material during the same and/or subsequent tropical
storms, and rise of the pipeline through this material,
further local scour of this seabed material is likely to have
occurred alongside the trunkline.
The results of the seabed-stability assessment indicate that
both liquefaction and scour have played a significant role
in the stability of the trunkline. Also, now that the pipeline
is exposed at the seabed, (predominantly) local scour is
likely to have a significant influence on the stability of
this pipeline.
Seabed stability assessment
The pipeline engineers who undertook the assessment
recognized that the overall stability of the pipeline depends
on the tripartite interaction between the hydrodynamic
loads induced by tropical cyclones, the seabed comprising
predominantly calcareous sediments and the trunkline.
Consequently, an in-depth study of these processes
was undertaken.
As a starting point, the interaction between the hydrodynamic
loads and the seabed was assessed. Specialists were engaged
to undertake seabed liquefaction and scour analyses.
The seabed liquefaction analysis performed for this project,
which used the methodology described by Bonjean et al.
[4] concluded that, although the large hydrostatic pressure
fluctuations caused by tropical-cyclone-induced waves do not
have the capacity to induce free-field seabed liquefaction,
it is likely that loose and fine sediments deposited within
the open trench would have liquefied during a significant
storm event (such as TC Orson in 1989). This would have
caused lift of the pipeline by several tens of centimetres. The
end result, after a significant storm, created a picture which
was perceived at the time by many as a general lowering of
the seabed (due to regional scour), while in reality it could
well have been the pipeline which had risen.
This trench backfill material liquefaction theory is considered
to be the most likely explanation for the observed change
in the burial of the NRA trunkline. It casts doubt onto the
validity of the broadly accepted regional-scour theory, or at
least the depth extent of such seabed erosion and its effect
on submarine structures such as pipelines in this region.
The seabed-liquefaction assessment also concluded that
where soils are classified as sand, excess pore pressure build-
up around the pipeline during an extreme load condition
is expected to be small (5 - 10%), which is not expected to
impact pipeline stability. However, where soils are classified
as silty sand, excess pore pressure build-up around the
pipeline can be much higher under extreme load conditions
(60 - 70%) which may be expected to cause localized partial
flotation of the pipeline.
Description Symbol Value
Significant wave height H
s
12.94 m
Peak period T
p
14.76 s
Water depth d 55.8 m
Significant wave-induced current
(perpendicular to pipeline)
U
s
1.62 m/s
Steady-state current (perpendicular to pipeline) V
R
0.52 m/s
Maximum wave-induced current
(perpendicular to pipeline)
U
max
2.46 m/s
Table 1. NRA trunkline 100-year RP design metocean data (from KP 52 to KP 63).
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
section was consequently used as the basis for the physical
model testing programme and subsequent 3D dynamic
stability assessment.
Rock-berm stability assessment
In parallel with the pipeline-stability assessment, the
stability of the rock berms (which were installed as part of
the remedial stabilization project in 1991) was re-analysed
using the industrys latest reliable software. The rock-berm
stability software package PROBED [7] was used to calculate
the minimum rock-armour layer D
50
values that would be
statically stable for the 100-year return period conditions.
The PROBED software package has been developed by Delft
Hydraulics and is based on tests performed on schematized
structures. It allows for the design of graded rock structures
that are subjected to a combination of steady-state currents
and oscillating currents induced by non-breaking waves. The
software uses empirical and semi-empirical design equations
2D pipeline-stability assessment
The limitations of the available pipeline-stability
recommended practices [1, 2] were assessed in great detail.
It was decided that, initially, a 2D analysis using a modified
RP F109 approach would be performed based on absolute
stability criteria. In view of the relatively low pipeline
SG and the fact that the pipeline had been trenched
following installation, it was a safe assumption that should
pipeline break out occur, instability had been reached to
an unacceptable degree. It was recognized that this was a
conservative approach, with the aim of identifying which
areas of the trunkline needed further assessment. Thus, the
results from the 2D stability analysis would then be utilized
to prepare the scope of work for a more-realistic, but also
more-complex, 3D FEA analysis.
The results of the 2D stability assessment are summarized in
Fig.5. The region between KP 52 and KP 63 was identified
as the most likely to experience instability: this 11-km route
Fig.5. Pipeline signifcant stability results between KP 28 and KP 124.
Fig.6. The O-tube hydraulic testing facility at UWA.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
to be incorporated in the design assessment, it was decided
that physical model testing would be performed to provide
additional information specific to the conditions relevant to
this pipeline. To achieve the required results the minimum
parameters of a testing facility were defined, which resulted
in the following main conclusions:
It was considered impractical to build a facility which
would enable testing a 40-in weight-coated pipeline at
the prototype scale. To practically model the tripartite
interaction between hydrodynamics, seabed soils,
and pipeline, scaling would need to be limited to a
maximum of 1:5 1:6; the quantification of scour
processes around pipelines become increasingly
difficult to model at smaller scales.
At such a scale, the use of an open wave and current
flume would be impractical, requiring a flume depth
of at least 10 m with the ability to concurrently
model wave-induced, as well as steady-state, currents.
Existing conventional open flumes are plagued by
wave breaking and non-linear affects.
U-tubes, which have commonly been used in the past
for similar work, have significant limitations in relation
to varying the wave periods as well as including steady-
state currents. It is difficult to control the frequency
in a U-tube much away from its natural frequency.
After several meetings in which the physical model
testing aspects were discussed, the concept of the
O-tube [8] was developed by the University of Western
Australia specifically for this project. To obtain the
additional required funding to construct such an
ambitious facility, Woodside formed a collaboration
with Chevron Australia with the aim of undertaking
additional testing over and above that required for the
NRA trunklines stability assessment. In addition,
federal funding was successfully applied for through
the Australian Research Council.
A number of scaled physical model tests were performed in
the O-tube facility for various symmetric and asymmetric
initial embedment profiles. The tests were performed using
an appropriately scaled representation of a 100-year return
period irregular wave-induced and steady-state current storm
time series. Seven realizations of complete on-bottom wave and
current velocity storm time series were generated. The storm
realization with the largest peak flow velocity was selected as
the base-case flow velocity time series for use in the physical
model testing programme.
The following key data were measured from each test for use
in the subsequent 3D FEA stability model:
The lateral and vertical hydrodynamic loads were
measured during each test as a function of time.
In addition, the profile of the artificial seabed, created
using sediments sourced from the Northwest Shelf,
was monitored and measured throughout each test
using a small echo-sounder.
based on wave height, critical Shields parameter, shear stress,
and other inputs to calculate a D
50
rock particle size. Some
engineering judgement is required as graded rock structures
are essentially a collection of graded rock particles, and
the ability of an individual rock particle to withstand the
design hydrodynamic load depends not just on the weight
and dimensions of the particle, but also on the level of
protrusion and interlocking with adjacent particles.
The analysis performed using PROBED indicated that the
rock berm along the trunkline met the design functional
requirements.
Physical model testing
The 2D pipeline-stability analysis provided information as
to which sections of the pipeline along the area of interest
were critical from a hydrodynamic-stability point of view. It
was identified that within the limitations of existing design
codes it was not possible to demonstrate that the pipeline
satisfied on-bottom stability requirements. However, it was
recognized that these limitations overlooked potentially
beneficial effects from seabed mobility. To allow such effects
Fig.7. Extract of test results (example).
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
diameter subjected to a 100-year return period hydrodynamic
load in the form of a storm time-history with 3 hrs ramp-up,
3 hrs peak storm duration, and 3 hrs ramp-down (prototype
timescale), and with a scaled-down combination of irregular
wave-induced and steady-state currents.
It was noticed that seabed instability (local scour), for this
particular test, occurred before the peak of the storm, and
well before the model pipe became unstable. The model pipe
was displaced laterally after approximately 4 hrs (prototype)
from the start of the storm, upon which the model pipe
was restrained laterally (but not vertically i.e. constant
SG) to monitor the seabed response to hydrodynamic
loading in the event that adjacent sections of the pipeline
would be stable (due to more embedment and/or reduced
hydrodynamic loading).

The measured forces and seabed data of all testing performed
were subsequently analysed and captured in a numerical
model for input into the 3D finite-element analyses.
Soil lateral-resistance model
The development of a reliable and realistic pipe-soil resistance
model for the carbonate soil used here was a key element of
the trunkline stability assessment. Pipe pull-out tests were
performed in the O-tube using the model pipe embedded
into the seabed soils, for a range of embedment levels, to
define a pipe-soil resistance model specific for this pipeline
and soil combination.
DNV RP-F109 [2] recommends modelling soil resistance
using the Verley and Sotberg soil passive-resistance model
[9] in combination with a Coulomb friction factor of 0.6.
According to this model, the Coulomb friction factor is
Ancillary tests were also performed to provide additional
information needed for the stability analysis, including:
Hydrodynamic loading of the test pipe on a rigid
seabed, to obtain lift and drag forces for a range of
KC numbers.
Pull-out tests to define a pipe-soil resistance model
specific for this pipeline and soil type.
It is recognized that the 1:5.8 scaling used for the testing
programme introduced several scaling issues, particularly
in view of the inability to use a similarly scaled soil for the
tests (the tests were performed using prototype soil). This
problem was addressed during the preparation phase of
the physical model testing campaign. An assessment was
therefore performed, in particular to quantify the level
of error associated with the onset and extent of scour
development in the physical model relative to what was
expected in the prototype.
Despite the inevitable scale of 1:5.8 used during the testing
campaign in the O-tube, it is noted that this is closer to full
scale compared to previous (similar) physical model testing.
To date, physical model tests involving subsea pipelines
have typically been performed at scales of 1:20 or smaller,
and this leads to scour-scaling distortions in the order of
100 to 200%. In comparison, the time required to reach
equilibrium scour in the O-tube has been assessed to be
approximately 35% relative to prototype scale. The soil-
particle size will not affect the equilibrium scour depth or
the initiation of local scour.
An example of the results of one test is provided in Figs 7 and
8. This particular test was intended to simulate a 2D pipeline
slice embedded symmetrically to a level of 50% of its external
Fig.8. Extract of test results (example), lift force (top), and drag force (bottom).
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
the current embedment through cyclic movements from
the surface: the pipeline was post-trenched by ploughing
following installation, and it is suspected that reduction of
its embedment has subsequently occurred during cyclonic
activity. The O-tube method of placing sand around the pipe
to achieve embedment is considered more representative of
this situation, compared to the cyclic-oscillation process used
to achieve embedment in the database of earlier pipe-soil
model tests from which the Verley model was calibrated.
The Verley model is based on a force-displacement curve,
as shown in Fig.9, where F
r1
, F
r2
, and F
r3
represent the peak
elastic, peak break-out resistance, and residual passive forces
respectively, and y represents the corresponding lateral
mobilization distances.
Relative to the O-tube (and other similar soil models
developed for pipelines in the Northwest Shelf of Western
Australia), the Verley and Sotberg passive-resistance model
significantly overestimates both peak breakout resistance
(F
r2
) and peak break-out distance (y
2
). This is consistent with
findings recorded in the SAFEBUCK JIP [10].
For the purpose of this project, the Verley model has been
modified in order to alter the magnitude and shape of the
passive-resistance curve to be in line with the O-tube pull-out
assumed to be constant, while the passive resistance is
dependent on both the pipe-soil vertical contact force and
pipeline embedment.
The Verley model has been obtained from physical model
tests performed on silica soils, while the NRA trunkline is
located in carbonate soils. In addition to this, the method
used to achieve pipeline embedment with the Verley model is
different from the method used for the O-tube pull-out tests.
The physical model tests used to develop the Verley model
achieved pipeline embedment by running a predefined
number of small, constant-amplitude, oscillations. In the
O-tube, the specified embedment was achieved by positioning
the model pipe in a prepared trench with a flat seabed either
side, prior to a test commencement. In all the O-tube tests
performed, there was no evidence of pipeline embedment
generated through a process of small pipeline oscillations.
In all cases, the test pipe remained essentially static until
a large wave-induced current caused displacement leading
directly to break out, or loosening the soil around the model
pipe through local scour, allowing subsequent wave-induced
current oscillations to complete the break-out process. This
suggests that the preparation method used for the O-tube
pull-out test is the more suitable for the development of
a soil-resistance model capable of analysing the stability
of the NRA trunkline. The trunkline did not achieve
Fig.9. Verley pipe-soil passive-resistance model [9].
Fig.10. Modifed Verley pipe-soil passive-resistance model.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
The analysis software was validated against the O-tube initial
embedment test results, which included scaling the test results
from the model scale to prototype scale. The following steps
were undertaken to define the sub-sections to be analysed:
A three-point moving average was applied to the
pipeline embedment data to define the initial
embedment profiles at 5-m intervals along the
pipeline.
The initial embedment level along the trunkline
varies, and eight such embedment cases were
modelled in the O-Tube (i.e. in terms of left:right
embedment level as a percentage of pipeline diameter
(D): 0D:0D, 0D:50D, 0D:75D, 50D:0D, 50D:50D,
50D:75D, 75D:75D, and 90D:90D). The initial
embedment profile (at 5-m intervals) was rounded
down to the nearest test profile. For instance, where
the survey data shows a section of the trunkline
route with an initial embedment of 0D:25D, the
initial embedment has been rounded down to the
nearest available test data, which is the 0D:0D initial
embedment test data. This is considered to be a
conservative approach.
The extent of each analysis model has been
determined by identifying regions that will provide
highly stable, effectively fixed-end, conditions such
as long regions (50-100 m) of fully buried or rock
dumped pipeline.
The schematic presented in Fig.11 is a representation of the
loads applied to the beam in an analysis time increment.

Based on the 3D dynamic stability analysis results, it
was concluded that the most critical region of trunkline
(between KP 52 to KP 63) is not expected to break out of its
test data, and comparisons between the modified model and
the O-tube pull-out tests are shown in Fig10. This allows the
un-conservatism inherent in the Verley model when applied
to this particular carbonate soil type to be removed, while
maintaining the generic form of loading-unloading-break-
out behaviour described within the Verley model in the 3D
dynamic FEA modelling of the NRA trunkline.
3D fnite-element analyses
A 3D finite-element pipeline stability analysis was performed
to translate the results from the 2D scaled physical model
testing to the 3D prototype 40-in diameter pipeline. This
dynamic on-bottom stability analysis was performed using
the CORUS-3D analysis software [11]. The 11-km section
of trunkline route (KP 52 - KP 63) was partitioned into
ten sub-sections, varying in length from 220 m up to 1,830
m, with each sub-section analysed independently. The
assessment was performed by subjecting the trunkline, in its
current condition, to a 100-year return period hydrodynamic
loading condition.
The objective of using the CORUS-3D analysis software over
and above the 2D physical model testing was to account
for the stabilizing effects of adjacent pipeline sections with
higher initial embedment levels on sections of pipeline
which would otherwise be considered unstable in 2D.
CORUS-3D is also able to reduce the conservatism present
in the 2D modelling by simulating 3D wave loading.
This FEA model incorporated the interactions that exist
between a pipeline and the surrounding fluid (hydrodynamic
effects), a pipeline and the seabed (passive resistance) and,
to a limited extent, the dynamic interaction between the
seabed and the fluid.
Fig.11. Schematic of the loads applied during an analysis increment.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
to provide input into and validate the 3D FEA software was
strategic and effective.
Acknowledgements
The writers of this paper are grateful for the contribution by
the following people and organizations: Neil Kavanagh, Nino
Fogliani, Andrew Pearce, and Alistair Walker from Woodside
Energy Ltd; Olav Aamlid and Jorn Spiten from Det Norske
Veritas; Carl Erbrich from Advanced Geomechanics; Gary
Parslow from BP; Chevron Australia; and the Australian
Research Council.
References
1. Det Norske Veritas, 1988. RP E305: On-bottom stability
of submarine pipelines.
2. Det Norske Veritas, 2007. RP F109: On-bottom stability
design of submarine pipelines.
3. A.C.Palmer, 1996. A flaw in the conventional approach
to stability design of pipelines, Offshore Pipeline
Technology Conference, Amsterdam, Netherlands.
4. D.Bonjean, C.T.Erbrich, and J.Zhang, 2008. Pipeline
flotation in liquefiable soil. Proc. Offshore Technology
Conference, Houston, Paper OTC 19668.
5. R.Soulsby, 1997. Dynamics of marine sands.
H.R.Wallingford, Thomas Telford.
6. N.Z.Flores and J.F.A.Sleath, 1998. Mobile layer in
oscillatory sheet flow. J. Geophy. Research, 103, C6,
pp 12783 12793.
7. Delft Hydraulics, 2006. PROBED: Design tool for bed
protections subjected to currents and waves.
8. L.Cheng, D.White, A.C.Palmer, E.P.Jas, A.Czajko,
A.Fogliani, R.Fricke, and H.W.An, 2010. A new facility
for research on the stability of pipelines on unstable
seabeds. Offshore Pipeline Technology Conference,
Amsterdam, Netherlands.
9. R.L.T.Verley and T.Sotberg, 1994. A soil resistance
mModel for pipelines placed on sandy soils. J. Offshore
Mechanics and Arctic Engineering, 116, pp145-153.
10. D.J.White, 2010. Safebuck Joint Industry Project Pipe-
soil interaction models for lateral buckling design Phase
II: A data review for sandy soils. Centre for Offshore
Foundation Systems Geomechanics Group UWA, Doc.
No. GEO10520v2.
11. Atteris Pty Ltd, 2011. CORUS 3D On-bottom
stabilisation analysis software for submarine pipelines.
embedment during 100 year return period conditions due
to hydrodynamic loads alone. Furthermore the analysis was
shown to have a significant margin of safety, as demonstrated
by the fact that the pipeline remains essentially stable even
considering reduced soil passive resistance of up to 70%.
The effects of seabed mobility were incorporated in the 3D
FEA model by applying at each pipe element the changing
embedment experienced in the relevant 2D physical model
tests, which captured the scour during the life of the storm.
It was demonstrated, however, that the potential exists for
lateral buckling driving loads, acting in combination with
the hydrodynamic loads, to cause sections of the pipeline to
break out. However, although the pipeline may break out of
its embedment at discrete locations, this will not result in
overstressing, and is not considered likely to lead to global
instability of the pipeline. Consequently it is considered that
this risk can be adequately addressed through monitoring
of the pipeline following significant tropical storms, and
intervention to restabilize the pipeline as necessary.
Conclusions
The need for the application of a new pipeline stability
analysis method became apparent during the early stages of
the NRA trunkline stability assessment. It was decided to
perform physical model testing to identify seabed mobility
effects, and apply these observations to a 3D FEA program
to simulate the prototype pipeline. Existing 3D FEA software
packages, which are commonly used in the industry, were
viewed to be inadequate due to their inability to model seabed
material instability. The significant effect of seabed material
instability on pipeline stability is a proven phenomenon
and has been addressed in the past by Professor Andrew
Palmer [3]; the physical model testing performed as part of
this project confirms this.
The application of the 3D dynamic FEA software package
described in this paper, and which simulates pipeline
response to hydrodynamic loading while accounting for
the effects of seabed instability during ramp-up, peak, and
ramp down of a severe storm, has proven to be effective
in assessing the stability of a pipeline in a mobile seabed
environment.
The decision taken to develop a custom-built physical
model testing facility and to run a series of large-scale tests
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
P
IPELINE RELIABILITY analysis appears at frst sight to be related to the probability analysis to which
everyone is accustomed. In reality, it is substantially different, and the numerical failure probabilities it
arrives at are nominal and unrelated to real probabilities. This matters because it misleads the engineer
and the wider community, and because it may lead to an illusion of confdence and safety that the analysis
and the underlying data do not begin to justify. The paper discusses the problem, and how codes might be
better written.
Authors contact details:
tel: +65 6516 4601
email: ceepalme@nus.edu.sg
by Prof. Andrew Palmer
Keppel Professor, Centre for Offshore Research and Engineering, National University of Singapore
10
-6
and all that: what do failure
probabilities mean?
T
HE NOTION OF probability was developed by gamblers.
Cardano (1501-1576) asked how many times you would
need to throw two dice to have an even chance of two sixes.
He is described in one book [1] as physician, philosopher,
scientist, astrologer, religionist, gambler, murderer, and as
it happens he got the wrong answer. The ideas were picked
up and developed by many mathematicians, some of them
known in the context of pipelines, among them Pascal, De
Moivre, Bernoulli, Fermat, Gauss, and Poisson. If you throw
a fair dice, the probability of a six is 1/6. If you throw two
dice, the probability that the sum of the pips will be 10 is
3/36, and so on. We have a clear idea of what that means,
and we can use it to inform decisions.
Turning to pipelines, we frequently come across statements
such as:
the nominal target failure probability level shall be
based on the failure type and safety class as given in
Table 2-5[2, section 2 C 503]
and there follows a table which says that the nominal failure
probability per pipeline per year for safety class very high
shall be 10
-6
for ultimate, fatigue, and accidental limit states,
and 10
-7
to 10
-8
for pressure containment, and so on.
A reasonable question to ask is how those failure
probabilities relate to the customary understanding of
what probability means. Or does the weasel word nominal
take care of the question?
The point was made more forcefully by Richard Feynman
[3] in his account of his discussions with NASA following
the Challenger space shuttle disaster:
As range safety officer at Kennedy, Mr Ullian had to
decide whether to put destruct charges on the shuttle.
(If a rocket goes out of control, the destruct charges
enable it to be blown into small bits. Thats much less
perilous than a rocket flying around loose, ready to
explode when it hits the ground.)
Every unmanned rocket has these charges. Mr Ullian
told us that 5 out of 127 rockets that he had looked
at had failed a rate of about 4%. He took that 4%
and divided it by 4, because he assumed a manned
flight would be safer than an unmanned one. He came
out with about a 1% chance of failure, and that was
enough to warrant the destruct charges.
But NASA told Mr Ullian that the probability of
failure was more like 1 in 105.
I tried to make sense of that number. Did you say
1 in 105?
Thats right: 1 in 100,000.
That means you could fly the shuttle every day for
an average of 300 years between accidents every day,
one flight, for 300 years which is obviously crazy
Yes, I know said Mr Ullian.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Again there are standard responses. One is to assert that
the probability distribution of each of the relevant variables
is one of the standard distributions such as Gaussian or
log-normal applies, so that the distribution is completely
characterized by a mean and a variance. That assertion is
totally unjustified: there is no reason at all why a form that
describes the middle should also describe the distant tails. It
has been well said that mathematicians believe the Gaussian
distribution to be a law of physics, and physicists believe it
to be a law of mathematics, but of course neither is correct.
Some arguments appeal the Central Limit Theorem, but
that is unlikely to be applicable.
The other standard response is to revert to the argument
described earlier, and to say that the calculated probabilities
are only nominal.
The difficulty has of course been recognized. The DNV
classification note [8] has this to say:
The analysis models are usually imperfect, and the
information about loads and resistances is usually
incomplete. The reliability as assessed by reliability
methods is therefore generally not a purely physical
property of the structure in its environment of actions,
but rather a nominal measure of the safety of the
structure, given a certain analysis model and a certain
amount and quantity of information.
Correspondingly, also the estimated failure probability
is dependent on the analysis model and the level
of information, and it can therefore usually not be
interpreted as the frequency of occurrence of failure for
that particular type of structure. An ideal frequentistic
interpretation of the estimated failure probability would
require a large population of the particular type of
structure in conjunction with perfect analysis models
and full knowledge about the governing uncertainties.
This will practically never be fulfilled.
which acknowledges the difficulty. It might be thought better
to replace usually by always in the first paragraph, to delete
generally later in the same paragraph, to replace usually
not by never in the second, and to delete practically in
the last sentence.
Does it matter? How to move on
Ideally, we would throw out all these spurious numbers. They
are not probabilities, and they do not help us to make decisions.
The reality is that ideas of this kind nowadays have so much
momentum, and have an industry of structural reliability
analysts in the background, that it is probably politically
unacceptable to dispense with them completely. Another
possibility is that we make a careful distinction between
the different ideas, retain the term probability for the
A conventional reply
This point has of course been made many times before
[4-6], and Goldberg [4] amusingly cites still more bizarre
examples, such as an electronic component with a failure
rate quoted as 5.93 10
-92
per hour.
The proponents of structural reliability analysis have a
ready-made response. You are being nave, they say, you
are not meant to treat those probabilities as it they were the
same kind of frequentist probability you are used to. That
is why we call them nominal. They express some kind of
confidence in the safety of the system.
The difficulty with that response is that it conflates two
quite different ideas.
When we attach a number to a failure probability, we are
consciously or semi-consciously exploiting the conventional
understanding of what that number signifies, and trying to
give ourselves or someone else confidence from it:
10
-6
per pipeline per year! That is pretty good, isnt
it, the public can sleep easily. A thousand pipelines
for a thousand years, and only one failure.
In reality, numbers like that neither say what they mean
nor mean what they say. They have no meaning that could
possibly be justified by any calculation or any available data.
If the number is taken seriously, and is not just some kind of
hollow public-relations exercise, the difficulty is that people
will come to take decisions based on the numbers, as the
NASA example demonstrates. In that instance, NASA made
a foolish statement, without any justification. Fortunately,
Ullian was more thoughtful, arrived at a rough-and-ready
but justifiable calculation, and based his decision on it.
Tail sensitivity
Much of the difficulty is that the data that ought to be used
to justify the numbers are not available (and can never be
available). That point too is obvious, and again has been
made many times. It is illustrated schematically in Fig.1 which
plots a probability density distribution for strength against
a strength parameter: it can of course trivially be extended
to multiple loads and multiple components of strength.
If the probability of failure is to be 10
-6
, the important part of
the probability density distribution is the extreme left-hand
tail. The data such as they are are all near the middle of
the probability density distribution, but the middle of the
distribution is of no importance to the failure probability. There
are no data from the left-hand end, but that is the only part
of the distribution that is important to the failure probability.
Tail sensitivity is the central problem of structural reliability,
but theorists devote astonishingly little time to it. Melchers
book [7], for instance, gives to it half a page.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
That suggests that clauses C501 through C503 could be
replaced by:
Compliance with the LRFD format used in
this standard, incorporating the application of
the numerical values given for each safety class,
ensures that the design achieves an acceptable level
of safety.
References
1. R.A.Epstein, 1977. The theory of gambling and
statistical logic. Academic Press, New York.
2. DNV, 2007. OS-F101Submarine pipeline systems. Det
Norske Veritas, Hvik, Norway.
3. R.Feynman, 1988. What do you care what other people
think? W.W.Norton.
4. H.Goldberg, 1981. Extending the limits of reliability
theory. Wiley, New York.
5. A.C. Palmer, 1996. The limits of reliability theory and
the reliability of limit-state theory applied to pipelines.
Proc. 28th Annual Offshore Technology Conference,
Houston, 4, 619-626, OTC8218.
6. A.C.Palmer, C.Middleton, and V.Hogg, 2000. The
tail sensitivity problem, proof testing and structural
reliability theory. Structural Integrity in the 21
st
Century, Proceedings, Fifth International Conference
on Engineering Structural Integrity Assessment,
Cambridge, 435-442.
7. R.E.Melchers, 1988. Structural reliability: analysis and
prediction. Ellis Horwood, Chichester.
8. DNV, 1992. Classification note 30.6 Structural
reliability analysis of marine structures. Det Norske
Veritas, Hvik, Norway.
9. B.Mller and M.Beer, 2008. Engineering computation
under uncertainty- capabilities of non-traditional
models. Computers and Structures, 86, 1024-1041.
frequentist interpretation, and call the numbers from the
other interpretation psuedoprobabilities. The addition of
the word nominal is already a long step in that direction.
However, the prefix psuedo is likely to be unacceptable.
It would be wrong to think that there is nothing to be done
beyond nave probability theory. Much effort is being given
to more sophisticated approaches, particularly in the context
of decisions. Mller and Beer [9] have written a useful review.
Anyway, all this only matters if the numbers are going to be
used to help make some decision, and that is rather unlikely.
What we can do is to incorporate sensible wording into codes.
Codes
Most users of codes rightly do not trouble themselves with
the analysis. Confronted with the DNV wording [2, clauses
2 C 501 and 503] for instance:
C501 As an alternative to the LRFD format specified
and used in this standard, a recognized structural
reliability analysis (SRA) may be applied
C503 As far as possible, nominal target failure
probability levels shall be calibrated against identical
or similar pipeline designs that are known to have
adequate safety on the basis of this standard.
A user sensibly opts for the LRFD format given and used
in this standard. He concludes that DNV has based its
conclusions on some analysis of its own, and that if he
sticks to the numbers given elsewhere in the code, he can
argue that the target reliability level for each safety class
has been reached. He can leave the C501 alternative on
one side. Only the most conscientious and unusual user
would embark on the SRA approach, and if he were to do
so he would soon encounter difficulties that could not be
surmounted in any honest way.
Fig.1. Tail sensitivity.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
MAY 1516 2013
MARRIOTT WESTCHASE HOTEL
HOUSTON, TX, USA
Exhibition and sponsorship
options available, visit
www.clarion.org
UPSF_13_FP_Ad.indd 1 8/11/12 9:45 AM
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
T
HE INCREASING demand for oil and gas, currently estimated at 135 million barrels of oil equivalent
per day, keeps pushing the boundaries of offshore engineering into ever-deeper waters. For instance, in
the Gulf of Mexico, exploration and production activities are performed in water depths exceeding 3000
m. Such remote locations and challenging environments call for new procedures and solutions in the design
and installation of offshore pipelines.
In this paper, numerical modelling and analysis of offshore pipelines is reviewed and discussed. Finite-element
techniques to assist in pipeline design are introduced, and applied to pipeline routeing optimization. Special
emphasis is devoted to out-of-straightness and on-bottom stress analysis.
Contact algorithms allowing the simulation of pipelaying on an uneven seabed (using bathymetry) are
reviewed, and recent developments in modelling of pipe-soil interaction are highlighted. The importance
of free-span detection and evaluation is stressed. In addition, it is shown how fnite-element analysis can
contribute to the prediction and mitigation of both upheaval and lateral buckling of subsea pipes. At the
end of this paper, pipeline walking on an inclined seabed is simulated, and the importance of seabed friction
on the walking rate is demonstrated.
*Corresponding authors contact details
tel: +32 497 548 916
email: filip@vikar.be
by Dr Filip Van den Abeele* and Raphael Denis
Fugro GeoConsulting Belgium, Brussels, Belgium
Numerical modelling and analysis
for offshore pipeline design,
installation, and operation
O
IL AND GAS exploration and production is embarking
into ever greater water depths. Consequently, offshore
pipeline engineering is continuously pushing the boundaries,
installing flowlines and export pipelines in water depths
exceeding 3000 m. The availability of high-performance
computing systems and dedicated software tools enable
pipeline engineers to cope with the challenges associated
with design of subsea completions.
In this paper, an overview is presented of numerical
modelling and analysis for offshore pipeline design,
installation, and operation. SAGE Profile 3D [1-3] is used
to demonstrate the added value of numerical modelling as
a design aid and decision tool throughout the entire life of
an offshore pipeline, covering:
preliminary pipeline design
route selection and optimization
offshore pipeline installation
free-span assessment
on-bottom stress analysis
SAGE Profile 3D uses a transient dynamic explicit integration
kernel, which enables the efficient simulation of the
pipelaying process and the response of the subsea pipe
when subjected to hydrodynamic loading and operational
conditions (time-dependent pressure and temperature
profiles). In this paper, the numerical algorithms governing
pipeline laydown, pipe-soil interaction, and numerical
integration are briefly covered, and some examples on
free-span evaluation, lateral buckling, upheaval buckling,
and pipeline walking are highlighted to demonstrate the
versatility of finite-element methods as a powerful support
tool in offshore pipeline design.
Pipeline route selection and
optimization
One of the early tasks for the pipeline engineer is to determine
the preliminary route and evaluate the feasibility of the
selected pipeline corridor. An informed route selection
cannot be made without information on the seabed
topography and geotechnical data [4].
Performing an initial desk study before embarking on
an extensive (and expensive) marine survey can save a
considerable amount of time and money [5]. In SAGE Profile,
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
of seabed elevation versus KP is updated simultaneously,
which allows evaluating the on-bottom roughness of the
selected route already during pre-processing, without any
requirement for computing power.
At the same time, the allowable bending radii can be quickly
screened. Each pipeline bend radius R should be large
enough to ensure that the bending stresses do not exceed
the allowable stress
a
:
! >
! !
!
2 !
!

(1)
where E is the Youngs modulus of the pipeline steel and
Do is the outer diameter of the pipe. Moreover, the pipeline
requires sufficient frictional force to resist being dragged
over the seabed by the lay barge. Hence:
! >
!
! !
!

(2)
with the lateral friction factor, T the lay tension, and w
s
the
submerged weight per unit length. In addition to bathymetric
considerations, selection of the optimum pipeline route also
depends on a broad spectrum of other factors, including:
politics and regulatory requirements
crossing of existing pipelines or submarine cables
iceberg plough marks, pockmarks
areas of very soft or very hard seabed
boulder fields, rock outcrops
risk of anchor damage and trawling gear impact
proximity of other subsea installations
cost-efficiency of installation
environmental and ecological issues

The SAGE Profile pre-processor allows introduction of
different layers of information, by importing additional
information such as admiralty charts, test locations, existing
pipelines, and shipwrecks. In Fig.3, for instance, a proposed
pipeline route is shown on a digital-terrain model and, in
addition, an overlay plot is made to display data associated
the seabed topology can easily be created or imported from
survey data, either as:
kilometre point (KP) versus seabed elevation
Easting-Northing-elevation (ENE) coordinates
full 3D digital-terrain model (DTM)
In Fig.1, two corridors imported from survey data are
compared. In the northern corridor, a curved pipeline
route has been drawn, whilst a straight pipeline section is
proposed for the southern part.
The pipeline route can be easily imported, or constructed
through a user-friendly and straightforward graphical
interface. This interface will convert the constructed route
automatically into a proprietary route format, with successive
sections of straight lines and circle arcs. The straight sections
(like the green route in Fig.1) are defined by a start and end
point, whereas the circular arcs (for example, the middle
section of the red route shown in Fig.1) are defined by the
tangent points and the centre of the circle linking these
tangent points.
As demonstrated in Fig.2, the user interface enables an early
assessment of seabed topography and on-bottom roughness.
Whilst modifying the proposed pipeline route, the graph
Fig.2. Early assessment of seabed topography.
Fig.1. 3D digital-terrain model based on survey data.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
at either side by nodes. The distributed mass of the pipe
is lumped at these nodes. The finite-element kernel uses
an explicit solver, which computes the dynamic motion
of the pipe and is therefore ideally suited to simulate the
pipelaying process.
During this pipeline installation process, new pipe elements
are continuously created and the pipe is laid along the target
path defined on the seabed. The lay tension T, applied at
the barge, is used as an input and the unstressed length
L
0
of the last element is updated such that the axial force
corresponds to the applied lay tension:
! !
!
!
!
!" !
!
= !
(3)
with L the original element length,
! =
!
4
!
!
!
!
!
!

(4)
the cross-sectional area of a circular pipe with inner diameter
D
i
and outer diameter D
o
, and
!
!
= 1 2! !
!
!
!
!
!
!
!

(5)
the pressure induced axial force component, accounting
for both the internal pressure p
i
and the (hydrostatic)
external pressure p
o
. As a result, both empty and water-filled
installation can be simulated. In Equn 3, is the Poissons
coefficient of the pipeline steel, where A
i
and A
o
are the surface
areas of the interior and exterior of the pipe respectively.
When the unstressed element length:
!
!
=
! ! !
! +!
!
+! !

(6)
with the pockmarks. This layered presentation of information
offers the pipeline designer an intuitive dashboard with a
wealth of data to select the most appropriate pipeline route.
In addition to overlay plots, contour maps, and slope angles
can easily be visualized, which provides additional input to
assess potential geohazards.
Simulating pipe laydown and
installation
Offshore pipeline installation is performed from a laybarge,
typically in S-lay configuration. For smaller diameters,
pipeline reeling can be the most cost efficient solution,
whereas J-lay is the only feasible approach in (ultra-) deep
water. Depending on the installation method, the pipeline
is subjected to different load patterns during installation,
including hydrostatic pressure, lay tension, and bending on
the stinger and in the sagbend. A comprehensive overview
on the mechanics of installation design can be found in [6].
The simulation of the pipelaying process is one of the most
challenging tasks once the optimum route has been selected.
Implementing pipeline installation in a general-purpose
finite-element package can be a time-consuming and tedious
job, in particular when importing vast amounts of seabed
data. Most often, advanced scripting techniques are required
to define the seabed profile, select the optimum pipeline
route, and simulate the laydown process. In addition, the
available constitutive models for pipe-soil interaction may
not comply with industry standards.
Finite element tools like SAGE Profile have been tailored
to assist the pipeline engineer during offshore pipeline
design. Using an explicit integration algorithm, the actual
pipeline-installation process can be approximated. The pipe
is simulated by discretising the entire pipeline into section
of finite length. These sections are represented by beam
elements with 12 degrees of freedom (DOF), bounded
Fig.3. Digital-terrain model with pockmark indications.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
where is the angle between the pipe and the target path,
and h is the height of the feeding point above the seabed.
Replacing the laybarge with a feeding point close to the
seabed allows for a significant reduction in calculation time,
without losing accuracy. Given the inherent complexity of
pipeline laying, an accurate and robust steering mechanism
of the feeding point is of paramount importance. In SAGE
Profile, this steering mechanism is governed by a proportional-
integrating-differentiating (PID) controller, providing a
smooth movement of the feeding point and ensuring that
the pipeline is installed on the pre-defined target path
(shown in red in Fig.5).
In addition to the concept of a feeding point, an efficient
element-killing procedure has been implemented to
control the computational effort during pipeline laydown.
Indeed, it would be too expensive to simulate the entire
length of the pipe from its starting point up to the feeding
point. In order to reduce the required calculation time,
elements that are already lying on the seabed and are no
longer moving will be removed from the simulation. If the
magnitude of the velocity vector for a node is lower than
a predefined threshold, the associated element has little
or no contribution to the simulation results and can be
killed without losing accuracy. In Fig.5, the elements that
have been killed are also shown.
Evaluation of free-spanning pipelines
Accurate prediction of free spans (location, length, and
height) is an important prerequisite in offshore pipeline
design. Indeed, free-span lengths should be maintained
within an allowable range [7], which is determined during
the design phase. Pipelines installed on a very rough
seabed can cause a high number of free spans that can be
difficult to rectify. A judicious assessment of free spans
can dramatically reduce the costs associated with seabed
intervention (trenching, rock dumping, and span supports).

Figure 6 demonstrates that finite-element analyses enable
the simulation of pipeline installation on an uneven seabed,
and allow detection of free spans. The colour code on
Fig.6 reflects the local span height, i.e. the gap between
the pipeline and the seabed. After the pipelay simulation
has been completed, SAGE Profile automatically detects
the spans over the entire pipeline route, and plots the
span location, length and height in comprehensive and
easy-to-read design charts, as shown in Fig.7.
Once a free span that is longer than the allowable span length
occurs, the span may suffer from vortex-induced vibrations
(VIV) which can induce fatigue damage in the pipe. It was
only recently that the commonly used pipeline design codes
allow free vibrating spans, as long as the structural integrity
of the pipeline system remains assured [8].
Span checks can be performed to assess whether an installed
pipe is compliant with the guidelines recommended in
becomes longer than twice the initial length, the element
is split in two new elements. An additional node is placed
along the last element such that the newly formed element
obtains the original unstressed length. This algorithm
accurately reflects the continuous supply of welded pipe
joints from a moving laybarge. Gravity, applied during the
pipelay simulation, will force the newly created pipe elements
into place; Fig.4 shows the typical catenary shape during
pipeline installation.
For long pipelines and significant water depths, simulating
the entire laydown process (from the barge down to the
seabed) tends to be time-consuming and is computationally
expensive. The sophisticated architecture of the currently
available numerical solvers allows for a significant reduction
in the resources required to simulate pipeline laydown. By
default, the laybarge and most of the free-hanging pipe is
replaced by a single feeding point in the water column moving
close to the seabed, as shown in Fig.5. This feeding point
acts as a submarine laybarge, generating new pipe joints
as it moves forward. The lay tension is now applied at the
feeding point, generating a residual on-bottom tension in
the laid pipe section.
Assuming a catenary shape [6], the lay tension at the feeding
point can be expressed in terms of the submerged weight
per unit length w
s
:
! =
!
!
tan
!
!
1 + 1 + tan
!
!
(7)
Fig.4. Pipeline catenary shape during S-lay installation.
Fig.5. Defnition of feeding point and target path.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
of the pipe-soil interaction, which is the most important
parameter governing the design. The elastoplastic
constitutive behaviour of the pipeline steel can be described
by the Ramberg-Osgood equation [11-12], connecting pipe
DNV-RP-F105 [9]. For each detected span, SAGE Profile
will calculate the associated reduced velocity:
!
!
=
!
!
+!
!
!
!
!
!

(8)
where U
c
is the mean current velocity (normal to the pipe),
U
w
the significant wave-induced flow velocity, and f
1
an
approximation [9] for the lowest natural frequency given by:
!
!
1 +!"#
!"
!
!
!
!
!
1 +
!
!
!
!"
+ !
!
!
!
!
!

(9)
with SCF the stiffening effect of the concrete coating, L
e
the effective span length [10], m
e
the effective mass, F
e
the
effective axial force, the static deflection and C
3
the end
boundary coefficient. The moment of inertia for the hollow
circular pipe is given by
! =
!
64
!
!
!
!
!
!

(10)
and the critical buckling load can be calculated as
!
!"
= 1 +!"# !
!
!
!
!
!
!"
(11)
where C
2
is an end boundary coefficient as well.

In addition to the reduced velocity (Equn 8), the software
calculates the stability parameter:
!
!
= 4 !
!
!
!
!
!
!
!
!
!

(12)
for each span, where
T
is the total modal damping ratio,
comprising structural damping, hydrodynamic damping and
soil damping. Based on the values of the reduced velocity from
Equn 8 and the corresponding stability parameter from Equn
12, the software will check whether the conditions for the
onset of in-line or cross-flow VIV are met in full compliance
with DNV-RP-F105. This powerful capability provides a quick
and easy tool to evaluate the severity of free spans for a given
pipeline route, and hence can save a tremendous amount of
time and money associated with seabed rectification.
In the next sections, some operational analyses are presented
to evaluate the susceptibility of high-temperature subsea
pipelines for buckling and walking. First, some details and
recent developments on numerical modelling of pipe-soil
interaction are reviewed.
Numerical modelling of pipe-soil
interaction
The key to a successful simulation of offshore pipeline
installation and operation is a profound understanding
Fig.6. Free-spanning pipeline on an uneven seabed.
Fig.7. Overview of span location, height, and length.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Ramberg-Osgood formulation in SAGE Profile takes into
account the combined effects of plasticity, ovalization [14-
15], axial force, and hydrostatic pressure ([15-16].
The pipe is assumed to be in contact with the seabed when
the difference between the z-coordinate of a pipe node and
the corresponding seabed elevation at this (x,y) location
is less than the external pipe radius R
o
. Once contact has
been detected, a soil response will be exerted depending on
the type of seabed soil. The soil response is captured by a
combination of vertical, axial and lateral springs.
The bearing capacity Q
u
is reflected by the vertical soil
reaction. For sands, DNV-RP-F105 recommends:
!
!
!
!
=
!
!
!
!
2
!(!
!
) + !
!
!
!
!
!
!(!
!
)
(16)
where
s
is the submerged unit weight,
!
!
= exp ! tan! tan
!
!
4
+
!
2

(17)
with the friction angle, and
!
!
=
3
2
!
!
1 tan!
(18)
The bearing width B depends on the pipe penetration z
p
,
as is schematically shown in Fig.8, and can be calculated as:
! !
!
=
2 !
!
!
!
!
!
0 !
!
!
!
2

!
!
otherwise

(19)
For clays, DNV-RP-F105 recommends:
!
!
!
!
= 5.14 !
!
+!
!
!
!
! !
!

(20)
where C
u
is the undrained shear strength. Figure 9 compares
the vertical soil-spring reaction forces for a medium-dense
sand (with a friction angle = 33 and a submerged unit
weight
s
= 8.5 kN/m) with the soil reaction of a soft clay
(with undrained shear strength C
u
= 30 kPa and a submerged
unit weight
s
= 7.5 kN/m).
In addition to the vertical soil springs recommended by
DNV-RP-F105 [9], other soil models for both cohesive and
cohesionless materials are described in DNV- CN30.4 [17-
18]. For very soft clays (C
u
< 20 kPa), a buoyancy formulation
could be used, assuming that the soil behaves like a liquid and
that the soil-induced buoyancy of the pipeline is equal to the
vertical soil reaction:
!
!
!
!
=
!
!
6 !(!
!
)
3 !
!
!
+4!
!
(!
!
) !
!

(21)
curvature with bending moment M through:
!
!
!
=
!
!
!
+ !
!
!
!
!

(13)
where the nominal curvature
0
and bending moment M
0

are related by:
!
!
!
!
= !"
(14)
and the parameters and are chosen to fit the moment-
curvature relationship obtained by integrating the stresses
across the section A for a given curvature:

! = !
!!
! !"
!

(15a)
Hence, Equn 13 is equivalent to the well-known stress/
strain relationship [11, 13]:
! ! =
!
!
+ !
!
!
!
!

(15b)
with
y
the yield stress and {K,n} the parameters describing
the hardening behaviour of the steel grade. The enhanced
Fig.8. Bearing width as a function of pipe penetration.
Fig.9. Vertical soil reaction for sand and clay.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
The combination of a vertical, axial and lateral soil spring
fully defines the pipe-soil interaction. In addition to the
commonly used soil-spring models, presented here, SAGE
Profile offers dedicated and more enhanced soil models to
describe complex soil behaviour such as berm formation,
buried pipes, and trenching operations [19]. Moreover, an
application programming interface (API) can be used to access
an advanced soil library based on the incremental plasticity
approach described by Zhang [20-21]. In this approach, the
load-displacement relationship for an elastoplastic soil model
is expressed in its incremental form:
!" = ! !"
(24)
where the vector of incremental loads {dF} is connected to the
resulting displacements {dU} by the compliance matrix [C].
In addition to an extensive library of predefined soil models,
user-defined constitutive laws can be implemented as well to
construct the compliance matrix.
Accurate pipe-soil interaction is a key requirement for the
reliable prediction of the on-bottom behaviour of offshore
pipelines. Significant development efforts are being conducted
to continuously improve the predictive capability of the pipe-
soil interaction parameters [22]. Recently, much R&D effort
has been devoted to the development of coupled soil springs
to comply with the guidelines of the SAFEBUCK JIP [23-24].
In the next sections, some case studies on pipeline instability
(lateral buckling, upheaval buckling, and pipeline walking)
are presented to demonstrate the importance of pipe-soil
interaction in finite-element simulations. First, the numerical
architecture of the transient dynamic solver is briefly explained.
Loading patterns and explicit
integration
Offshore pipelines are subjected to hydrodynamic loading
(combined actions of currents and waves), internal and external
pressure, operational loads (temperature and pressure), and
external loads. As is schematically shown in Fig.11, these
loads can be either defined directly (for example, lay tension,
For rock, either a rigid seabed or clay with a (very) high
undrained shear strength, can be modelled. In addition to
the vertical soil spring, reflecting the bearing capacity of the
seabed, axial and lateral springs are included in the formulation
of pipe soil interaction. As shown in Fig.10, such springs allow
for energy dissipation through friction and have a memory
component which is instrumental for simulating plastic soil
deformation, for example the pipeline walking phenomenon
discussed below. As soon as contact is made with the seabed,
the coordinates of the first point of seabed touchdown are
saved for each node. The distance the node subsequently
travels while in contact with the seabed determines the amount
of compression exerted on the lateral and axial soil springs.
The input values required to define the axial and lateral
soil springs are:
the threshold displacement d
lim
after touchdown
the adhesion F
adh
the dimensionless friction factor
The distance d
f
to the touchdown point in axial and lateral
direction is computed, and converted into a friction force
defined by the soil springs. If the displacement is smaller than
the threshold, a friction force:
!
!
=
!
!
!
!
!
!"#

(22)
is applied, where the maximum friction force is given by:
!
!"#
= !
!"!
+ ! !
!

(23)
with F
V
the vertical load per unit length. If the displacement
is larger than the threshold, than the friction force is equal
to F
max
, i.e. the soil behaves as perfectly plastic. As indicated
in Fig.10, the unloading path differs from the loading curve,
which allows for frictional energy dissipation.
Fig.10. Frictional spring with memory component.
Fig.11. Load patterns acting on a subsea pipeline.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
and a drag force:
!
!
=
1
2
!
!
!
!
!
!
! !"#$ +!"#$% ! !"#$ +!"#$%
(26)
are imposed based on the Morisons equations [25], where
C
L
and C
D
are the lift and drag coefficients respectively.
On top of that, the wave induced acceleration a gives rise
to an inertia force:
!
!
= !
!
!
!

! !
!
!
4
! !"#$
(27)
with C
I
the inertia coefficient. For combined wave and
current action, the hydrodynamic coefficients {C
L
,C
D
,C
I
} can
be selected [26] based on the surface roughness on the pipe,
the Reynolds number Re, the Keulegan-Carpenter number K,
and the gap between the pipe and the seabed [27]. Current
and wave velocity and incidence angle can be supplied directly
by the user. The wave parameters can also be calculated based
on the JONSWAP spectrum [28]. In this modification of the
Pierson-Moskowitz spectrum [29] for a developing sea state in
a fetch-limited situation, the significant flow velocity amplitude
and the mean zero-up crossing period are calculated based on
linear Airy wave theory [30].
Finally, the pipeline loading history, which is fundamental for
the realistic simulation of its response, is tracked by defining
successive load cases. Loading sequences such as shut-down cycles,
can be modelled using restart capabilities, which allows a load
case to start from the final configuration of a previous condition.
All of the load patterns described above are converted to
nodal forces. The forces acting upon each node are summed,
resulting in an out-of-balance force F(t). According to Newtons
law, this force implies an acceleration:
! ! =
!(!)
!

(28)
with m the total mass lumped at this node.
To calculate the position of the pipeline over time, a transient
dynamic solver uses an explicit integration method. Indeed,
the velocities v are obtained from the acceleration using the
central difference integration scheme:
! ! +
!
2
= ! !
!
2
+ ! !(!)
(29)
and a similar scheme is used to update the nodal positions p:
! ! +! = ! ! + ! ! ! +
!
2

(30)
A similar scheme applies for the rotational degrees of
freedom. The explicit integration algorithm is conditionally
operational pressure and temperature) or modelled using
either uniformly distributed loads (such as to model buried
pipe sections), point loads (such as to reflect the additional
mass of sacrificial anodes), or pre-described displacements
(such as the pipeline being lifted by the plough grabs during
trenching operations).
Internal and external pipe pressure are modelled by taking
into account the water depth at each node. Temperature
profiles can vary with KP, reflecting the temperature gradient
between the hot end (close to the wellhead or manifold)
and the cold end (riser tie-in). Pressure and temperature
variations (as well as the residual bottom tension after lay-
down) contribute to the nodal forces. Assuming that waves
are approaching the pipeline with a velocity u and at an angle
, and the current with steady velocity V is approaching at
an angle , both a lift force:
!
!
=
1
2
!
!
!
!
!
!
! !"#$ +!"#$%
!

(25)
Fig.12. Introduction of quasi-static loading.
Fig.13. SAGE Profle simulation of lateral buckling.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
In Fig.13, the plan view of an operational pipeline is shown
when subjected to increasing temperatures. The simulation
has been performed with SAGE Profile, using the vertical
soil spring for medium dense sand (shown in Fig.9), and
lateral and axial friction factors equal to = 0.7.
The pipeline path deviates from its original route, and a
lateral buckle develops. In post-buckling behaviour, the
effective axial force in the buckle zone (which reached the
critical value to trigger instability) drops because of the
additional length that feeds into the buckle. Simulations
have shown [37] that the length of the buckle, its maximum
amplitude, and the effective axial force mainly depend on
the mechanical material properties, the pipeline geometry
and weight, and the pipe-soil interaction.
Susceptibility to upheaval buckling
Buried pipelines subjected to a large temperature increase
and axial restraint are prone to upheaval buckling. As
a benchmark, a straight 100-m long pipeline, pinned
at both ends, was simulated on a rigid frictionless soil.
An initial imperfection was introduced at mid-length to
invoke upheaval buckling. The SAGE Profile simulation
results, shown in Fig.14, show excellent agreement with
the Abaqus simulations for both the onset of buckling
and the post-buckling behaviour.
In addition to these predictive capabilities, the software
offers tools to prevent or mitigate the problems associated
with upheaval buckling. For instance, the tendency of
a buried pipe to lift is decreased by the weight of the
backfill soil and the shear resistance generated along the
potential failure surface in the backfill soil. This backfill
behaviour can be accounted for by including the submerged
weight of the soil cover, and incorporating the equivalent
shear resistance of the backfill using the Schaminee
model [38]:
!
!
= !
!
!
!
! 1 +!
!
!
!

(33)
stable [30], provided the time increment t is sufficiently
small. For the (dominant) axial deformation mode, critical
time step is closely connected to the acoustic velocity, and:
! < !
!

!
!
!
1 +!
!
!
(31)
must be satisfied to avoid that information propagates over
more than one element during one time increment. In (31),
s
is the density of steel, and is a damping factor. While
an explicit solver offers the potential for dynamic loading
and realistic pipelay simulation, the condition in Equn
31 indicates that the initial element length and the time
increment must be judiciously chosen to obtain convergence.
By introducing the concept of damping, the explicit (and
hence inherently dynamic) approach can also be used to
obtain solutions to (quasi-)static problems. Quasi-static load
cases are phased in over a period of time, called the ramping
time. Once the ramping time is reached, the load reaches
its full extent, and the pipe is given time to settle down
and reach equilibrium. This quasi-static load introduction
is explained on Fig.12.

The total sum of the kinetic energy over all nodes:
!
!"#
=
!
!
2
!
!!!
!
!
!
(!)
(32)
is a scalar value that indicates the energy content of the
entire pipeline as a function of time. As shown in Fig.12,
the simulation is assumed to have reached equilibrium when
the kinetic energy drops below a pre-defined threshold value
E
th
, or when the equilibrium time is reached.
Numerical prediction of lateral
buckling
Pipelines operating at high temperature are susceptible to
global buckling. The basics of buckling were first developed by
Euler [32], who established the critical load for long, slender,
structures under compression. In pipeline engineering, Hobbs
[33-34] was one of the first to develop a semi-empirical method
to calculate buckling. His approach was based on solving
the linear differential equation for the deflected shape of a
spring-supported beam-column under axial load. The most
important limitations of this method are the assumptions on
linear-elastic material and small rotations, and the idealized
straight pipeline.
It is recognized [35] that lateral-buckling modes tend to occur at
lower compressive forces than the vertical (upheaval) buckling
mode. Hence, unless horizontal displacements are restrained
(like for buried pipelines) or a prevailing vertical imperfection
is present, pipelines tend to buckle laterally. It has even been
argued to use lateral buckling as a design tool [36-37] to relieve
and control axial compression in the pipeline.
Fig.14. SAGE Profle upheaval buckling simulation.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
with w
p
and w
c
the submerged weight of the pipe and the
backfill cover respectively.
The download wd can then be converted to the rock dump
volume V
d
shown on Fig.16:
!
!
= ! !
!
+!
!
+
!
!
+!
!
!
tan!

(38)
with a the crest width, the slope angle and Z
d
the required
rock dump cover depth above the pipe. Note that this approach
remains valid for a pipeline sitting on the seabed, with:
! = 0 = !
!
(39)
Numerical tools can be used to evaluate the influence of
seabed modifications, such as the installation of sleepers to
promote buckling, rectifying a rough seabed, or finite-element
simulation of trenching operations. The post-trench soil
behaviour can be reflected by intelligent backfill soil springs
as shown in Fig.17, which account for the combined effects
of pipe mobilization, cover download, and backfill shear
resistance. Such tools allow optimizing rock dumping and
controlling the costs associated with seabed interventions.
Pipeline walking simulation
Observations and analysis [41] have shown that short subsea
flowlines operating at high temperatures can exhibit pipeline
walking [42] and axial creeping [43]. A recent, comprehensive
overview on pipeline walking is presented in [44]. Here,
SAGE Profile is used to demonstrate the importance of pipe
soil interaction (and in particular seabed friction) on the
likelihood of pipeline walking.
As explained in [44], pipeline walking can cause cumulative
axial displacement of an entire pipeline, which can induce
damage at termination units, expansion spools and riser tie-
ins. The rate of walking depends not only on the temperature
profiles, but also on the magnitude of axial resistance, the
mobilization distance and the seabed topography.
The main driving mechanisms for pipeline walking are:
tension, associated with a steel catenary riser
global seabed slope along the pipeline length
thermal transients during start-up and shut-down
Although the origin may be different, the walking mechanism
for each of these three cases is governed by the effective axial
force profile of the pipeline. For a fully restrained, closed-
ended pipeline, the effective axial force F
e
is the sum of the
forces due to axial elongation, internal and external pressure
(including end effects), and the temperature gradient T:
!
!
=
! !
!
!
!
!" + 1 2! !
!
!
!
!
!
!
!
!" !!
(40)
or the Pederson formula [39]:
!
!
= !
!
!
!
! 1 +0.1
!
!
!
+!
!
!
!
1 +
!
!
2!
!

(34)
where Z is the cover depth (counted from the top of the pipe)
and is an uplift coefficient, accounting for shear strength.
The subtle difference between the soil failure surfaces assumed
by Schaminee and Pederson is explained in Fig.15.
In addition to predicting the susceptibility to upheaval
buckling, finite-element software can be used as a design
tool to optimize the required volume of rock dump to
avoid buckling. The design envelope approach described by
Richards [40] gives an estimation of the required download:
! =
!
!
!
!
! +
!
!
!

(35)
where a and b are constants, h is the amplitude of the
imperfection, l the corresponding imperfection wavelength,
and:
! = !
!
!
!"

(36)
For a buried pipe, the rock dump should then exert a
download equal to:
!
!
= !!
!
!
!

(37)
Fig.15. Backfll according to Schaminee and Pederson.
Fig.16. Required volume of rock dump.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Generally, the slope of the force profile is defined by the
axial friction f = w
p
. On an inclined seabed, the pipe weight
promotes expansion in downhill direction, but counteracts
the uphill expansion. This is similar to modifying the
friction coefficient:
!
!
= !
!
! !"# ! !"# !
(42)
which causes an asymmetric force profile envelope [44]. This
situation gives rise to a rigid body displacement: the pipe
starts to walk down-hill. Figure 19 clearly demonstrates that
both pipeline ends move the same amount at the end of
each cycle. The walking rate can be approximated by [44]:
!
=
! +!
!
! !"# ! !
!
! ! !"# ! ! !"# !
! ! !

(43)
When the friction factor (or the pipeline length) is increased,
the fully constrained force is still sufficient to overcome
friction on first load, but not enough to mobilize the pipe
during cool-down, as schematically shown in Fig.20. Indeed,
the axial friction during cool-down is enough for a certain
section of the pipeline to reach fully constrained conditions.
with the coefficient of thermal expansion for steel.
During cyclic loading of the pipeline, the change in fully
constrained force:
! = ! 1 2! !
!
!" !!
(41)
will dictate the pipeline walking response [44-45]. In this
paper, SAGE Profile is used to evaluate the influence of
seabed friction on the walking rate of a pipeline on an
inclined seabed. A straight pipeline of 5000 m length
was modelled on an inclined seabed with a slope = 15
and subjected a temperature gradient T = 20. The soft
clay, shown in Fig.9, was used as a vertical soil spring. The
horizontal soil behaviour was modelled by uncoupled axial
and lateral soil springs as shown in Fig.10, i.e. allowing for
perfect plasticity and with a memory component; the latter
capability is fundamental to accurately model walking, and
only the seabed friction is varied.
Figure 18 shows the general force profiles in fully heated
and cool-down conditions for a short pipeline, i.e. the
force envelope does not exceed the fully constrained force
F. In this situation, the pipeline is fully mobilized and can
freely expand and contract around a virtual anchor point.
Fig.17. Smart backfll soil springs for buried pipelines.
Fig.19. Pipeline walking over fve subsequent cycles.
Fig.18. Force profles for fully mobilized short pipeline.
Fig.20. Force profles for cyclically constrained pipeline.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
the pipeline becomes fully constrained. As shown in Fig.22,
the pipeline is anchored over a certain length during both
cool-down and heat-up. Since the anchored lengths during
both phases overlap, the pipeline cannot walk. This is
clearly demonstrated by the vanishing axial displacements,
shown in Fig.23.
Thanks to the plastic soil springs with memory component
(Fig.10), modelling pipeline walking is feasible with SAGE
Profile. The results for pipeline walking on an inclined
seabed, presented here, are in excellent agreement with
the theory [44, 45].
Conclusions
In this paper, recent developments in numerical modelling
and analysis of offshore pipelines were reviewed and
discussed. Finite-element techniques to assist in pipeline
design were introduced, and the SAGE Profile software
suite for offshore pipeline analysis was used to evaluate free-
spanning pipelines, simulate lateral and upheaval buckling,
and address pipeline walking. The main conclusions from
these case studies are:
Simulation of the pipelaying process is one of the
most challenging tasks. An incremental solution
was presented, where new elements are fed-in from
an anchor point close to the seabed. This elegant
approach allows simulating the actual installation
process from a laybarge in an efficient fashion, and
enables a quick and straightforward assessment of
the on-bottom roughness.
Finite-element simulations of pipelaying can
contribute to route optimization, and the assessment
of free-spanning pipelines can save a significant
amount of time and money associated with seabed
rectification.
The key to a successful simulation of offshore
pipeline installation and operation, is a profound
understanding of pipe-soil interaction. The pipeline
steel can be modelled by a Ramberg-Osgood material
model, whereas the soil response is captured by a
combination of vertical, axial, and lateral springs.
The correct calibration of the pipe-soil interaction
parameters is of paramount importance to reach
reliable solutions.
Transient dynamic solvers are based on an explicit
integration algorithm. The central differences
scheme is conditionally stable, provided the time
increment is sufficiently small. By introducing the
concept of damping, the explicit (dynamic) approach
can also be used to obtain quasi-static equilibrium.
SAGE Profile can predict the occurrence of lateral and
upheaval buckling. In addition to these predictive
Over this length, the pipeline remains anchored and hence
cannot move during cool-down. Moreover, since the virtual
anchor point of the heat-up stage is located within the
anchored length of the cool-down cycle, the pipeline becomes
anchored at that location, which prevents walking. The axial
displacement during heat-up and subsequent cool-down,
predicted by the software, is shown in Fig.21.
When increasing the friction until:
!
!
>
2 !
!
!
!

(44)
Fig.21. Axial displacement for partially constrained pipe.
Fig.22. Force profles for fully constrained pipelines.
Fig.23. Axial displacement for fully constrained pipe.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
14. S.Gellin, 1980. Plastic buckling of long cylindrical
shells under pure bending. Int.J. Solids and Structures,
16, pp397-407.
15. H.Soren and Y.Bai, 1999. Bending moment capacity
of pipes. Proc. International Conference on Offshore
Mechanics and Arctic Engineering, OMAE.
16. P.E.Winter, J.W.B.Stark, and J.Witteveen, 1985. Collapse
behaviour of submarine pipelines. Shell structures,
Chapter 7, pp221-226.
17. Det Norske Veritas, 1992. Foundations: Class Notes
DNV CN30.4.
18. R.Verley and K.M.Lund, 1995. A soil resistance model
for pipelines placed on clay soils. Proc. International
Conference on Offshore Mechanics and Arctic
Engineering, pp.225-232.
19. J.C.Ballard, H.Falepin, and J.F.Wintgens, 2009. Towards
more advanced pipe soil interaction models in finite
element pipeline analysis. Proc. Annual Conf. of the
Society for Underwater Technology, Perth, Australia.
20. J.Zhang, D.P.Stewart, and M.R.Randolph, 1999. An
elasto-plastic model for pipe soil interaction of unburied
pipelines in calcareous sands. Proc. International
Symposium on Offshore and Polar Engineering ISOPE,
pp185-192.
21. Idem, 2002. Modelling of shallowly embedded
offshore pipelines in calcareous sand. J.Geotechnical and
Geo-Environmental Engineering, 128, 5, pp363-371.
22. R.Denis and C.De Brier, 2010. Deep water tool for soil
pipe interaction measurement: recent development
and system improvement. Proc. Offshore Technology
Conference, OTC-20630.
23. R.A.Jewell and J.C.Ballard, 2011. Axial pipe soil
interaction: a suggested framework. Proc. Offshore
Technology Conference, OTC-22010.
24. D.J.White, S.A.Ganesan, M.D.Botton, D.A.S.Bruton,
J.C.Ballard, and T.Langford, 2011. Observations on axial
pipe soil interaction from testing on soft natural clays.
Ibid.
25. J.R.Morison, M.P.OBrien, J.W.Johnson, and
S.A.Schaaf, 1950. The forces exerted by surface waves
on piles. J.Petroleum Technology, 189, pp149-154.
26. Det Norske Veritas, 2010. Recommended Practice DNV-
RP-C205, Environmental conditions and environmental
loads.
27. F.Van den Abeele and J.Vande Voorde, 2006. Stability
of offshore pipelines in close proximity to the seabed.
Proc. 6th Pipeline Technology Conference, Hannover,
Germany.
28. R.M.Isherwood, 1987. A revised parameterisation of
the JONSWAP spectrum. Applied Ocean Research, 9, 1,
pp47-50.
29. Pierson and Moskowitz, 1964. A proposed spectral
form for fully developed wind seas based on similarity
theory of S.A. Kitaigorodskii. J.Geophysical Research, 69,
pp5181-5190
30. T.Sarpkaya and M.Isaacson, 1981. Mechanics of wave
forces on offshore structures. Van Nostrand Reinhold,
New York.
capabilities, the software offers tools to prevent or
mitigate the problems associated with buckling. Smart
backfill soil springs can account for the combined
effects of pipe mobilization, cover download, and
backfill shear resistance, and calculation of the
required download to prevent upheaval buckling
contributes to rock-dump optimization.
Pipeline walking due to seabed slope has been
modelled, and the results show excellent agreement
with the theory: migration only occurs when the
effective axial forces along the entire pipeline remain
below the fully restrained conditions during both
heat-up and cool-down. Again, the influence of
seabed friction on the simulation results was stressed.
Finite-element tools, such as that described in this
paper, provide added value as a design aid and
a decision tool throughout the entire life of an
offshore pipeline, covering (among other aspects)
preliminary design, route optimization, pipelaying,
span assessment, and on-bottom stress analysis.
References
1. J.F.Witgens, H.Falepin, and L.Stephan, 2006. New
generation pipeline analysis software. Offshore Pipeline
Technology Conf.
2. Fugro GeoConsulting, 2010. SAGE Profile 3D pipeline
analysis software. Technical Brochure.
3. http://www.sage-profile.com (accessed 2012).
4. A.C.Palmer and R.A.King, 2004. Subsea pipeline
engineering. PennWell Books.
5. R.Denis and I.Nash, 2012. Desk-top study: the un-
exploited resource. World Pipelines.
6. Y.Bai and Q.Bai, 2005. Pipelines and risers, Chapter
34: Installation design, pp 597-636.
7. H.S.Choi, 2001. Free spanning analysis of offshore
pipelines. Ocean Engineering, 28, pp1325-1338.
8. O.Fyrileiv, 2010. Effect of internal pressure on free
spanning pipelines. 8th Int. Pipeline Conference,
IPC2010-3162.
9. Det Norske Veritas, 2006. Recommended Practice
DNV-RP-F105, Free spanning pipelines.
10. R.E.Hobbs, 1986. Influence of structural boundary
conditions on pipeline free span dynamics. International
Engineering, OMAE.
11. W.Ramberg and W.R.Osgood, 1943. Description of stress
strain curves by three parameters. National Advisory
Committee for Aeronautics, Technical note no. 902.
12. W.F.Chen and D.J.Han, 1985. Tubular members in
offshore structures, Chapter 2: Moment-curvature
relations, pp 21-30.
13. C.E.Murphey and C.G.Langer, 1985. Ultimate pipe
strength under bending, collapse and fatigue. Proc. 4th
International Conference on Offshore Mechanics and
Arctic Engineering, OMAE, pp467-477.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
39. P.T.Pederson and J.Michelsen, 1988. Large deflection
upheaval buckling of marine pipelines. Proc. Symposium
on the Behaviour of Offshore Structures, pp965-980.
40. D.M.Richards, 1990. The effect of imperfection shape
on upheaval buckling behaviour. In: Advances on subsea
pipeline engineering and technology, Ed. Ellinas, pp51-66.
41. I.Konuk, 1998. Expansion of pipelines under cyclic
operational conditions. Proc. International Conference
on Offshore Mechanics and Arctic Engineering,
OMAE.
42. D.Perinet and I.Frazer, 2005. Movements of deep
water flowlines as a result of pressure and temperature
loads and the influence of steel catenary risers. Deep
Offshore Technology Conference.
43. K.Tornes, J.Jury, B.A.Ose, and P.Thomson, 2000. Axial
creeping of high temperatures flowlines caused by soil
racheting. Proc. International Conference on Offshore
Mechanics and Arctic Engineering, OMAE.
44. M.Carr, F.Sinclair, and D.Bruton, 2006. Pipeline
walking: understanding the field layout challenges
and analytical solutions developed for the SAFEBUCK
JIP. Proc. Of fshore Technology Conference,
OTC-17945.
45. C.S.Gaillard and K.Williams, 2005. Pipeline walking.
Proc. Offshore Pipeline Technology Conference.
31. K.J.Bathe, 1996. Finite element procedures. Prentice-
Hall Inc.
32. S.P.Timoshenko and J.M.Gere, 1961. Theory of elastic
stability, 2nd Edn, McGraw-Hill.
33. R.E.Hobbs, 1981. Pipeline buckling caused by axial
loads. J.Constructional Steel Research, 1, 2.
34. R.E.Hobbs and F.Liang, 1989. Thermal buckling
of pipelines close to restraints. Proc. International
Engineering, OMAE.
35. M.Carr, D.Bruton, and D.Leslie, 2003. Lateral buckling
and pipeline walking: a challenge for hot pipelines. Proc.
Offshore Pipeline Technology Conference OPT.
36. D.Bruton, M.Carr, M.Crawford, and E.Poiate, 2005. Safe
design of hot on-bottom pipelines with lateral buckling
using the guideline developed by the SAFEBUCK JIP.
Proc. Deep Offshore Technology Conference.
37. D.Kaye, D.LeMarchand, P.Blondin, and M.Carr, 1995.
Lateral buckling design of the Dunbar-North Alwyn
double wall insulated pipeline. Offshore Pipeline
Technology Conference.
38. P.E.L.Schaminee, N.F.Zorn, and G.J.M.Schotman, 1990.
Soil response for pipeline upheaval buckling analyses:
full scale laboratory tests and modelling. Proc. Offshore
Technology Conference, OTC-6486, pp563-572.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
*Corresponding authors contact details:
tel: +31 6 2420 2452
email: rob.bos@europims.nl
G
ASUNIE OPERATES the frst cross-border network in Europe, a network that ranks amongst the largest
high pressure gas pipeline grids in Europe consisting of over 15,000km of pipeline in the Netherlands
and northern Germany, dozens of installations, and 1,300 gas receiving stations. Due to the reliability and
strategic location of the grid in relation to expanding international gas fows, the Gasunie network forms
the core of what is called the northwest European gas roundabout. In 2010, the annual gas throughput
totalled approximately 125 billion cubic metres. Safety, reliability, and sustainability have always been the
key elements in Gasunies operational practice and, as a consequence, much effort has been made to keep
the condition of the infrastructure, which originates from 1960, up to current standards.
GTS, a wholly-owned subsidiary of Gasunie, is the designated national transmission system operator in
the Netherlands. One of GTSs key missions is the non-discriminatory provision of safe and reliable gas
transmission services. Such services should enhance security of supply and the proper functioning of the gas
market. As a consequence, GTS has a strong drive for continuous improvement on the pipeline-integrity-
management system (PIMS) that is already in place.
Gasunie and GTS are utilizing high-integrity standards for their assets. This is not only refected in the design
standards that have been applied, but is also expressed in daily operations and maintenance practice of the
company combined with a strong drive for continuous improvement.
Managing the integrity of pipelines is such a comprehensive matter that it requires the involvement of
several disciplines. Therefore, an integral approach is very important. This paper provides an overview of
the relevant aspects for the integrity management of pipelines and shows how these aspects are related.
by Rob Bos*
1
, Suzanne Mooij
1
, Leen Pronk
2
, and Wessel Bergsma
2
Risk control at lower cost
PIMS practice in GTS
GTS has extended the operational PIMS practice in 2011
with a new approach in which transparency and a continuous
improvement are key issues. As a result, the process of
continuously improving PIMS, which used to be focussed on
the regular operational, maintenance, and safety elements,
has recently been enhanced with a focus on the following
aspects of PIMS:
Bow-tie risk analyses. Risks are evaluated for the four
sub-areas: transport, technical integrity, environment,
and organization.
Performance indicators. Performance indicators are
derived from the Bow-tie risk-analysis approach in
order to monitor integrity performance.
PIMS-plan. Periodically, (annually, in principle) a
PIMS plan is formulated which contains the most
important achievements of the past year and which
lists the proposed improvements.
A major part of the improvement of the PIMS in GTS
involves the introduction of the Bow-tie risk-analysis method.
This method measures the actual integrity management
of the pipelines or facilities concerned, and determines
opportunities for improvement in a systematic manner.
The approach to risk management carefully considers both
results in terms of integrity and cost. In this paper, the
Bow-tie risk-analysis method is described in more detail.
Legal developments
With this new method of integrity and risk management,
GTS complies with the new legislation in the Netherlands
on pipeline-integrity management. This new law, called the
BEVB (on external safety of pipelines) has strengthened GTS
in its objectives to become the best-in-class. This has resulted
in the enhanced risk-management methodology which is the
main subject of this paper, not limited to external safety, but
extended to a number of areas of risk as will be described
below. In fact, in this study, external safety has proven to
be no issue at all. The paper describes the review of current
condition of the 17,000 sub-surface valves that are part of
the pipeline system and the way the Bow-tie technique has
1 PIMS International, Haren, Netherlands
2 Gasunie, Groningen, Netherlands
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
is required. This has been begun, and the maintenance /
replacement programme for these sub-surface valves in the
40-bar grid is being intensified.
Aims
GTS has taken the decision to execute this sub-surface
valve assessment using a risk-based approach, and the
recently developed risk-assessment methodology (based on
supported GTS in the decision-making process regarding
investment decisions which are based on both technical
and economical parameters.
Scope
The main sections of the GTS grid operate at 40 bar and
contain around 17,000 sub-surface valves from a wide range
of manufacturers which include, in alphabetic order:
Argus
Audco / Rockwell / Nordstrom
Cameron / TK / Cort
Christensens
Elita / Porringern
Grove
M&J / LPG / NDSM
Mokveld
Newman Milliken / Hattersley
Pibiviesse
RMA
Saut du Tarn
Thevignot
WKM
The number of operational valves from these manufacturers
varies between just a few to several thousand. In order not to
affect the manufacturers reputation, results of the analyses
of these reports have been made anonymous, as shown in
Table 1 in which the valves are categorized by type.
A review of the construction dates of these sub-surface valves
is given in the Fig.1. As can be seen a considerable number
of these valves is have been in operation for around 40 years.
Recent reports on corrosion and other time-related ageing
effects have strengthened GTS in its opinion that a review
Manufacturer Valve type Number in
operation
A Axial on-off 457
B Ball 108
C Ball 1558
D Ball 402
E Ball 7
F Ball 397
G Ball 13
H Gate 3572
I Gate 865
J Knife gate 790
K Plug 4900
L Plug 957
M Plug 1355
N Plug 227
Fig.1. Construction dates of GTS sub-surface valves.
Table 1.Valve types according to manufacturer.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
commissioning
operations
maintenance
third-party interference prevention
mothballing / removal
The sub-division in these life-cycle barriers enables
the comparison of the efficiency of mitigation
measures taken for different life cycles (for example,
design measures vs maintenance measures) to find
the most efficient package of mitigation measures
for the whole lifetime of the assets. In general,
most asset managers have to deal with existing
infrastructure, in which case the inventory of
mitigation measures for the life cycles of design,
construction, and commissioning do not appear
relevant; however, the inventory of escalating factors
and accompanied mitigation measures for these life
cycles clearly show:
whi ch mi ssi ng measures from desi gn,
construction, and commissioning need to be
compensated for;
which adaptations in design, construction,
and commissioning should be applied in new
constructions to improve the lifetime efficiency.
incorporation of effectiveness and costs of any of
the mitigating measures;
incorporation of the Deming cycle: transparent
plan-do-check-act (PDCA) by the inventory and
implementation of new mitigation measures, based
on efficiency considerations (improvements);
incorporation of life-cycle cost optimization (LCC)
by the inventory and ranking of all (existing and
new) mitigation measures, based on efficiency
considerations for all life cycles;
inventory of the actual risk level by means of a
risk matrix and the implementation of risk-
mitigation measures when the matrix criteria are
not met.
This results in the basic configuration of the Bow-tie
shown in Fig.2.
Bow-tie risk-analysis method
The Bow-tie method developed by GTS and PI is based on
the inventory of costs and effectiveness of all mitigation
measures (existing and improvements) applied by the
pipeline operator to minimize the failure frequency and
the consequences of unwanted events.
(Note that the availability of high-quality data in the pipeline asset
register is of great help for the (semi-)quantification of threats,
consequences, and measures for different levels of the pipeline
system; however, if such data are not available, expert opinion
can do the job as well. In that case, the assessment should be
considered to be a semi-quantitative risk analysis.)
the advanced Bow-tie technology) was selected. The major
objectives of the project are:
to investigate the risk level of the sub-surface valves
(the risk-matrix parameters) on:
safety
security of supply
costs
environment
reputation
to investigate the average lifetime of the sub-surface
valves in the 40-bar grid
to find the main cause of deterioration of the valves,
for example:
ageing (corrosion, wear, third-party interference,
etc.)
geographic conditions (soil/ground water
composition, medium composition, etc.)
manufacturer (type, material, fittings, etc.)
function (isolation, by-pass)
to reconsider current maintenance practice and
replacement policy if it is concluded that certain valves:
do not meet the above-mentioned criteria, i.e.
have reached end-of-life status
require investigation of the possibilities for repair,
modification, and / or replacement if they have
met their end-of-life status
are far from their end-of-life status
to define maintenance / operational measures to
be taken during the transition period for those
valves which need replacement, if applicable.
Methods
General
The methodology applied, that has been recently developed
by GTS and PIMS International BV in close co-operation
with Gaz de France Suez and Total EP, is based on the
conventional Bow-tie methodology but offers a considerable
amount of additional management information because
it has been extended with:
four sub-sectors of risk assessment covering the
risk-management aspects of:
environment
organization
transport
technical integrity
a sub-division of the main barriers to the threats and
consequences in the seven life-cycle barriers, both
for preventive and suppressive mitigation measures:
design
construction
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
GTS risk matrix on the following aspects: financial
damage; environment; reputation; security of supply;
and safety. Aspects are score based on the expert
opinion / knowledge of the expert team.
(Note: The ranking of the current risk level is based on
the assessment of the worst-case scenario: what happened
or might happen in the near future and at what interval
when no additional mitigations are applied.
The appointment of two independent teams, in particular
for high-consequence decisions, is highly advised to
guarantee maximum objectivity of their conclusions and
recommendations; this has not been applied in this study.
If applied, two teams have to be established, one each to
undertake:
the inventory at the current risk level
the inventory after implementation of the new mitigation
measures.)
4. Development of one fully developed Bow-tie for each
of the subsectors; for each Bow-tie:
a. The identification and quantification (effectiveness
and costs) of threats that could result in an unwanted
event (preventive measures, Fig.2, left part).
This Bow- tie risk analyses method involves the
following steps:
1. Assignment of workshop members. This team
should at least have the following expertise: valves
(in this particular case), third-party interference,
cathodic protection, coating, corrosion phenomena,
inspection techniques, maintenance engineering,
operations, engineering, and financials, and should
be managed by a facilitator who is experienced in
the field of integrity management.
2. Inventory of unwanted events for each of the
following subsectors: GTS focuses on the following
unwanted events (which is a legal requirement):
Subsector Unwanted event
Environment Loss of containment
Organization Backlog in carrying out
maintenance
Transport
1
Failure of contractual
obligations
Technical integrity Failure of function
3. Inventory of the consequences for each of the
unwanted events including a check on the current
risk level of each of these consequences using the
1.
The standard unwanted event is Consequential loss of erroneous medium composition. Because Gasunie deals only with inert gases, it has been decided also to introduce
the unwanted event Failure of contractual obligations.
Fig.2. The GTS Bow-tie diagram.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
vi. Mothballing / cleaning / removal: removal of
superfluous objects, registration of condition
of removed objects.
e. The definition of the seven suppressive life-cycle
barriers that are installed to prevent and / or minimize
the consequences of the unwanted event, to be
defined for each of the consequences (Fig.2, right).
f. The identification of escalating factors for each of
the suppressive life-cycle barriers that could reduce
the effectiveness of the mitigation measures, to be
defined for each of the consequences.
g. The identification of existing and potential measures
for each of the escalating factors that could prevent
and/or limit the consequence of the unwanted event
(such as loss of containment) defined per life cycle.
Examples of these measures are:
i. Design: gas detection and/or ignition source
availability.
ii. Construction: prevention of spark-inducing
materials in ground bed.
iii. Commissioning: quality check of ignition-
initiating sources.
iv. Operations: emergency plan.
v. Maintenance: application of non-igniting tools,
gas-detection tool.
vi. Mothballing/removal: procedures to prevent
nearby damage whilst removing superfluous
objects.
b. The definition of the seven preventive life-cycle
barriers that are installed to prevent and / or
minimize occurrence of the unwanted event.
c. The identification of the escalating factors for each
of the preventive life-cycle barriers that could reduce
the effectiveness of each of the barriers.
d. The identification of existing and potential
measures for each of the escalating factors that
could prevent and / or limit the effect of the factor
in initiating the unwanted event, defined per life
cycle. Examples of these measures are:
i. Design: code applied, steel grade, coating type,
wall thickness, cathodic protection-design.
ii. Construction: supplier qualification,
supervision to meet design and construction
specs.
iii. Commissioning: quality check of: pressure
control system, welds, fittings and cathodic
protection, test criteria / conditions /
medium, hand-over of asset register operating
instructions and maintenance requirements.
iv. Operations: education on operations manual
and ESD-system, incident reports.
v. Maintenance: maintenance concept, support on
maintenance planning and other organizational
issues, material supply / alternatives, education
and training, requirement for maintenance /
failure reports.
Fig.3. Detail of the Bow-tie diagram for sub-surface valves.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Bow-tie detailed view
Figure 3 shows a detail of the Bow-tie diagram for sub-surface
valves (safety aspects), and represents the preventive part of
the Bow-tie (the left part) including some of the mitigation
measures taken to prevent the unwanted event occurring (the
blue box). In this example, the mitigation measures with a
green tab are already implemented, the mitigating measures
with the yellow tab are not implemented yet but have been
proposed by the expert team during the workshop as a new
mitigation measure (i.e., an improvement).
LCC calculations
The Bow-tie includes both the effectiveness of the mitigation
measures and the corresponding costs, enabling the user to
evaluate the economics of current and new measures to either
gain the same effectiveness of integrity management at lower
costs or to achieve a higher level of threat control at the same
costs, as is demonstrated in Fig.3. The red line represents
current practice: which barrier value (x-axis) can be achieved
at what cost (y-axis). The blue line represents the potential of
new proposed mitigation measures. Both lines are constructed
on the basis of the costs and the effectiveness of the mitigation
measures; however, these are not shown on the x-axis due to
the fact that there are too many (around 130).
Figure 3 very clearly shows the opportunities for LCC-
optimization, which are:
same barrier of mitigation measures at lower costs =
cost saving
same costs of mitigation measures at higher barrier =
increase of effectiveness
5. Calculation and ranking of effectiveness and
efficiency of all of the existing and new proposed (i.e.
improvements) mitigation measures.
6. Calculation of the most efficient set of mitigation
measures based on a mix of existing and new
mitigation measures.
7. Calculation of life-cycle efficiency and life-cycle
optimization.
8. Calculation of the risk-reducing effect of the new
barriers in the risk matrix, both on failure frequency
and consequence.
9. Calculation of performance indicators (actual
and potential).
Workshops
The Bow-tie risk assessments are performed in a number
of workshops, in general taking three days to investigate
each object-type. These workshops are attended by
around eight experts in the relevant fields, and involve
the following steps:
Development of a full developed Bow-tie for each
of the subsectors (four in total).
Data processing and analyses.
Expert judgement of the results of the analyses.
Trend analyses based on maintenance reports
(the SAP module Plant Maintenance (PM) is
applied in GTS for maintenance reporting) and
incident reports.
Definition of conclusions and recommendations
including a review of existing maintenance and
replacement policies.
Report to management.
Fig.4. Opportunities for LCC-optimization.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
analysis are reported to support management decisions
on maintenance optimization.
This means that when a loss of containment is experienced,
the cause should be reported in SAP PM. This can be
done by selection from a table with a choice of threats
reported in the Bow-tie analyses. For example: (1) external
interference (excavation), (2) external interference (ram
piling), (3) system gains higher pressure than design
pressure, etc. The consequences need to be selected as
well, for which a choice should be made from the short list
of (consequences) shown on the right side of the Bow-tie.
For example: (1) gas emission, (2) environmental damage,
(3) fire, (4) explosion, etc.
In a similar way, the causes and consequences of other
undesirable events for the other subsectors are reported
in SAP.
Management information
Success factors (SF)
Success factors are used to represent the result of the
companys integrity-management practice. They are
visualized and quantified by the number of incidences of
the unwanted events as applied in the development of the
Bow-tie analyses.
Performance indicators (PI)
Performance indicators represent the way the SFs are
managed by the company; for example, the number of the
loss-of-containment incidences can be influenced by the
(Note: The above-mentioned analysis include the effect of
mitigation measures for the full lifetime of the assets. For existing
infrastructure, revision of the design, construction, and commission
mitigation measures cannot be applied. For this reason, the analysis
offers the functionality to skip these life cycles in the inventory
of improvements, and compensation of the missing barriers has
to be found in the other life-cycle barriers.)
Priority-setting of mitigation measures
Priority-setting of mitigation measures is performed by
analysing the results of the Bow-tie risk analysis. Figure 5
shows the relationship between improvement measures
(a mix of all life-cycle-related measures), failure frequency,
and cost. Since measures are ranked according to efficiency
(= max. barrier value/), the choice for a new failure
frequency directly shows the most efficient set of measures.
(Note: It is assumed that all mitigating measures are independent.
Figure 5 shows the cumulative effect of only the new mitigation
measures (improvements) on the reduction of the failure frequency
(red line). The black line represents the cumulative costs (at five-
year intervals). As a consequence, the effect of for example
measure 5 is based on the impact of measures 1 to 5. It is clearly
demonstrated that spending only 20% of the cost will result in an
approximate 80% reduction of the failure frequency.)
In order to fully support decision-making with regard to
the proposed measures, the threats and consequences that
were identified during the Bow-tie sessions are incorporated
in the standard SAP PM maintenance reports. This means
that as well as the (usual) identification data of the object,
the frequency, cause, and consequence of the unwanted
event as they were analysed during the Bow-tie risk
Fig.5. The cumulative effect of the new mitigating measures (improvements) on the reduction of the failure frequency.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
PIs are derived from the Bow-ties and are quantified by the
calculation of the combined effect of all mitigation measures
for the neutralization of all threats and all consequences.
The following PIs are derived from the Bow-tie method:
PI
Threat control

PI
Preventive life-cycle threat control
PI
Consequence control
PI
Suppressive life-cycle threat control
implementation and/or avoidance of certain mitigation
measures. The selection of the mitigation measures that
have to be implemented (or not) to manage the SF properly
is based on the policy of the company on that particular SF:
identifying what is the accepted number of unwanted events.
The effectiveness of the implemented measures represents
the integrity-management quality of the company, and is
determined with the Bow-tie methodology. Examples of PIs
are given in the following paragraphs.
Fig.6. GTS conventional performance indicators (PIs).
Fig.7. The score on GTS enhanced preventive life cycle PIs.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
infrastructure). The method supplies a direct focus
on the effectiveness of the life-cycle-related mitigation
measures and, as a consequence, enables the operator
to focus on the actual (economic) performance
of the companys life-cycle management practice.
The performance of the (whole of the) company is
measured by the PIPreventive life-cycle threat control
.This indicator represents the reduction in threat
level for each individual life-cycle phase that has
been achieved by the preventive measures that are in
place and the maximum reduction level that can be
achieved by the implementation of the new mitigating
preventive measures (the preventive improvements).
The analyses of all Bow-tie parameters results in
Fig.7, which represents the score on GTS enhanced
preventive life-cycle performance indicators.
Figure 7 indicates that a reduction of the frequency
of an unwanted event can be achieved by the
implementation of new mitigation measures for the
following life cycles:
Life cycle Threat-reduction
potential
Design from 20% to 5%
Construction from 16% to 9%
Maintenance from 17% to 8%
Third-party interference from 13% to 10%
As has been mentioned above, design and
construction barriers are not an option for an existing
infrastructure. This means that the majority of threat
reduction has to be achieved by the implementation
of new maintenance measures (reduction from 17%
to 8%) and reduction of the third-party interference
(from 13% to 10%). The selection of the most
efficient new mitigation measures should be made
on the basis of the available Bow-tie diagram and the
efficiency plots.
Shift in the risk matrix resulting from new mitigation
measures
The effect of the proposed mitigation measures can be
visualized in the matrix shown in Fig.8 which includes the
risk level of the current situation as it has been ranked by
the experts in the workshop: the small black squares. Next to
that, the effect of the new mitigation measures that have been
proposed by the experts is (after processing by the software)
are presented as the blue squares. As can be seen, the blue
square has a shift to the left (reduction of failure frequency
affected by the new preventive mitigation measures) and a
shift to the top (reduction of consequence resulting from the
new suppressive mitigation measures). This is achieved by the
introduction of an x-axis scale which is a log scale related to
the frequency of the unwanted event, and a y-axis log scale
of the consequence which, in this case, is related to the cost.
(Note: In-depth calculation of the PI-level of an object requires
connection between the Bow-tie diagram and the asset register of
that object, in order to distinguish between different objects with
the same function. Difference in object parameters can result in
different effectiveness of the mitigation measures that are in place.
As an example: changes in depth of cover of a pipeline, a mitigation
measure against third-party interference, will result in changes in
effectiveness of this mitigation measure along the pipeline route.)
The value of each PI is calculated using PIMS International
software, similar to the method that is used to calculate the
combined effectiveness of mitigation measures as described
earlier. To illustrate this, an example of the results of
the preventive PI calculation has been given to illustrate
the usefulness of these PIs in the characterization of the
management control. Options for improvement can be
found easily, and the prospects for each of the life cycles is
presented in Figs 6 and 7. The major difference between
the PI characteristics of threat control and life-cycle control
have been emphasized to provide a clear understanding of
the strength of this new method of risk assessment.
1. The conventional way of PI definition is to focus on
the management of threats. This ends, in general,
with the definition of the PIThreat control which is
the reduction in threat level for each individual threat
that has been achieved by the preventive measures
that are in place and the maximum reduction
level that can be achieved by the implementation
of the new mitigation preventive measures (the
preventive improvements). The analyses of all Bow-
tie parameters results in Fig.6, which represents the
values of GTS conventional performance indicators.
Figure 6 shows that the actual management of threats
has resulted in a substantial reduction of the basic
threat levels (in an estimate made by the expert team).
Next to that, it demonstrates that additional (new)
mitigation measures can be implemented to achieve
even better control. Whether these new measures
are going to be implemented or not depends on
the companys policy regarding its experienced
frequency of the unwanted event (in this case, loss
of containment). The figure shows that further
reduction is optional if this frequency (the SF) is
considered to be too high. The selection of new
mitigation measures can be made on the basis of the
available Bow-tie diagram and the efficiency plots
(see paragraph above on priority setting).
2. The enhanced method introduced by PIMS
International and GTS should be considered as a
very practical extension to the conventional method.
It enables the asset manager to be realistic, and to
skip the opportunities of threat reductions that are
out of scope (for example, a design review is not
an option for reducing the risk level of the existing
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Results
Ranking of the risk matrix
In the following discussion, all the results have been made
anonymous. 14 different valves were ranked in an exercise
carried out by an expert team, and this resulted into the
matrix shown in Fig.9 (a random example relating to
one of the 14 valve manufacturers). As can be seen, two
criticals (red scores) have been reported as a consequence
of disintegration resulting from the loss of containment
(in this case a relative small leak) unwanted event:
safety risk
financial damage
(Note: The current level of the safety aspect of disintegration (score
Vv) gives a yellow score (see the left blue circle), and no injuries
are reported up to that point. However, the expert team holds
the opinion that further deterioration of the valves may result
in serious injuries to the field engineers in the near future if no
additional mitigation is taken, and has decided to caution GTS
management that direct action is required in order to prevent this
happening. As a result, the team assigned the red score to the
safety aspect of disintegration (see the right blue circle).
Closer investigation of the cause of this red score has
resulted in the conclusion that this is the effect of corrosion
of the drain and seal accessories of the valves. In particular,
operation of the corroded gas taps might result in the near
future in disintegration of the tap with fitting material
rapidly dispersed. Whenever this happens, serious injury
As an example, consider she score of the safety-effect
2
of the
disintegration consequence (Z
v
) that has been ranked by the
expert team on 2Ihigh = red (see light blue circle). The ranking
is positioned in the right-bottom corner of the corresponding
area instead of the conventional method of ranking (indicating
the complete area), i.e. the right bottom corner represents the
score 2Ihigh.
The shift to the left which results from the mitigating
effect of the new preventive measures and the shift to the
top which results from the mitigating effect of the new
suppressive measures brings the score out of the red zone
into the orange zone.
As been demonstrated in this example, leaving the orange
zone is not an available option and, as a result, a decision has
to be made whether the orange++ zone is acceptable for the
company or not. If not, more fundamental actions have to be
taken to meet the company policy on asset integrity, such as
renovation, replacement, etc. If the orange zone is acceptable,
additional analyses have to be made to find the minimum
number of mitigations required to leave the red zone. Although
the cost of the mitigation measures have to be included, the
method delivers the most efficient set of mitigation measures
(preventive and suppressive) that is required to manage the
risk level in an appropriate and sophisticated way.
In the end, this enhanced Bow-tie method supplies the asset
manager with all the information required to support risk-
management practice in a structured way, and enables the
(legal) integrity requirements to be met in the most transparent
and efficient manner.
2.
Note that each risk area does have its own risk definition, and that risk areas cannot be compared to each other. Obviously, these risk areas also differ in priority.
Fig.8. The effect of the proposed mitigation measures.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
of the field engineer who is operating the tap can be
anticipated; according to GTS policy, the red level score
requires immediate action. On the basis of this, GTS has
decided to take immediate action:
A technical bulletin was published with the
instruction to stop the operation of drain and seal
taps, and if this is not possible, to protect the field
engineers when operating the taps.
A feasibility study to be executed on the potential
of risk reduction by additional mitigation measures.
Feasibility study
The Bow-tie method was selected to investigate the shift
of the risk scores in the matrix by additional preventive
and suppressive mitigation measures (improvements) as
proposed by the expert team, and Fig.10 shows the result of
this exercise. In Fig.10, the risk level of the current situation
as ranked by the experts in the workshop is shown
as the black squares. Next to that, the effect of the new
mitigation measures on the shift of the ranking has been
calculated on the basis of the additional (new) measures
that have been proposed by the experts, and processed
by the software. The calculated scores are presented in
the graph as the blue squares. The figure shows that a
reduction of both safety risk and financial damage risk is
achieved by the additional mitigation measures, although
unfortunately not enough: both scores are still in the red
area. This resulted in the conclusion that further action was
required to meet GTS risk standards, and the following
study was started:
1. The investigation of the main cause of the
deterioration of the valves: the following potential
causes were investigated (similar to a Pareto analysis):
ageing (including corrosion, wear, third-party
interference)
geographic conditions (including soil/ground
water composition, medium composition)
manufacturer (including type, material, fittings)
function (isolation, by-pass).
The manufacturer-type showed up as the major
driver in the mechanism of deterioration.
2. The inventory of valve types that meet GTS risk
standards and those that do not.
3. If required, replacement of the valves would be a time-
consuming operation, and it was therefore decided
that prioritization of the replacement programme
was required. The prioritization has been executed
on the basis of the score of the different makes on
all aspects of the risk matrix.
4. Investigation of the score of new valves on the basis of
the Bow-tie parameters to rank them on the potential
risk, both now and in the future.
Fig.9. Risk-matrix ranking for one of the 14 valve manufacturers.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
interference, and vandalism are the most important threats
at this moment (each of which is 5%-6%, which means
that all of them are controlled for about 80%). All of these
threats have the option to reduce the threat level by the
implementation of the additional mitigation measures that
have been proposed by the expert team during the workshop
and which are shown by the dark blue bars. The risk matrix
that has been discussed above, however, shows that these
additional mitigation measures will not have the effect of
moving the valve integrity out of the red zone.
Inventory of valve types that meet GTS risk
standards
A review of the score of all the valve concerned is presented
in Table 2, which clearly shows that most valves will not
meet the safety criteria in the near future. The main threats
to valve integrity are shown in Fig.11, in which the light
blue bars represent the actual threat level, based on the
application of the mitigating measures that are in place, the
Performance Indicators. Wear, external corrosion, external
Manufacturer Type Average year of construction Number Estimated future safety risk score
B Ball 1991 108
K Plug 1973 4900
C Ball 1979 1558
L plug 1992 957
J Knife-gate 1978 790
D Ball 1985 402
H Gate 1977 3572
A Axial on-off 1974 457
M Plug 1972 1355
E Ball 1996 7
F Ball 2007 397
G Ball 1976 13
N plug 1974 227
I Gate 1972 865
Fig.10 Feasibility study result.
Table 2. Review of the future safety risk score for all valve types.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Prioritization replacement programme and
experimental study
It was clear to GTS that the valves that had a red score on
the safety aspect need to be replaced. Because the lead time
of such a project is rather long, it was considered desirable
to find a method of sub-ranking of the valves that need
replacement, in other words to differentiate between the
bad and the ugly. For this reason, a pilot study was been
carried out to investigate the possibility of further ranking
by scoring the matrix parameters.
The following relative-ranking factors were introduced
to enable ranking of valves from the various different
manufacturers:
Colour scores: Ranking Aspects of matrix Ranking
Red 3 Safety 10
Orange 2 Costs 2
Yellow 1 Environment 5
Green 0 Reputation 3
Security of supply 8
The yellow score is acceptable under the GTS condition of
As Low As Reasonably Acceptable (ALARA). For this reason
the standard is calculated on the basis of the yellow score
in combination with the above-mentioned aspect ranking,
a technique of virtual ranking. The number of criticals
(consequences that have red scores) in the overall assessment
is six, i.e. the level of risk of the following six consequences
was concluded not to be acceptable:
disintegration (failure of drain and seal connections)
control on loss of containment during calamities
isolation fails during maintenance work
grid separation fails
failure of function causes unwanted downstream
effects
doesnt meet requirements on License to Operate.
The scores of these aspects of the matrix on these six
consequences were then been used to differentiate between
all the valves. This brings the virtual standard (i.e. the
maximum allowable risk level) to: 6 x (10 + 2 + 5 + 3 + 8)
= 168. All valves having a score above 168 do not meet this
virtual risk standard. The results of the analysis is presented
in Fig.12, where clearly the ball and gate valves do not meet
this virtual standard with one exception: ball valve F.
Pilot study: ranking new valves on the basis of design
parameters and theoretical mitigations
Because the use of ball valves as the main isolation valves
is considered by GTS to be essential for the execution of
pig runs, the company has decided to extend the Bow-tie
analysis to more recently designed ball valves to find out if
more recent design types have a better risk score than the
valves of older designs which are in operation.
This pilot study was carried out by the same group of experts.
Analyses were carried out to find correlations between the
Bow-tie parameters and the final score in the risk matrix. In
the end, the summarized value of all the preventive mitigation
measures proved to be the major factor in predicting the
risk score of the new valves with unknown risk-profiles.
Fig.11. The main threat to valve integrity.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
expected virtual risk level in the matrix of 108, which means
that the new valve will meet the virtual risk criteria now and
for up to 40 years of operation. The development of the Bow-
ties for other comparable valves (and their manufacturers)
is continuing and the values of R
p
will be used as selection
criteria for the valve (both manufacturer and configuration)
that will be selected for the replacement programme.
It should be noted that this methodology applied can be used
and is going to be used for all GTS infrastructure assets.
Summary and conclusions
A new method of integrity and risk control has been developed
with the following benefits:
full transparency in threat mitigations and consequence
control
complete integration of all relevant risk areas
perfect match between risk control and cost
containment.
A new and enhanced method of risk assessment has been
developed by GTS and PIMS International that enables the
pipeline operator to:
Meet governmental requirements on risk management
by controlling not only the safety risks but also the risk
of environmental damage, loss of reputation, security
of supply, and costs, in a highly transparent way.
Determine the performance of integrity management
with performance indicators that can easily be derived
from the Bow-tie diagram.
This is shown in Fig.13 where the relationship between the
summarized preventive barrier values of the Bow-tie (blue line)
and the virtual risk level (red line) can be seen. In summary:
high barrier value low risk level in matrix (and
vice versa)
The green line in Fig.13 shows the relationship between risk
level and the virtual standard (i.e. the maximum allowable
risk level):
R
p
= summarized barrier value > 1.68 gives an
acceptable virtual risk level
This results in the definition of R
p
as the predicted risk level
based on the summarized barrier value of all mitigation
measures from the Bow-tie analyses.
This factor R
p
has been used to compare the risk score of
unknown, new, valves as a selection criterion. An example
of the result of this way of risk ranking is given in the
following: the new design of valve C has been investigated by
the expert team and all mitigation measures were reassessed
on the basis of the valve parameters that were supplied by
the manufacturer. Next, the summarized preventive barrier
value was calculated with the result: R
p
(new design) = 1.78.
Conversion of this calculated R
p
value into the virtual risk
value can be done via the green dotted line in the graph of
Fig.14, which shows the increased value of the R
p
for the new
valve C compared to that of the old design that originates
from around 40 year ago. This higher value is the result of
the modifications that the supplier has made to the original
design and, as a consequence, the modifications result in an
Fig.12. Example of the virtual standard.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
21-23 OCTOBER 2013
CONFERENCE
21-23 OCTOBER 2013
EXHIBITION
W
Global
Webb
2123 October 2013, Bahrain

GULF CONVENTION CENTRE, BAHRAIN
www.pipelineconf.com
CONFERENCE
Technical streams presented by industry leaders
covering a wide range of subjects will run over the
two and a half day event.
Some of the subjects to be discussed;
Planning, design, construction and materials
Operations and maintenance
Asset integrity management
Inspection and cathodic protection
Repair and rehabilitation
Automation and control
Leak detection
Paper abstracts are now being accepted.
Join leaders in the international pipeline industry as they converge for the Best Practice
in Pipeline Operations and Integrity Management Conference and Exhibition in Bahrain.
EXHIBITION
A comprehensive exhibition will be part of the
event, allowing companies from around the world
to showcase their products and services. Contact
us today to book your space.

NETWORKING
Throughout the event there will be ample
opportunities to network with participants to
further your business relationships. Meet with
industry leaders from around the world.

Registrations w
ill open soon m
ake sure you attend this landm
ark event.
ORGANIZERS
Held under the Patronage of His Excellency Shaikh Ahmed bin Mohamed Al Khalifa, Minister of Finance,
Minister in Charge of Oil and Gas Affairs, Chairman of National Oil & Gas Authority, Kingdom of Bahrain
bahrain_conf. announcement_ad.indd 1 20/09/12 9:42 AM
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Save money in daily practice by the application of
life-cycle cost optimization which is supported
by a software tool that is based on the extended
Bow-tie methodology, and which enables balancing
of all existing and new mitigation measures for
all life cycles based on efficiency, allowing the
most economical way to be found of operating
the infrastructure.
Find improvements on risk management (new
mitigation measures) in a highly structured way,
and extend the effectiveness of expert-team risk-
assessment sessions to a higher level of quality.
Make decisions on risk-based equipment selection
on the drawing board by analysing the parameters
of a Bow-tie diagram that has been developed using
the equipment characteristics.
Fig.13. Relationship between the preventive barrier values of the Bow-tie (blue line) and the virtual
risk level (red line).
Fig.14. The increased in value of Rp for the new design of valve C compared to that of the old design.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
1 - (1 0.95) x (1 0.9) x (1 0.8) = 0.999 i.e. a
99.9% barrier.
This means that the implementation of the yellow
barriers results in a reduced value of the main
construction barrier down to:
20 x (1 0.001 x 0.25) = 19.995% i.e. a reduction
of 0.005%.
7. The total effect of both escalating factors for the
remaining value of the main construction barrier
when all mitigating measures are implemented is a
reduction down to:
20% x (1- (0.030 + 0.005)) = 19.965%.
8. This reduction of the main barrier will result in
an increase of the effect of the threat of ground
subsidence to the frequency of the loss of
containment unwanted event of:
0.035 x 9.9% = 0.35% (approx.).
9. The effect of all the mitigating measures in the Bow-
tie diagram is calculated in a similar way. Whether
implementation of a certain mitigation measure is
efficient depends on its cost. Once the cost of all
the mitigation measures have been incorporated,
the efficiency of any individual measure can be
calculated, and it can then be ranked on the basis
of its individual efficiency, no matter where it is
positioned in the diagram (or life cycle). This final
table with all the mitigation measures ranked on
their efficiencies enables the operator to take the
next steps required to implement life-cycle cost
optimization.
(Note: The Bow-tie has been developed on the right-hand side of
the unwanted event as well, which enables not only economic
optimization of suppressive measures to be evaluated, but also
enables GTS to find the most efficient control of risk by balancing
between preventive and suppressive measures.)
1. The sum of the effectiveness of the main barriers
should always be 100%, and is factor representing
the relative contribution of each of the life cycles
(design, construction, etc.) for the prevention of
the unwanted event occurring (in this case: loss
of containment).
2. The value of the effectiveness of the main barriers
will be reduced by the effect of escalating factors. An
explanation of this reducing effect is given below.
3. The incorrect assembly escalating factor represents
75% of the reducing effect of both escalations to the
current barrier value of construction. This means
that without any mitigation measure for this factor,
only 25% of the barrier value of construction would
remain, i.e. 25% x 20% = 5%;
4. At this moment only one mitigation measure against
the incorrect assembly escalating factor has been
implemented. This measure is called welding
instructions according to procedure and has a
barrier value of 90% (green barrier). This means
that 90% of the escalating factor is compensated
for and, as a result, 10% of the reducing effect of
the escalation factor still has an effect which results
in a reduction of the main barrier construction
down to:
20 x (1 (0.1 x 0.75)) = 18.5%.
5. The total value of all mitigation measures (incl. the
yellow ones the improvements) for the incorrect
assembly escalating factor is:
1 - (1 - 0.9) x (1 - 0.9) x (1 - 0.8) = 0. 998 i.e. a
99.8% barrier.
This means that the implementation of all barriers
including the yellow ones results in a reduced value
of the main construction barrier down to:
20 x (1 (0.002 x 0.75)) = 19.970%, i.e. a reduction
of 0.030%.
6. The total value of all mitigation measures (including
the yellow ones, the improvements) for the incorrect
application of materials escalating factor is:
Assessment of the integrity of the sub-surface valves in
GTS 40-bar grid has been carried out by using the new
risk-assessment method with the following conclusions:
The sub-surface gate and ball valves are meeting
their end-of-life after 40 years of operation.
Plug, knife-gate, and axial on-off valves last longer
than gate and ball valves; however, they cannot be
used on pipelines which are required to be piggable.
The parameters of ball valves of recent designs
perform better under the combined risk of safety,
environmental damage, loss of reputation, security
of supply, and cost when compared to the 40-year
old designs.
Appendix 1: Algorithms used for effciency calculation
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Appendix 2: Gasunies risk matrix.
F
i
g
.
1
5
.

G
a
s
u
n
i
e
s

r
i
s
k

m
a
t
r
i
x
.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
GET THE
VERY
BESTJOB
www.pipelinesinternational.com/jobs
Visit
and nd your job today
Qualied. Professional. Experienced. Get the very best.
Weve created a dedicated jobs board
for people in the pipeline industry.
Our jobs board exclusively lists jobs in
the pipeline industry, which means you
can nd the very best role that ts your
skills, knowledge and passion.
PIN_Jobs_FP_2012.indd 1 20/04/12 11:28 AM
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
The international gathering of the global pigging industry!
Organized by
Introducing
Do you know a colleague or employee of a pipeline operating, service
or supply company who has demonstrated a valuable and original
contribution to the field of pipeline integrity?
Nominate them now for the PPIM Young Achievement Award.
For more information on award criteria or to nominate a colleague visit
www.clarion.org
The PPIM YOUNG ACHIEVEMENT AWARD 2013
PPIM13_Awards_FP_NEW.indd 1 1/08/12 2:29 PM
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
of pipeline pressure-testing requirements, what
records have been specifically required, how
those records relate to establishing MAOP, why
so-called grandfathered pipelines exist, and the
significance of recently-articulated criteria for
records accuracy.
9.20 [2] IMPACT OF SHALE PLAYS ON PIGGING
by David Wint, TD Williamson, Tulsa, OK,
USA
The high liquids content of production from
shale formations causes significant issues with
slugging, high differential pressures (liquids
loading) and corrosion.High levels of paraffin
and other flow-reducing contaminants can
require pigging some lines on a daily basis. The
great number of lines to be pigged, and the
need to pig often, requires the installation of
multiple automated pig and/or sphere launchers.
Automation of pigging systems offers compelling
economics when compared to traditional manual
systems. This paper will discuss the options
available to meet this need and the savings of
automation over standard manual systems.
9.55 [3] THE USE OF ILI IN ASSESSING THE
LONG-TERM PERFORMANCE OF
SEVERAL COMMERCIALLY AVAILABLE
COMPOSITE-REPAIR SYSTEMS
by Shawn Laughlin, Clock Spring Company,
L.P., Houston, TX, USA, and Jesse Mitchell,
Centurion Pipeline, Houston, TX, USA
Pipelines have been repaired with composite
material for more the two decades. Reports
regarding long-term performance of several
commercially available systems have raised
concerns recently. This paper reviews investigation
of installed composite repairs, in various
TUESDAY, FEBRUARY 12
5.00 Welcome reception and opening of the
Exhibition
WEDNESDAY, FEBRUARY 13
8.00 Registration; Continental breakfast
8.30 Opening remarks
8.45 [1] PRESSURE TESTING AND
RECORDKEEPING: RECONCILING
HISTORIC PIPELINE PRACTICES WITH
NEW REQUIREMENTS
by Michael J Rosenfeld, Kiefner &
Associates, Worthington, OH, USA, and
Rick W Gailing, Sempra Utilities, Los
Angeles, CA, USA
The NTSBs recommendations to US pipeline
operators and regulators, based on their
investigation of the 2010 San Bruno, CA gas
pipeline incident, have focused attention on
validation of MAOP of gas pipelines installed
prior to federal or California state regulations.
The industry is in the process of understanding
the viability of continued operation of
grandfathered pipelines and pipelines that
have experienced loss of records necessary to
give full confidence of the quality of installation.
A full understanding of the implications
requires knowledge of how pressure-testing and
recordkeeping practices and requirements have
evolved over time. This paper explores these
issues, historically and currently, nationally and
in California. Similar issues will likely arise in
other states. The paper describes the evolution
Pipeline pigging conference in
Houston: 25 years
T
HE publishers of the Journal of Pipeline Engineering Great Southern Press of Melbourne, and Clarion
Technical Publishers and Conferences of Houston are proud to announce that the 25th annual Pipeline
Pigging & Integrity Management (PPIM) Conference and Exhibition will be held in Houston, Texas, on 11-14
February, 2013. The frst event in Houston in 1989 built on the strengths of earlier events held in Europe, and
the subsequent development of the series has very much mirrored the success of the international pigging
and inspection industry. We anticipate that over 1,500 pipeline operators and engineers, manufacturers,
and suppliers from around the world will converge on Houstons Marriott Westchase hotel for the event,
to attend presentations on the latest technical issues of interest to the industry, to update their skills at
one of the training courses preceding the conference, and to receive comprehensive product and service
information from many of the industrys biggest names at the accompanying (and sold-out) exhibition.
The following pages preview the conference presentations and give registration information: more
comprehensive and up-to-date information can be found at www.ppim2013.com. Organizers John Tiratsoo
and BJ Lowe, and their colleagues, very much look forward to the opportunity of meeting Journal readers
at the event in February, where old acquaintances can be renewed and new acquaintances developed in
the friendly atmosphere that the event engenders.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
12.45 Lunch in Exhibition area
1.45 [7] THE CHALLENGES OF A LOW-FLOW/
LOW-PRESSURE ILI TOOL PROGRAM
by Bryce Brown, Rosen, USA, Houston, TX,
USA
As natural gas gathering operators begin
assessing pipeline segments as part of their asset
management programs, they are being confronted
with challenging pipeline conditions that are
difficult to inspect with existing ILI technology.
Because of transient operation, low flow rates,
and low system pressures, these gathering
pipelines, which are being voluntarily assessed,
are more difficult to inspect than transmission
lines requiring assessment. Rosen USA and
Access Midstream Partners joined together to
address inspection technology needs for these
lines. The tool development and production was
completed in June 2012, and a successful first run
was executed in July 2012. The success of the
first run can be attributed to the extensive testing
that took place onsite allowing for a better
understanding of the tool performance in live
survey conditions.
2.20 [8] ASSESSMENT OF VINTAGE GIRTH
WELDS AND APPLICATION OF ILI
TOOLS
by Yong-Yi Wang and Dr Jing Ma, Center
for Reliable Energy Systems, Dublin, OH,
USA, and Satish S Kulkarni, Chevron Pipe
Line Co, Houston, TX, USA
The girth welds on many pre-1970s pipelines were
not 100% non-destructively inspected at the time
of construction. These welds may contain flaws
that can lead to failures when the lines experience
stresses beyond those of normal operating
conditions. Recent accidents on vintage pipelines
have led to calls for more effective integrity
management. Fitness-for-service (FFS) principles
offer the best approach to assessment of these
welds. Although general FFS procedures are well
established, enhancement of these procedures is
necessary to take into account the unique features
of vintage girth welds. The paper explores key
elements necessary for enhancing FFS procedures.
The limitations and potential of various ILI tools
in detecting and characterizing girth weld flaws are
presented based on the fundamental principles of
tools and published data.
2.55 Coffee break in Exhibition area
3.30 [9] ENHANCEMENT OF PIPELINE
INTEGRITY MANAGEMENT PLANS
WITH ADVANCED ACOUSTIC
LEAK-DETECTION MONITORING
TECHNOLOGY
by Jeff Robbins and Maurino DeFebbo,
Asel-Tech, Houston, TX, USA
Most internal leak-detection monitoring
systems in use today are based on mass-balance
technology. Although acoustic leak-detection
systems have been around since the late 1970s,
they did not catch on because of inconsistent
performance. Now, advances in digital and
electronics technologies have led to development
environmental conditions and regions of the world,
employed in various types of pipelines. Units have
been removed from service and investigated. The
results are presented and issues are addressed.
11.00 [4] DIRECT ASSESSMENT AND THE GIS
CONNECTION
by Ed Nicholson, NiSource Gas
Transmission and Storage, Charleston,
WV, USA, Amy Jo McKean, Willbros
Engineering, Kansas City, MO, USA, and
Brad Leonard, Willbros Engineering,
Pittsburg, PA, USA
The paper provides an account of ECDA,
SCCDA and ICDA Assessment Program
utilizing GIS databases for compiling, analysis,
field verification and overall data recovery and
storage for the Direct Assessment processes. It
will also show how the results and early finds
saved considerable time and money, along with
Lessons learned that will hopefully save time
and effort by other companies that elect Direct
Assessment as their method of Assessment.
11.35 [5] ESTABLISHING INTEGRITY OF
MARINE PIPELINES DURING
PRECOMMISSIONING AND
COMMISSIONING
by Raed Gasim, Intecsea WorleyParsons
Group, Houston, TX, USA
This paper discusses precommissioning
and commissioning processes and appraises
techniques and best practices for pipeline
filling, cleaning, flushing, chemical treatment,
hydrostatic testing, dewatering, and drying,
including the filling and cleaning of pipelines
that contain corrosion-resistant alloys. The main
causes of internal corrosion during pipeline filling
and hydrostatic testing are also examined.
12.10 [6] HIGH-RESOLUTION CALIPER
INSPECTION TO IDENTIFY PIPE
EXPANSIONS
by Jason Matocha,TD Williamson, Aurora,
CO, USA
Ruby Pipeline Company LLC recently
constructed a 680-mile, 42-inch diameter
pipeline, which was designed under the
Alternative MAOP rule (80% SMYS design).
Due to potential quality issues on previously
manufactured high-yield-strength pipe, PHMSA
mandated the use of high-resolution caliper
technology for operators to inspect for pipe
expansions resulting from hydrostatic testing.
Additional requirements were to perform a
minimum of two verification digs per segment
inspected, and to identify all expansion
greater than 0.6% for 42-inch pipe. The tool
specifications required multi-finger sensors that
contact the pipe internal diameter and have an
accuracy of +/- 1% or less to identify expanded
pipe and dents. The paper reports on how T.D.
Williamson was able to assist Ruby Pipeline in
meeting this requirement, providing project
management services for 341 miles of inspection,
tracking and non-destructive evaluation (NDE).
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
THURSDAY, FEBRUARY 14
7.30 Registration; Continental breakfast
8.00 [12] RISK CONTROL AT LOWER COST
by Vishal Pooran, Sasol Gas, Johannesburg,
South Africa, Leen Pronk, Gasunie,
Groningen, Netherlands, Peter Baars,
GDF SUEZ, France, and Rob Bos,PIMS
International, Haren, The Netherlands.
Integrity management is a process of continuous
improvement and should be an integral part of
the daily practice of the pipeline operator. Risk
assessment plays a key role in this process: (new)
risks are identified and mitigated by appropriate
(new) measures that are the basic elements for
the Annual Periodic Integrity Management Plan.
Any Safety Management System (SMS) has a
primary focus on the effectiveness of integrity and
risk control but, in general, costs are of secondary
importance. The methodology described in this
paper brings these ideas together: it supplies the
tools that enable the pipeline operator to manage
integrity and risk in an economical way.
8.35 [13] VALIDATING EMAT ILI
PRIORITIZATION CRITERIA FOR
THE DETECTION OF SCC CRACKS: A
PROGRAM REVIEW
by Nader Al-Otaibi and Mashaan Al-
Shammari, Saudi Aramco, Dharan, Saudi
Arabia
Following an EMAT ILI program undertaken
over the last few years, Saudi Aramco discovered
a number of SCC defects in its pipelines.
Prioritization criteria based on operating
parameters, age and condition were established
to manage utilization of the EMAT technology.
An extensive field-verification program was
carried out that demonstrated the capabilities of
the tools. In response to the discovery of SCC,
Saudi Aramco managed the issue by building
on the collective knowledge and experience of
its engineers. The paper will present the main
findings of the inspection and assessment program.
A case study of a gas pipeline in which SCC
cracks were found will be provided to elaborate on
the actions taken in response to discovery of the
cracks, and how the prioritization criteria aligned
with inspection results.
9.10 [14] DETERMINATION OF CORROSION
GROWTH RATES USING PIPELINE
OPERATORS FORUM DEFECT
CLASSIFICATIONS
by Toby Fletcher, Wood Group Integrity
Management, Newcastle upon Tyne, UK
Assessing how quickly defects may corrode
can be a complex undertaking. To simplify
such assessments, a single corrosion growth
rate may be applied to all defects in a pipeline;
this approach is frequently taken as the basis
for determining an ILI re-inspection date. The
Pipeline Operators Forum provides guidance
on classifying defects based on pipeline wall
thickness, defect axial length and defect
circumferential width. Examples include pitting-
type defects, general-corrosion-type defects and
of new acoustic systems based on more-
sophisticated architecture and advanced signal
processing techniques. The paper will review
these new systems, which work substantially
better than their predecessors, and in most cases
better than other CPM/mass-balance systems.
4.05 [10] A FREE-SWIMMING ILI ACOUSTIC
LEAK-DETECTION SYSTEM FOR
WATER AND PETROLEUM PIPELINES
by Anis Somani, Pure Technologies,
Columbia, MD, USA, and David W Kurtz,
Pure Technologies, Dallas, TX ,USA
A joint academic-industry research initiative
funded by PHMSA has led to the refinement
of a free-swimming tool capable of detecting
leaks as small as 0.150 liters per minute in oil
product pipelines. The tool swims through the
pipeline being assessed and produces results at
significantly reduced cost compared to current
leak-detection methods. GPS-synchronized,
GIS-based above-ground loggers capture low-
frequency acoustic signatures and digitally log
the tools passage through a pipeline. A tri-
axial accelerometer system gives the odometric
position of the ball, and has the accuracy of
standard instrumented pigs. This paper will focus
on the technology, analysis and performance of
the device in petroleum pipelines.
4.40 [11] PIPELINE RISK ASSESSMENT: WORKING
TOWARDS A UNIFIED APPROACH
by W Kent Muhlbauer, WKM Consultancy,
Houston, TX, USA
Some regulators who have expressed increased
skepticism regarding how pipeline operators are
measuring risks. Regulators recent criticisms
are not unjustified. There is currently great
disparity in approaches and level of rigor
applied to risk assessment. This is largely
due to the absence of complete standards or
guidelines covering this complex endeavor. The
disparity leads to inconsistent and problematic
oversight by regulatory agencies. Without
some standardization or at least consistency
of understanding, regulators cannot readily
determine where deficiencies may lie. On
the other hand, too much standardizationa
mandated, prescriptive approachis inefficient
and stifles innovation. A better solution is to
establish guidelines of essential ingredients
necessary in any pipeline risk assessment. Critical
elements would be identified and it would be left
to the operator subject matter experts (SME)
to detail those elements. This paper presents
such a set of essential elements for pipeline risk
assessment.
5.15 Reception in Exhibition area
6.30 End of Day 1
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Based on this experience and understanding, it is
considered that EMAT ILI can be used in lieu of
hydrostatic testing for SCC threat management,
provided that the uncertainties are properly
understood and addressed during the crack-
detection, sizing and evaluation procedures.
11.25 [17] SUPPORT PIPELINE INTEGRITY
MANAGEMENT BY IDENTIFICATION
AND ASSESSMENT OF PIPELINE
COATINGS USING EMAT ILI DATA
by Richard Kania, TransCanada Pipelines,
Calgary, AB, Canada, Ralf Weber, ILI
Consulting, Karlsruhe, Germany, and
Stefan Klein, Nikola Jansing, and Michael
Meuer, Rosen Innovation Center, Lingen,
Germany
TransCanada Pipelines asked Rosen in 2012
to investigate the feasibility of EMAT ILI data
towards identification and assessment of tape
coating at girth-welds. The regulatory demand was
to replace all tape coating at welds of a 24 FBE
coated gas transmission line. TransCanada and
Rosen initiated a feasibility study and inspection
project using an EMAT ILI tool. Successful
identification of individual tape-coated welds from
the EMAT data and results of direct examination,
successfully allowed for a repair program meeting
all regulatory demands and timelines. It helped
TransCanada save costs with respect to direct
examination of all welds considered in this project.
This paper will discuss the technical background
of the applied EMAT technique, procedures and
results of this project.
12.00 [18] A PIPELINE INSPECTION CASE STUDY:
DESIGN IMPROVEMENTS ON A NEW
GENERATION UT ILI CRACK TOOL
by Mark Slaughter, Weatherford, Houston,
USA, Michael Huss, Adria-Wien Pipeline
GmbH (AWP), Klagenfurt, Austria, and
Yuriy Zakharov and Andrey Vassiljev,
Weatherford, Moscow, Russia
The predominant ILI applications utilizing
ultrasonic technology have been for wall loss
and crack inspection. Despite a high success rate,
technological improvements are needed. Among
them are higher confidence in the Probability of
Detection (POD), improving detection reliability
under different conditions, increased ranges
for pipeline operating parameters and utilizing
combo WM-CD mode in one run. In 2010,
Weatherford Pipeline and Specialty Services
(P&SS) commissioned its new generation fleet of
ultrasonic wall measurement and crack-detection
tools. This paper reviews the latest design
improvements for the tools and presents a case
study on a recent survey conducted on the Adria-
Wien Pipeline (AWP).
12.35 Presentation of PPIM Young Achievement
Award
12.50 Lunch in Exhibition area
pinholes. This paper examines the use of these
classifications to estimate corrosion growth rates
by fitting the dimensions of defects to different
statistical distributions. The use of these methods
helps to refine estimates of corrosion growth rates
and determine more accurate intelligent pig re-
inspection intervals.
10.15 [15] PIPELINE CRACK ASSESSMENT: NEW
TECHNOLOGY VS OLD-SCHOOL
METHODS
by Ted L. Anderson, Quest Integrity Group,
Boulder, CO, USA
The integrity of pipelines with cracks and other
planar flaws has received renewed interest by
operators and regulators, due primarily to a
number of high-profile incidents. The pipeline
industry currently assesses planar flaws with
methodology that dates back approximately 40
years. The traditional crack assessment models
can lead to gross errors in the prediction of burst
pressure. This paper points out the problems
with these models, using both theoretical analysis
and a comparison with burst test data. The field
of fracture mechanics has advanced considerably
in the last 40 years, and improved methods are
available. This paper describes several such
methods, and uses burst test data to demonstrate
that state-of-the-art models lead to much more
accurate predictions of failure pressure and
critical flaw size.
10.50 [16] USING EMAT ILI IN LIEU OF
HYDROSTATIC TESTING TO
MANAGE THE THREAT OF STRESS-
CORROSION CRACKING: ISSUES TO BE
CONSIDERED
by Dr David Batte, Macaw Engineering,
Newcastle upon Tyne, UK, Dr Ray R Fessler,
Biztek Consulting, Evanston, IL, USA, Mark
L Hereth, PPIC, Bloomeld, CT, USA, Jim
Marr, TransCanada Pipelines, Calgary, AB,
Canada, and Steve C Rapp, Spectra Energy,
Houston, TX, USA
During a recently-completed Joint Industry Project
(JIP), eight major gas pipeline operators collated
and reviewed their experience with current-
generation EMAT ILI for detection, sizing and
evaluation of stress corrosion cracking (SCC).
Results were available from over 45 pipeline
inspections totalling more than 3000 miles, during
which over 100 features have been confirmed
by excavation to be SCC that would probably
have failed a hydrostatic test. This experience
has demonstrated that features of the size that
would just survive a hydrostatic test can readily be
detected using EMAT ILI. Other work completed
during the JIP has established that the failure
pressures of pipe containing SCC can be predicted
conservatively to within +/- 20%.. Finally, the
reliability of these methods for SCC threat
management has been confirmed by recent field
experience with 16 valve sections where EMAT
ILI has been followed by hydrostatic testing.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
3.00 [21] STUDY OF PIPELINES THAT RUPTURED
WHILE OPERATING AT A PRESSURE
BELOW 30 % SMYS
by Michael Rosenfeld, Kiefner & Assoc./
Applus RTD and Robert Fassett,
Kleinfelder, Santa Rosa, CA, USA
This paper provides data on pressure-related
ruptures due to interacting threats on pipelines
that were operating below 30% SMYS. The
broader discussion will consider areas of
technology that the industry may want to focus on
to address issues that may affect local distribution
companies more than interstate natural gas
operators. The paper will also discuss the types of
interaction that caused the ruptures and provide a
high-level decision tree that will allow operators
to begin to address how to model their systems
to determine if they may have these same threat
combinations.
3.35 [22] LEVERAGING ADVANCED DATA
MANAGEMENT AND GEOGRAPHIC
INFORMATION SYSTEMS
TECHNOLOGY IN A CHANGING
REGULATORY CLIMATE
by Colin Cochran, Williams Gas Pipeline,
Houston, TX, USA
When coupled with a structured relational data
model, GIS provides a sustainable platform for
the unique data integration requirements of oil
and gas pipeline companies, especially in support
of risk assessment and integrity management
activities. This presentation presents lessons
learned by Williams in the implementation
of GIS as an enabling technology for integrity
management and regulatory compliance. The
presentation includes a review of best practices
to support High Consequence Area (HCA) and
class location analysis, and ongoing management
of modeling results. WGP best practices for
integration of inspection and survey data
from multiple vendors and sources will also be
reviewed, along with risk modeling techniques.
Finally, the presentation will include a summary
of next steps that are underway to proactively
respond to the changing regulatory environment
in the US.
4.10 End of conference
1.50 [19] ENHANCED USE OF ILI DATA TO
IMPROVE INTEGRITY DECISIONS
by Collin Taylor, Enbridge Pipelines Inc,
Edmonton, AB, Canada
This paper will explore the benefits of a statistical
alignment method utilizing joint
characteristics such as length, long-seam orientation
(LSO), wall thickness (WT) and girth weld (GW)
counts to ensure precision data alignment between
ILI inspections. By leveraging the fingerprint-
like morphology of a pipeline system, many
improvements to data and records systems become
possible, including but not limited to:
Random ILI Tool performance errors can be
detected and compensated for.
Repair history and other records become
rapidly searchable.
New statistically accurate descriptions are
created by leveraging the sensitivities of
various ILI technologies.
2.25 [20] IMPROVEMENTS IN THE ENGINEERING
ASSESSMENT PROCESS FOR METAL
LOSS IN-LINE INSPECTIONS
by Ian Smith and Lisa Barkdull, Quest
Integrity Group, Houston, TX, USA
This presentation will demonstrate the application
of an all-inclusive ILI data set engineering
assessment using API 579 fitness-for-purpose
methodology and a comparison with the traditional
data analysis/engineering assessment approach.
By utilizing compression wave ultrasonic ILI data
and API 579 fitness-for-purpose Part 5 Level 2
methodology, the engineering assessment process
can be applied directly to an ILI data set, not
just the threshold- and boundary-limited data
found in a spreadsheet or database. The API 579
methodology evaluates the remaining strength of
the complete pipeline, using all of the ultrasonic
wall thickness measurements, not just assessing
areas where metal loss has been identified and
boxed through data analysis process. By assessing
the entire ILI data set, thresholds and boundaries
no longer limit information the operator receives
about the pipeline. Advantages of this approach
include accuracy, no dependence on metal-loss
interaction criteria, repeatability and more
informative run comparisons.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
www.ppimhouston.com
Cardholders Name

Expiration date: /
Credit card number CCV No:
REGISTRATION FORM COURSES | CONFERENCE | EXHIBITION
VENUE & ACCOMMODATION
Fees do NOT include accommodation.
The event will be held at Houston Marriott Westchase Hotel
2900 Briarpark Dr. Houston, TX 77042 USA
+1 713 978 7400 or +1 800 452 5110 +1 713 735 2726 (FAX)
Our group rate is $153 + tax. Refer to Group Code PIPPIPA.
Dont forget to say you are attending the Pipeline Pigging &
Integrity Management Conference and Courses to take
advantage of the special rate (limited availability).
CONFERENCE ONLY February 13-14
$995
EXHIBITORS ONLY (for Companies who are exhibiting)
Conference only - exhibitor company: $895 (multiple registration discount not applicable)
Exhibition-only - exhibitor company: $95
COURSE ONLY February 11-12
A. Pipeline Risk Management: $2695
B. DOT Pipeline Safety Regulations: $2695
C. Introduction to Excavation Inspection: $2695
D. Pigging & In-line Inspection: $2695
E. Pipeline Defect Assessment Calculations Workshop: $2695
F. Pipeline Defect Assessment: $2995
G. Pipeline Integrity Management: $2695
H. Pipeline Repair Methods / In-Service Welding: $2695
COURSE (Any) + CONFERENCE ($3490 for A,B,C,D,E,G,H. $3790 for F)
A B C D E F G H
NOTE: If you decide to register for an additional event (course or conference) separately, or at a later date, the combined rate will apply.
VISITOR REGISTRATION (EXHIBITION ONLY)
$25
I have a coupon. Enter coupon code.
NOTE: "Visitors" are here defined as persons who are not employees, representatives or affiliates of exhibiting companies. Personnel employed by or
representing exhibiting companies must register as additional exhibitor staff.Exhibition hours: see page 9.
DISCOUNTS (applies to conference and course fees only)
Early registration prior to January 6, 2013: $50 off
Multiple registrations from the same company: $100 off per person
(Please submit a separate registration form for each registrant. We will credit the discount to each person.
It is not necessary for everyone to send their form in at the same time.)
PIPE members - 10% discount (may not be combined with other discounts. Individual members only.)
PIPE member number:
PAYMENT OPTIONS (check as appropriate)
I will mail a check payable to Clarion Technical Conferences
Please send me an invoice
Bank transfer
JP Morgan Chase Bank NA, 600 McGowen St., Houston, TX 77006, USA. Tel. +1 713 533 1309.
SWIFT #CHASUS33. Routing #111000614. Account #875665622 (Clarion Technical Conferences)
Credit card (statement billing address required - see below) Mastercard Visa American Express
FOR FASTEST REGISTRATION, REGISTER ONLINE NOW
at www.clarion.org (secure server.) OR fax or mail this form to:
CLARION Technical Conferences
TM
3401 Louisiana Street, Suite 255, Houston, TX 77002
FAX +1 713 521 9255 TEL +1 713 521 5929
NOTE: If you decide to register for an additional event
(course or conference) separately, or at a later date,
the combined rate will apply.
Name Title
Company Email
Billing Address
City

State Postcode/Zip
Country Telephone
BILLING AND CONTACT INFORMATION
CANCELLATIONS/SUBSTITUTIONS
Cancellations made in writing and received on or before 10 business days prior to the event will
be refunded less a US$100 handling fee. Cancellations received later than 10 business days prior
to the event will not be refunded, and the full invoice fee will be payable regardless of whether
you attend the event or not. Substitutions may be made at any time. Confirmation will be made
in writing as soon as possible upon receipt of payment. This confirmation will be sent to the
address given on the registration form, unless otherwise required. The organizers reserve the
right to cancel any event due to insufficient enrollment. In this event fees will be refunded in full.
However, the organizers assume no liability for travel or any expenses other than fees paid.
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
Pipelines International Premium is the international
oil and gas pipeline industrys foremost in-depth
source of information, comprising a digest of
high-quality papers covering the latest technology
and reviews of the pipeline industry worldwide, and
a comprehensive project database.

It is comprised of:
Pipelines International Digest which provides a
monthly update of papers covering all areas of the
industry from key projects, and engineering and
construction issues, to environmental, regulatory,
legal and nancial issues.
Pipelines International Projects which allows subscribers
to access a searchable database of completed and
current projects.

Subscribe or get a free 14 day trial now at
www.pipelinesinternational.com/premium
PIN_Premium_FP.indd 1 26/11/12 12:51 PM
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n
S
a
m
p
l
e

c
o
p
y

n
o
t

f
o
r

d
i
s
t
r
i
b
u
t
i
o
n

JPE Dec 2012 - Sample Issue

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

JPE Dec 2012 - Sample Issue

Uploaded by

Copyright:

Available Formats

December, 2012 Vol.12, No.

2123 October 2013, Bahrain

You might also like