VEDLIKEHOLDSSTYRING - KORROSJON UNDER ISOLASJON (KUI)

Ptil - Maintenance management -

corrosion under insulation (CUI)
Petroleumstilsynet

Report No.: 2021-4107, Rev. 02

Document No.: 1249481
Date: 2022-01-21
Project name: Vedlikeholdsstyring - Korrosjon under isolasjon (KUI) DNV AS Energy Systems
Report title: Ptil - Maintenance management - corrosion under Integrity Management
insulation (CUI) Veritasveien 25
Customer: Petroleumstilsynet, Professor Olav Hanssens vei 10

4007 Stavanger
4021 STAVANGER Norway
Norway Tel: +47 51506000
Customer contact: Morten A. Langøy 945 748 931
Date of issue: 2022-01-21
Project No.: 10301306
Organisation unit: Integrity Management
Report No.: 2021-4107, Rev. 02
Document No.: 1249481
Applicable contract(s) governing the provision of this Report:

Objective:
Review and evaluation of publicly available guidelines and operators’ in-house strategies for CUI management.

Prepared by: Verified by: Approved by:

Wiggen, Digitally signed by
Wiggen, Frode
Digitally signed by Jessen, Jan
Vasland
Digitally signed by Eikeland,
Kjetil
Frode Date: 2022.01.21
09:18:23 +01'00'
Date: 2022.01.21 09:36:23
+01'00'
Date: 2022.01.21 10:16:45
+01'00'
Frode Wiggen Jan Vasland Jessen Kjetil Eikeland
Senior Principal Engineer Principal Engineer Head of Section

Copyright © DNV 2022. All rights reserved. Unless otherwise agreed in writing: (i) This publication or parts thereof may not be copied, reproduced or
transmitted in any form, or by any means, whether digitally or otherwise; (ii) The content of this publication shall be kept confidential by the customer; (iii)
No third party may rely on its contents; and (iv) DNV undertakes no duty of care toward any third party. Reference to part of this publication which may
lead to misinterpretation is prohibited.

DNV Distribution: Keywords:

☒ OPEN. Unrestricted distribution, internal and external. CUI, PSA, Strategy, Methodology, Methods, PoF,
☐ INTERNAL use only. Internal DNV document. Risk
☐ CONFIDENTIAL. Distribution within DNV according to applicable
contract.*
☐ SECRET. Authorized access only.
*Specify distribution:

Rev. No. Date Reason for Issue Prepared by Verified by Approved by

0 2021-12-20 First issue, for comment by PSA

01 2022-01-14 Issue with comments from PSA implemented FWIG JVJ EIKE

02 2022-01-21 Minor update in chapter 2.1 FWIG JVJ EIKE

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page i

Table of contents
1 EXECUTIVE SUMMARY................................................................................................................................... 1

2 INTRODUCTION ............................................................................................................................................... 2
2.1 Background 2
2.2 Abbrevations 3

3 BASIS FOR WORK ........................................................................................................................................... 4

4 REGULATIONS, GUIDELINES AND STRATEGIES ......................................................................................... 4

4.1 Regulation and legislation 4
4.2 Publicly available guidelines and standards 4
4.3 Company specific strategies 8
4.4 Design requirements leading to increased CUI risk 9
4.5 CUI historical data 10

5 EVALUATION OF DIFFERENT STRATEGIES AND GUIDELINES ................................................................ 11

5.1 Guidelines 11
5.2 Industry approach and strategies 12
5.3 Implementation 12

6 CONTINUOUS IMPROVEMENT AND LESSONS LEARNED ........................................................................ 13

6.1 Trends 13
6.2 Best Available Technology 14
6.3 Operator experiences 15

7 CONCLUSION ................................................................................................................................................ 17

8 REFERENCES ................................................................................................................................................ 18

A1. WORKSHOP AGENDA ..................................................................................................................................... 2

A2. PARTICIPANTS ................................................................................................................................................ 2

A3. BACKGROUND ................................................................................................................................................ 2

A4. REGULATORY REQUIREMENTS .................................................................................................................... 3

A5. FEEDBACK QUESTIONAIRE ........................................................................................................................... 3

A5.1. Question: Main challenges 4
A5.2. Question: Detection of CUI 5
A5.3. Question: Most common immediate consequences 6
A5.4. Question: Assessment model 7
A5.5. Question: Parameters for probability assessment 8

Appendix A Work shop

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page ii

1 EXECUTIVE SUMMARY
DNV has on behalf of the Petroleum Safety Authority Norway (PSA) reviewed the current industry practice with respect
to management of CUI within the oil and gas industry in Norway. The scope of work includes both the review of public
available guidelines and operators’ in-house strategies for CUI management as well as identification of learning and
improvement opportunities. The work has been organised as review of guidelines, questionnaires to operators as well
as a workshop with 7 operators on the Norwegian continental shelf.

Norwegian:
DNV har på vegne av Petroleumtilsynet gjennomført oppdraget «Identifisering av KUI-aktiviteter med hensyn til fare og
ulykkesrisiko, risikoelementer, kompenserende tiltak og vedlikholdsstyring i petroleumsnæringen». Arbeidsoppgaven har
vært å beskrive hvilken praksis som benyttes i næringen og eventuelt hvilke nasjonale og internasjonale standarder den
bygger på samt å identifisere læringspunkter på tvers i en samlet næring. Arbeidet har vært gjennomført i form av
gjenomgang av retningslinjer, spørreundersøkelser og workshop sammen med 7 operatører.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 1

2 INTRODUCTION

2.1 Background
Corrosion under insulation is one of the major threats to the integrity of process plants in the oil and gas industry. A
study performed by PSA indicates that about 50% of the reported hydrocarbon leaks at onshore plants are caused by
CUI.

Figure 1 CUI related hydrocarbon leaks reported to PSA.

The figure above gives an overview of the incidents reported to the PSA over the last 23 years. Root cause analysis has
identified these leaks to be caused by CUI.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 2

2.2 Abbrevations

Abbrevation Explanation

API American Petroleum Institute

CoF Consequence of Failure

CRA Corrosion Resistant Alloy

CUI Corrosion Under Insulation

EFC European Federation of Corrosion

HDG Hot Dipped Galvanized

IOGP International Association of Oil and Gas Producers

NCS Norwegian Continental Shelf

PoD Probability of Detection

PoF Probability of Failure

PSA Petroleum Safety Authority, Norway

Ptil Petroleumstilsynet, see PSA

TSA Thermal Sprayed Aluminimum

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 3

3 BASIS FOR WORK
The basis for the work has been reference documents as listed in section 7, results from the workshop held with
operators on the Norwegian Continental Shelf (NCS) on November 9th, 2021, see Appendix A, and DNV experience
and knowledge related to CUI.

4 REGULATIONS, GUIDELINES AND STRATEGIES

4.1 Regulation and legislation

Norwegian legislative requirements for design and operation of offshore facilities are covered by the legislation from
Petroleum Safety Authority. Requirements related to design with relevance for CUI can be found in the facilities
regulations and requirements for insulation needs is covered both in paragraph §11, §12 and §29.

The Norwegian legislation, ref. the activities regulations chapter IX, including §47 – Maintenance programme, set
requirements to the maintenance program for “failure modes that may constitute a health, safety or environmental risk”.
The guideline to §47 details that “The DNV-RP-G101 guideline may be used to establish the inspection programme for
process plants and auxiliary systems”.

The DNVGL-RP-G101 Risk-based inspection of offshore topsides static mechanical equipment was updated in January
2021 and the current revision refers to DNVGL-RP-G109 Risk-based management of corrosion under insulation for
assessment of risk related to CUI, while the 2017 revision of DNV-RP-G101 gives a rate calculation formula for CUI.

The Management regulations § 23 Continuous Improvement and the Framework regulations § 24 Use of recognised
standards is also relevant to management of CUI

4.2 Publicly available guidelines and standards

Based on a thorough review of available guidelines in the industry the following documents have been identified as the
most relevant and are further detailed and compared in this report:

• API 581, “Risk-Based Inspection Methodology”/1/

• API 583, “Corrosion Under Insulation and Fireproofing”/2/

• DNVGL-RP-G109, “Risk Based Management of Corrosion Under Insulation”/3/

• Energy Institute, “Guidance for corrosion management in oil and gas production and processing”/6/

• EFC no. 55, “Corrosion-Under-Insulation (CUI) Guidelines”/5/

• HOIS-G-023, “Guideline for in-situ inspection of corrosion under insulation”/7/

• NACE SP0198, “Control of Corrosion Under Thermal Insulation and Fireproofing Materials – A systematic
Approach”/4/

The 5 first documents are considered guidelines for management of corrosion under insulation, while the NACE
document focuses on design to avoid CUI. The HOIS document focuses on NDT to detect CUI damage.

Highly relevant design documents such as NORSOK M-004 “Piping and equipment insulation” and NORSOK M-501
“Surface preparation and protective coating” were not included in this review as they are considered specific design

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 4

documents for insulation and coating. These documents will also be the main input content of the new specifications
issued by IOGP under the JIP33 initiative:

S-715 Coating and Painting for Offshore Coastal and Subsea Environment

S-738 Insulation for Piping and Equipment

4.2.1 Technical content of the reference documents

Table 4-1 describes the referred documents in 4.2 with respect to their technical content in the following categories:

• Risk

• Material degradation
• Coating

• Insulation

• Design
• NDT

• Maintenance

• Susceptible areas (Hot-spots / Focal points)

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 5

Table 4-1 Overview of technical content in selected documents relevant for managing CUI.

Standard Risk Material Coating Insulation Design NDT Maintenance Susceptible Comment

degradation areas (hot-spots)

API 581 Quantitative Calculate corrosion Use 3 different levels Adjust Corr rate for Extensive minimum Uses NDT results NA NA Quantitative risk analysis for
approach rate. Basis is of coating protection insulation type (x wall calculation incl (Inspection refineries, Part 2 section 16
temperature and (0, 5, 15 y) 1,25 or x 0,75 or x 1) stress calculations efficiency) in calc of calculate damage factor for
water wetting wall thickness CUI in Carbon steel. Very
conditions detailed calculations using
many simplifications.

API 583 Annex A Describe the CUI Listing different Describe total 10 Giving advice on Describe 12 different Giving advice Chapter 5 (6 Describe different elements
(informative), degradation coating systems but different types of how to design to inspection methods on pages) identifying related to CUI. Advice on
qualitative mechanism and refer to NACE insulation and some avoid CUI, chapter with advantages and maintenance in total 57 focus design (chapter 9),
approach identify factors SP0198 for details jacketing (cladding) 9. disadvantages and repair areas distributed maintenance and mitigation
affecting CUI types with strategies with on general (20), (chapter 11)
advantages and respect to CUI vessels (10),
disadvantages. piping (22) and
Insulation techniques tanks (5)
described in Annex
B

DNVGL-RP- Qualitative Material degradation Lifetime of coating Insulation and Diameter and wall Referring to HOIS-G- Effect of Generic list of 11 Focus on risk management
G109 approach is assessed based assessed based on cladding assessed thickness is included G16 but set maintenance focus areas
on operating 13 different coatings based on design in assessment requirement to PoD is included in
temperature and quality of solution and risk update
workmanship workmanship

EFC no.55 Describe the Describe the Semi-Q assessment Semi-Q assessment Design for Listing 13 different As part of CUI Susceptible areas Address elements such as
RBI process chemical differentiating TSA differentiating prevention of CUI is inspection methods mitigation listed for different cost, quality assurance,
for qualitative, degradation and organic coating engineering addressed in chapter with limitations, (chapter 7), types of organization, responsibility,
semi- mechanism. Also as well as QA/QC standards, age and 8, this cover design advantages and advise are equipment (piping, inspection strategies and CUI
quantitative present corrosion and “poor/high” maintenance to avoid water disadvantages. given on best vessels, heat mitigation.
and qualitative rate as a function of quality coating. Also programs. Local wetting and use of Table 6.2 gives design / exchangers) in
analysis. temperature (same separate appendix D environment is also CRA materials as effect of different upgrade to chapter 5.3
Users are as NACE SP 0198). refer to NACE SP addressed in the well as TSA local conditions on avoid CUI.
cautioned Upper temp limit set 0198 and Appendix semi-Q approach. the ability of the NDT This includes
against to 175°C E addressing TSA Insulation also method. NDT TSA, CRA
commercial covered in appendix methods also material and
RBI programs F and G in EFC 55 covered in appendix limited use of
that attempt to I of EFC 55I insulation.
calculate Appendix C1
corrosion rate cover
for CUI “Maintenance
and
remediation
issues”

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 6

 Standard Risk Material Coating Insulation Design NDT Maintenance Susceptible Comment

degradation areas (hot-spots)

Energy Referring to Relate degradation General comments General comments Limited input Refer to HOIS-G-023 Give some The term “prone To a large degree adjusted to
Institute UK regulations of CS to age, and practical advice. and practical advice. for NDT methods advice related areas” are used fit UK regulation. All activities
for risk temperature and Promote TSA if to fabric and described at a related to plan-do-check-act
«Guidance management. water wetting without applied correctly Claim upgrade and Refer to DNV-RP- maintenance high level in principles in the improvement
for corrosion (Annex B3). giving any formula or opposed to organic control of insulation G101 for inspection and insulation I.12.5.3 loop. Annex I.12 address
management Describing approach for coatings. solution to be more effectiveness maintenance CUI. Referring to EFC no. 55
in oil and general work assessing the PoF. relevant in late life
gas process for than coating Recommend a set scheme
production corrosion risk condition. (ref I.12.5.2) for a CUI RBI
and assessment scheme given for High risk:
processing” (Annex F) Advice NOT to give regular CVI with removal of
referring to too much weight to insulation in prone areas
API 580/581 insulation type in (every 2-4 year) combined
and DNVGL- CUI PoF with full strip and full CVI in
RP-G101 assessment less frequent intervals (every
7-10 year). For Medium risk
less frequent inspection
scope is recommended

HOIS-G-023 NA NA NA Discuss different NA Very extensive NA NA Main focus is assessment of

insulation solutions different NDT methods ability
vs NDT to find CUI. Describing
application and limitations for
both standard and advanced
NDT methods

NACE NA Describe the Giving advice on Describing Design to avoid CUI Listing 9 methods NA Listing total of 13 Primarily giving advice for
SP0198 degradation different types of application and is addressed in without any details focus locations design.
mechanism and coating to be used limitation of 8 section 3 on application and and 23 awareness
influencing factors. under different different groups of limitations elements.
Present CUI conditions (temp.) insulation material
corrosion rate as a
function of
temperature
(graphical)

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 7

4.3 Company specific strategies
A workshop with participants from 7 operators on the Norwegian Continental Shelf and onshore assets under PSA
regulation was held November 9th, 2021. The main topic for this workshop was to discuss the respective operators’
strategies to manage CUI, discuss experiences and learn from each other.

All operators had a specific strategy to manage CUI, two of these were based on internally developed methodologies,
while 5 were based on acknowledged recommended practices: DNVGL-RP-G109 (3) and DNVGL-RP-G101, rev. 2017
(2). The latest revision of DNVGL-RP-G101, rev. 2021, refers to DNVGL-RP-G109 for assessment of CUI.

The differences in strategies also reflect the different type of assets the operator is responsible for. Where one operator
might only have one relatively modern asset with extensive use of CRA materials, other operators have a wide variety of
both old and new assets.

The involved companies have different work processes, although all could relate to plan-do-check-act principles, or the
maintenance loop as described back in the 1990’s. Obvious differences relate to how well the relevant disciplines are
involved in the CUI management. One operator properly involved relevant disciplines in weekly meetings to discuss and
manage CUI, while others had problems with alignment and cooperation between maintenance and technical integrity
for example.

In the evaluation of probability of CUI, there exists large differences in complexity and number of parameters used in the
assessments. The companies using the DNVGL-RP-G109 generally take more parameters into account than those
using DNVGL-RP-G101. For the two internally developed methodologies there are large differences in complexity and
use of parameters, where one methodology is close to DNVGL-RP-G109 in terms of number of parameters and
approach, while the other methodology is rather simplistic, with few parameters used.

Several of the companies had specific CUI strategies for deck penetration and HDG bolts in CRA systems as this these
were identified as specific design details that were exposed to CUI. Other noted challenges were low quality in the
workmanship of insulation and coating for nozzles that led to increased CUI challenges after 10-15 years in operation.

The different operator-specific strategies could be divided into 4 main approaches:

• Use of DNVGL-RP-G109

• Use of DNVGL-RP-G101 rev 2017

• Use of internally developed strategy based on rate model, complex

• Use of internally developed strategy based on rate model, simplistic

Most companies base their CUI inspection campaigns on risk identification and visual external inspection followed by
partial removal of insulation for close visual inspection of the coated steel.

4.3.1 Use of DNVGL Recommended Practices

The main difference between DNVGL-RP-G101, rev. 2017, and DNVGL-RP-G109 is that the former(G101) is based on
a quantitative approach using a rate model as a function of temperature, while the latter (G109) is a semi-
quantitative/qualitative model based on assessments of many more parameters, giving a probability of failure ranging
from very low to very high. The latest revision of DNVGL-RP-G101, rev. 2021, refers to DNVGL-RP-G109 for
assessment of CUI.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 8

The DNVGL-RP-G109 approach is a semi-qualitative approach where several parameters are included in the
assessment. The three companies using this approach use 13 to 17 parameters to assess the probability of CUI. This
approach will enable a dynamic risk assessment where the update of risk based on relevant mitigation is important. The
risk assessment will identify when the risk become unacceptable based on consequence of failure, condition of coating,
condition of insulation (water wetting), design, material selection and operational temperature.

4.3.2 Internal company developed strategies

The complex internally developed assessment methodology holds several of the elements described in DNVGL-RP-
G101 e.g., coating assessment. Time for mitigation is based on remaining life calculation given a calculated corrosion
rate. The main features of the probability assessment relate to:

• Design of pipes and vessels (e.g. material type, pipe diameter, wall thickness)

• Coating type and insulation type

• Operating temperature

• Design pressure

• External conditions

• Consequence category

The resulting probability assessment will normally be on the conservative side based on assumptions such as
consistently wet insulation, a corrosion model that assumes uniform corrosion opposed to the normally localised
corrosion, and remaining life based on design pressure. The model does not take into account temperature fluctuations
that might work the opposite way and remove some conservatism from the calculations. Note that temperature
fluctuations are not included in any of the described methodologies in this report.

The simplistic internally developed assessment models are primarily based on CUI corrosion rate model in DNVGL-RP-
G101 for coated and uncoated surfaces combined with a consequence category. Based on this rate model, a predefined
inspection strategy is implemented. This methodology doesn’t normally take into account coating type and coating
condition, insulation systems, water wetting, design details or relevant workmanship.

4.4 Design requirements leading to increased CUI risk

The use of insulation in oil and gas facilities are very extensive. Calculations for the need of passive fire protection
(PFP) are often based on very conservative assumptions including fire load from a jet fire.

The requirements with respect to passive fire protection, ref §29 in the facilities regulations, often lead to increased risk
of breach of the primary barrier (containment) to fulfil the requirements for the secondary barrier (escalation and
evacuation). Also, the effect of deluge is not included in such calculations. The different operators have different ways of
calculating the need for PFP.

There might be an unreleased potential in reducing the overall risk by better balancing the CUI risk with the PFP
requirements and thereby reducing the need for insulation. A cross-discipline assessment of the actual need for
insulation is recommended.

Regulations also sets requirements to testing fire water systems, which often leads to extensive water wetting of the
process plant with sea water, and thereby creating a more corrosive environment than during normal operation without
testing the deluge system.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 9

4.5 CUI historical data
In general, there are limited data collected and made available to extract historical information regarding CUI rates or
failure distribution. A study presented at EuroCorr in 2019 /8/ published the first marine upstream onshore plant CUI
dataset for carbon steel piping and compared it with other existing datasets and API 581 prediction.

Figure 2 Comparison of CUI corrosion rate versus temperature from three plant carbon steel datasets./8/

The paper emphasised that:

• Estimated corrosion rates are not the best way to present plant data due to assumptions of coating life and
water ingress – the best way to report CUI data is metal loss versus age with first cycle inspection after
extended periods from a large sample population particularly relevant. Failures versus age graphs are also
useful.

• API581 adjustment factor metadata was not supported by these plant datasets. In Figure 4-2 API 581
underpredicted marine plant CUI at ~100°C, the API 581 adjustment factors not matching the majority data
points.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 10

Figure 3 Comparison of marine wall loss versus age with API 581 prediction./8/

The comparison study with API 581 and measured wall loss shows that API 581 underpredict the corrosion rate. This
result is only based on limited data and will in such a context not be conclusive.

The same study /8/ also reviewed in total 10 guidelines and 8 in-house methods for CUI management and revealed
several gaps and inconsistencies between the methods. In conclusion, it is stated that existing guidance is based on
very limited actual plant data, with each dataset bringing valuable knowledge, but limited by lack of consistency of
reporting and relevant metadata; more recent papers, though not all publishing raw data, reporting learning that
optimizes timing of inspection and/or improved the hit rate of % inspection.

The in-house methods were reported as very simplistic.

Note that the study is not based on assets from the Norwegian Continental Shelf (NCS).

5 EVALUATION OF DIFFERENT STRATEGIES AND GUIDELINES

5.1 Guidelines
Based on the publicly available guidelines that have been reviewed, there are several similarities and none of the
documents contradict the others. The main differences relate to the level of details and advice. Several documents are
more descriptive (eg. API 583, NACE SP0198 and EFC no. 55) with respect to application and limitations rather than

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 11

explicitly stating that one way is better than the other. This should be seen in context with a wide range of specific
situations related to e.g. age, workmanship or local weather conditions to be considered and caution should be used to
not rule out solutions for specific cases.

When describing CUI, challenges and possible solutions to CUI, the EFC no. 55 document is the most extensive and
most relevant document. The document covers and exceeds all elements described in API 583 and NACE SP0198. The
EFC. No 55 document is descriptive with respect to the risk management process but does not cover the topic in a
sufficient way. The document describes a work process for risk assessment without describing a methodology as such.
Risk mitigating effect and risk update is not included in the document.

For risk management of CUI, the DNVGL-RP-G109 will give a good recipe for both risk assessment, risk mitigation and
risk update. The DNVGL-RP-G109 is based on a semi-qualitative approach to risk. For a quantitative method the API
581 should be considered.

NDT to detect CUI is described in several documents including the EFC no. 55. However, the only document that
actually evaluates the different methods is the HOIS-G-023 document and it represents the most updated information
related to NDT methods for CUI.

For a practical implementation of a CUI management system, it is also recommended to consider all the different
practical advices that can be found in the Annex I12 in the Corrosion Management guideline from the Energy Institute.

5.2 Industry approach and strategies

The CUI challenge varies for the different operators depending on the design and age of the plant. For process plants
designed with an extensive use of CRA materials, the CUI challenge will be reduced but awareness should still be
raised for 316 material and carbon steel bolts in CRA systems.

For systems primarily designed in carbon steel, the CUI challenge might be seen after some years in operation, and for
some systems with poor design or poor workmanship, CUI could be a challenge already in the first years of operation.

The industry approach, as seen through this study reflects the different designs, where the simplistic internally
developed methodology is suitable for a company with one installation with extensive use of CRA materials, and the
complex internally developed methodology using many parameters is suitable for an operator with multiple types of
design with many years in operation.

For old systems primarily designed in carbon steel material, a more thorough approach should be used for CUI risk
management or just be replaced by extensive inspection and refurbishment. This is not the case for all operators.

The use of DNVGL-RP-G101, rev. 2017, which describes a CUI corrosion rate based on temperature combined with a
simplified model for coating degradation, might give too low corrosion rates, and thereby a too low probability of failure,
followed by an underestimated risk.

5.3 Implementation
The implementation of a CUI management strategy will require continuous attention and updates of the present risk.
Such updates need to reflect both the type of mitigation performed, as well as the extent of such mitigation. The
continuous awareness will lead to a dynamic risk management approach based on present conditions, opposed to a
static mitigation program based on fixed intervals.

Such a recommended dynamic approach to risk management is not implemented by all operators and relative static
programs with predefined inspection content are used by several operators.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 12

Successful implementation will also require an organisational setup to enable efficient communication and including
involvement of all relevant disciplines. One operator facilitates the communication by arranging weekly meetings
involving the relevant disciplines to address CUI specifically, while other operators struggle to have efficient
communication with relevant disciplines. The use subcontractors in this setting might also reduce the effectiveness of a
good CUI management implementation dependent on the possibility to achieve good communication lines with all
involved stakeholders.

6 CONTINUOUS IMPROVEMENT AND LESSONS LEARNED

The operators from the oil and gas industry in Norway have a CUI forum that is arranged by DNV twice a year for
common sharing of knowledge and experiences. This forum helps the industry at large to become smarter and make
better decisions related to CUI.

6.1 Trends
The reported hydrocarbon leaks due to CUI ref Figure 1 has for the last 10 years (2011 – 2020) been on an average of 3
incidents on an annual basis without any specific positive or negative trend. Reported leaks before 2008 indicate an
average of 1 incident annually. This increase over the last years might be caused by older assets more exposed to CUI
but it can also be a result of increased awareness and better reporting routines with respect to root cause of the leak.

In the later years the material selection for static process systems on new builds has reflected more use of corrosion
resistant alloys, which in general reduce the CUI potential. However, CRA systems often are still designed with the use
of HDG carbon steel bolts and nuts, which will remain a CUI threat. The selection of coating systems has also been
more focused on proper CUI protection and thermal spray aluminium (TSA) is also used more extensively. TSA has
proven to provide significantly longer protection towards CUI provided it is applied correctly.

Over the last years new insulation systems based on aerogel have been qualified for piping systems. These products
claim to be hydrophobic, i.e. water repellent and they also require less volume to ensure proper insulation. Up to
recently, aerogel products have only been used by a few operators on NCS.

One operator in Norway has increased the CUI focus and has for many years performed extensive testing and
qualification of coating and insulation systems. This testing has been beneficial to the entire industry as it has been in
addition to the in-house testing performed by coating and insulation manufacturers.

The use of insulation spacers between pipe and insulation to improve water drainage has now become a well proven
technique. It was introduced as early as 2007 in Norway and has shown good results in terms of reduced water wetting
of the insulation.

Moisture monitoring in the insulation is being explored by several operating companies and several suppliers can deliver
different solutions. Such service solutions will require ATEX certified, low-cost sensors with long battery lifetime and an
efficient installation method. Furthermore, systems for data gathering and decision support need to be in place. This
monitoring technology will most probably be implemented for critical systems for several operators in the years to come.

Within the field of NDT for detection of CUI there has been many recent developments but most of the new promising
techniques will still only detect larger defects. In this context the work done by HOIS to quantify the ability to detect CUI
is very valuable and enables the industry to make better decisions regarding use of NDT methods. The most promising
method “Open Vision” is still not allowed in Norway due to radiation regulations.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 13

The competitive situation and cost focus by operators with respect insulation and coating discipline has lead to low rates
and poor recruitment to these trades, followed by increased use of foreign workers. The required trade certificates might
not always be in place and personnel are often on short-term assignments. These elements can all have a negative
impact on the workmanship and thereby the end quality of the final insulation and coating system.

6.2 Best Available Technology

For new builds, the best design for CUI will include the use of thermal spray aluminium (TSA) and insulation products
that do not retain water or enable efficient water drainage.

For assets in operation, it will remain important to manage the embedded risk of CUI through good dynamic risk
management, knowledge of the asset condition, and well-planned mitigation. This will include the use of:

• Well-implemented CUI risk management guideline

• Inspection and monitoring that confirm or adjust risk by increased knowledge and reduces uncertainty

• Mitigation plans that reduce the risk to an acceptable level

The current most relevant CUI guidelines are described in this document, all available for the industry. The introduction
of a barrier approach introduced in DNVGL-RP-G109 is well received by operators in Norway.

Advanced NDT methods developed lately to identify CUI are still not efficient enough, with the potential exception of the
“Open Vision” system. This system has shown good results but is not yet allowed in Norway due to radiation safety
regulations. Inspection methods based on guided wave ultrasonic and pulsed eddy current show potential but will still
need to achieve a better probability of detection for small, deep CUI flaws. There are ongoing processes to enable such
technology for the Norwegian market.

Moist sensors are available for the industry and will give better information of the water wetting situation in the insulation
and thereby help operators to better assess the risk of CUI.

For the mitigation of CUI several new insulation and coating systems has been qualified and taken into use. Company
strategies should include the use of new systems for better CUI management and not just use the same insulation and
coating systems that has been used earlier.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 14

6.3 Operator experiences

Operators consider CUI as the “biggest threat to the mechanical integrity of oil and gas industry facilities”. Most
operators consider CUI as a complex degradation where it is difficult to predict location of CUI attacks.

The general work process for the operators, independent of their CUI risk assessment model consist of the following
steps:

1. Gather data for risk assessment

2. Perform assessment
3. Perform general visual or close visual
inspection in field to detail the inspection
program
4. Perform detailed inspection, often a
combination of close visual inspection of
external cladding and close visual inspection
of piping and vessels after removal of
insulation
5. Extend removal of insulation if any findings
6. Report and update inspection program

It is considered important to use local facility knowledge in combination with analysis to identify potential location for
CUI. Experience from several operators indicate that the following object should be given high focus:

• Deck penetrations
• Carbon steel bolts in CRA systems
• Nozzles of pressure vessels
• Corrosion Under Pipe Supports (CUPS)

Some of the operators also have specific CUI programs to address these mentioned geometric features.

The implementation of DNVGL-RP-G109 is found to be easy and will give a good baseline for CUI-RBI. However, it is
also identified challenges to adapt historical data when using DNVGL-RP-G109. For new inspections reporting
parameters can be aligned with requirements for information in mentioned guideline. This will lead to a better data
driven decision process in the way forward. It is reported that 50% of the operators actively use historical data in their
CUI management. There is a large variation on which parameters the different operators use as input for CUI probability
assessment, but all or most operator use material data, temperature, coating type and inspection results as input

Most operators express that they use field survey by Field Engineers and Inspectors to identify the locations most
exposed to CUI and some operators are also having good experience with use of digital twin in planning phase.

Several of the operators are seeking to optimize the use of insulation. Optimize, in this context means to reassess the
actual need for insulation and potentially remove or partly remove insulation to eliminate CUI risk. In general, the

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 15

operators consider the insulation as wet until the opposite is confirmed. Thermographic inspections are also used to
identify wet areas of the insulation.
Workmanship within insulation and cladding from project phase has strong impact on condition during late life operation

One operator experience that a weekly cross-discipline forum and close cooperation with insulation inspectors give a
improved CUI management while others struggle to achieve a sufficient cross discipline internal cooperation.

Examples from inspection campaigns (radiographic and close visual inspection) for CUI indicate very high degree of
flaws in inspection campaigns, above 25% with one sever finding of 50% wall reduction. In general, the identified flaws
vary in size and few need to be mechanical repaired. Based on the feed-back from the operators the most common
consequence of CUI is coating repair.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 16

7 CONCLUSION
Corrosion under insulation is and will continue to be a major threat to the technical integrity of oil and gas assets. The
embedded risk of a potential hydrocarbon leak is managed in different ways by the operators. The public available
documents give guidance on the implementation of CUI management systems and some operators follow such
guidelines while other operators have developed in-house methodologies. Some of the public available guidelines and
some of the in-house developed documents might be simplistic and can underestimate the risk.

Operators differentiate significantly in the use of input data in the assessment of the CUI threat, and it is assumed that
more extensive use of data will promote more correct risk evaluations.

In general, the oil and gas industry give high focus to the CUI threat and there are several initiatives amongst the
operators that has led to successful implementation of better technical solutions.

Several improvements areas are identified and many of these can be solved through industry cooperation.

• Different approaches with respect to design of passive fire protection. A best practice could be identified and
should be shared with the industry

• There is no existence of good historical data sets for CUI damage. A joint effort might result in better data-
driven decisions and adjustments to current strategies

• Qualification of new technology could be organised with the involvement of multiple operators and sharing of
results to the benefit of the industry

• Facilities regulations to be updated with reference to EFC 55

• Activities regulation to be updated with reference to DNVGL-RP-G109

• Existing reference in activities regulations to DNVGL-RP-G101 to be updated with reference to DNVGL-RP-

G101 revision 2021.

Awareness should be raised for the following:

• A comparison study with API 581 and measured wall loss shows that API 581 underpredict the corrosion rate.

• The use of DNVGL-RP-G101, rev. 2017, which describes a CUI corrosion rate based on temperature
combined with a simplified model for coating degradation, might give too low corrosion rates, and thereby a too
low probability of failure, followed by an underestimated risk.

• Simplistic CUI assessment models with the use of few input parameters might underpredict the probability of
failure

• Different approaches and methodologies might cause different gaps and key learning across companies might
become less relevant

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 17

8 REFERENCES
/1/ API Recommended Practice 581, “Risk-Based Inspection Methodology”. Third edition April 2016, Addendum
1, 2019 and Addendum 2, 2020

/2/ API Recommended Practice 583 “Corrosion Under Insulation and Fireproofing” First addition May 2014.

/3/ DNVGL-RP-G109 “Risk Based Management of Corrosion Under Insulation” rev. December 2019.

/4/ NACE SP0198 “Control of Corrosion Under Thermal Insulation and Fireproofing Materials – A systematic
Approach” rev. July 2017

/5/ European Federation of Corrosion (EFC) Publication number 55 “Corrosion-Under-Insulation (CUI)

Guidelines” revised edition 2017

/6/ Energy Institute “Guidance for corrosion management in oil and gas production and processing” second
edition March 2019

/7/ HOIS-G-023, HOIS/OGTC Guideline for in-situ inspection of corrosion under insulation (CUI), Issue 1, March
2020

/8/ Eurocorr2019-226986 “Using industry data to compare performance of different risk-based methods for
management of corrosion under insulation” PhD C. Watt, PhD Chi-Ming Lee, S. Paterson, A. Jopen.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com Page 18

APPENDIX A
Work shop

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-1

 A1. WORKSHOP AGENDA
The agenda for the workshop was:

• Introduction and welcome

• Background

• Presenting CUI questionnaire feed-back

• Technical topics and discussions

o Main challenge

o Work process

o Probability assessments

o Findings and management of findings

o Reporting and learning

• Success and Challenges

A2. PARTICIPANTS
The following companies were represented during the workshop:

• AkerBP

• Alterra Infrastructure

• ConocoPhillips

• Equinor

• Gassco

• Lundin

• Norske Shell

• Vår Energi

Organizers:

• Petroleum Safety Authority

• DNV

A3. BACKGROUND
The objective and background for the workshop were presented by Petroleum Safety Authority.

Main objective was to identify how operators and rig owners on the Norwegian Continental Shelf (NCC) is working to
manage the CUI threat with regard to pressure containing equipment and how effective the different practices are
related to:

• Understand the application and limitations of the various approaches, including the use of standards

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-2

 • Understand how and to what extent the methodology is implemented and how the relation is between
theoretical approach and practical execution

• Discuss potential improvement areas and how to learn from each other's experiences

A4. REGULATORY REQUIREMENTS

The following regulations are considered as most relevant related to CUI management:

• The Facilities regulations § 12 Materials

• The Activities regulations § 45 Maintenance

• The Management regulations § 23 Continuous Improvement

• The Framework regulations § 24 Use of recognised standards

A5. FEEDBACK QUESTIONAIRE

Prior to the workshop all invited operators and rig owners were asked to give feedback on the following:

• What is your companies main challenge(s) with respect to avoiding a major accident due to corrosion under
insulation? See A5.1

• How do you detect CUI? See A5.2

• What is the most common immediate consequence of CUI? See A5.3

• How is your assessment model for CUI built? See A5.4

• Parameters used in CUI probability assessment. See A5.5

Feedback was received from 7 operator. In the following presentation of the results, all answers are anonymised and
are presented as “number of answers” per given category.

Note that difference in scale, age and design of the various operated assets may influence the answers given.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-3

 A5.1. Question: Main challenges

Figure 4 Summary of results Question 1

Figure 4 Summarises the answers to question 1. The following main take-away may be drawn from the answers and
discussions during the workshop:

• All operators states that there is enough relevant competence available

• Most operators find it difficult to predict where CUI will occur

• More than 50% of the operators indicates that there is a lack of suitable NDT methods available

• More than 50% of the operators points out that there are major costs connected to CUI management

General comments
Some of the predefined statements were somewhat difficult to interpret and answers may be influenced of each
companies’ interpretation.
Most operators gave separate comments which all can be summarised as highlighting the complexity of the CUI threat,
the challenges related to scope size, resource requirements and the varying quality of workmanship related to coating
and insulation.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-4

 A5.2. Question: Detection of CUI

Figure 5 Summary of results Question 2

Figure 5 summarises the answers to question 2. The following main take-away may be drawn from the answers and
discussions during the workshop:

• Most operators have developed a dedicated strategy for CUI inspection activities

• Most CUI indications are reported through inspection or fabric maintenance activities

• Some answers indicates that certain operators have assets with a robust design, e.g. large extent of corrosion
resistant alloys and adequate surface protection, minimizing CUI as a threat

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-5

 A5.3. Question: Most common immediate consequences

Figure 6 Summary of results Question 3

Figure 6 summarises the answers to question 3. The following main take-away may be drawn from the answers and
discussions during the workshop:

• The most common consequence of CUI is repair of surface protection or insulation system through fabric
maintenance

• Temporary repairs are also used, mostly in form of clamping or composite repairs. Some operators are
restricted by company specific procedures to limit use of temporary repairs.

• Mechanical repairs are mainly often or rarely a consequence of CUI

• Leakages, unintentional shut-down and personnel exposure to HC/hazardous substances are rarely an
immediate consequence of CUI

General comments
Some of the predefined statements were somewhat difficult to interpret and answers may be influenced of each
companies’ interpretation.
CUI incidents reported to the Petroleum Safety Authority is limited to leakages with leak rates larger than 0,1 kg/s.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-6

 A5.4. Question: Assessment model

Figure 7 Summary of results Question 4

Figure 7 summarises the answers to question 4. The following main take-away may be drawn from the answers and
discussions during the workshop:

• All operators have a specific model for CUI, but only three operators indicate the model used follows a
standard or RP. DNV-RP-G109 is the only standard/RP used

• Most operators express that they use field survey by Field Engineers and Inspectors to identify the locations
most exposed to CUI.

• About 50% of the operators replies that they use historical data in updating their CUI model / RBI.

• Comments provided in the questionnaire indicates, that for some operators, CUI assessment is included in the
general RBI assessment

General comments
Some of the predefined statements were somewhat difficult to interpret and answers may be influenced of each
companies’ interpretation, mainly related to the interpretation of “Point based model” and the difference between
“Quantitative”, “Semi-Quantitative” and “Qualitative” models.

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-7

 A5.5. Question: Parameters for probability assessment

Figure 8 Overview of results Question 5

Figure 9 Distribution of answers Question 5

Figure 8 and Figure 9 summarises the answers to question 5. The following main take-away may be drawn from the
answers and discussions during the workshop:

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-8

 • There is a large variation on which parameters the different operators use as input for CUI probability
assessment, but all or most operator use material data, temperature, coating type and inspection results as
input

• Figure indicates that 4 operators have a high number of input parameters included in their probability
assessment, and 3 of them are utilising DNVGL-RP-G109 as basis for their assessment model while 1 has an
inhouse developed model

• The discussions during the workshop pointed out the difference in CUI data availability based on insulation age
and design, which limits the possibility to utilize several of the mentioned input data sources

DNV – Report No. 2021-4107, Rev. 02 – www.dnv.com A-9

About DNV
DNV is the independent expert in risk management and assurance, operating in more than 100 countries. Through its
broad experience and deep expertise DNV advances safety and sustainable performance, sets industry benchmarks,
and inspires and invents solutions.

Whether assessing a new ship design, optimizing the performance of a wind farm, analyzing sensor data from a gas
pipeline or certifying a food company’s supply chain, DNV enables its customers and their stakeholders to make critical
decisions with confidence.

Driven by its purpose, to safeguard life, property, and the environment, DNV helps tackle the challenges and global
transformations facing its customers and the world today and is a trusted voice for many of the world’s most successful
and forward-thinking companies.