Pipeline Pigging & Integrity Management Conference
plus Training Courses and Exhibition
George R. Brown Convention Center and the Marriott Marquis Hotel

February 18-19 February 20-22 February 19-21


Conference Program

Program Advisory Committee

  • Dr Tom Bubenik – DNV GL
  • Dr Andrew Cosham – Ninth Planet Engineering Limited
  • Matt Hastings – Williams
  • Everett Johnson – Marathon
  • Dr Keith Leewis – L & A, Inc.
  • BJ Lowe – Clarion Technical Conferences
  • Jim Marr – Marr Associates Pipeline Integrity Ltd.
  • Garry Matocha – Enbridge
  • Bryan Melan - Tide Water Integrity
  • John O’Brien – Chevron Corporation
  • Steve Rapp – Spectra
  • Jerry Rau – RCP
  • Terry Shamblin – EQT Midstream
  • John Tiratsoo – Tiratsoo Technical
  • Nelson Tonui – Trans Mountain Corp.
  • George Williamson – BP


Wednesday 20 February


1.0 Plenary opening session


Opening remarks


[1] Managing complexity through collaboration will need the industry to move from a proactive to a predictive mindset, by Chris Yoxall, Rosen USA, Houston, TX, USA


[2] Achieving and demonstrating pipeline engineering capability – the role of competency standards, and their use for qualifications and registration, by Chris Harvey, Chris Harvey Consulting, Australia


[3] Assessing the competence of staff, by Michelle Unger, Rosen Group, Stans, Switzerland, and Dr Phil Hopkins, Phil Hopkins Ltd, Whitley Bay, UK




[4] A regulator’s perspective on pipeline integrity concerns, by Iain Colquhoun, National Energy Board, Calgary, AB, Canada


[5] Achieving a N American record for longest intelligent inspection of a natural gas pipeline, by Sheshi Epur and Aaron Schartner, TransCanada Pipelines, Calgary, AB, Canada, and Frank Sander, BHGE, Calgary, AlB, Canada (invited)


[6] Are you safer than you were 15 years ago?, by Joel Anderson, Enable Midstream, Oklahoma City, OK, USA


YPP Awards




2.1 ILI applications

3.1 Risk assessment & management

4.1 Evaluating dents reported by ILI for response and remediation


[7] MFL high-temperature solution, by Guenter Sundag, Thomas Stubbe, and Corey Richards, Rosen USA, Houston, TX, USA



[28] Dissecting new PHMSA risk assessment guidelines, by Sevinc Yeliz Cevik and William Kent Muhlbauer, WKM Consultancy LLC, Austin, TX, USA



Chair: Sergio Limon

     Elevara Partners, Salt Lake

     City, UT, USA


Engineering methods for evaluating and ranking dents reported by ILI tools

       Rhett Dotson, Rosen USA,

       Houston, TX, USA



2.1 ILI applications (cont’d)

3.1 Risk assessment & management (cont’d)

4.1 Evaluating dents (cont’d)


[8] Transitioning from hydrostatic testing to in-line inspection for pipelines with challenging seam welds, by J Bruce Nestleroth, Kiefner & Associates, Columbus, OH, USA, Matthew S Krieg, Marathon Pipe Line LLC, USA, Thomas Hennig, NDT Global, USA, and Harvey Haines, Applus RTD Technology Center, USA

[29] Quantitative risk assessment following an ILI survey (ILI-based risk assessment), by Jane Dawson, Ian Murray, Inessa Yablonskikh, and Thomas Hoffmann, BHGE, Cramlington, UK

Repair systems for dents

     Dr Chris Alexander

     ADV Integrity, Inc., Magnolia, TX,




What we have learned from decades of experimental research on dent behavior

     Aaron Dinovitzer

     BMT Fleet Technology, Ottawa, ON,





[9] The role of ILI for MAOP verification, by Simon Slater, Rosen USA, Huston, TX, USA

[30] Knowledge risk management, by John Godfree and Tara McMahan, DNV GL, Dublin, OH, USA






5.1 ILI analysis

3.2 Risk assessment & management (cont’d)

4.2 Evaluating dents (cont’d)


[10] The art of looking: an in-line inspection perspective, by Dr Mike Kirkwood, Dane Burden, and Miguel Maldonado, T.D. Williamson, Tulsa, OK, USA

[31] Pipeline risk modeling: a comparative analysis of modeling techniques, by Andrew Kendrick and Robin Echols, Kendrick Consulting, Aurora, CO, USA

A gas operator perspective on managing dents

     Rick Wang

     TransCanada, Calgary, AB, Canada


A liquids operator perspective on managing dents

     Justin Hardraker

     Colonial Pipeline Co., Alpharetta,

     GA, USA



5.1 ILI analysis (cont’d)

3.2 Risk assessment & management (cont’d)

4.2 Evaluating dents (cont’d)


[11] Challenges associated with pit-to-pit matching (or how to know when corrosion is taking place), by Dr Thomas Bubenik, Steven Polasik, and Zach Booth, DNV GL, Dublin, OH, USA

[32] Risk-based evaluation of asset integrity projects, by David Mangold, TRC Companies, Columbus, OH, USA

API 1183 Upcoming dent assessment and management recommended practice

     Mark Piazza

     Colonial Pipeline Co, Alpharetta,

     GA, USA




End of day, Exhibition reception

Thursday 21 February


5.2 ILI analysis (cont’d)

6.1 Engineering assessment

7.1 Materials


[12] Analysis of factors which reduce MFL sizing accuracy of pinholes, by Joel Falk, Desjardins Integrity Ltd, Calgary, AB, Canada

[33] Realistic burst pressure predictions in pipelines with non-ideal crack profiles, by Dr Ted Anderson, TL Anderson Consulting, Longmont, CO, USA 



[50] Nondestructive testing of pipeline materials: analysis of chemical composition from metal filings, by Mary Louie and Dr Monty Liong Exponent, Menlo Park, CA, USA, Dr Peter Veloo, Exponent, Los Angeles, CA, USA, Bill Amend and Melissa Gould, DNV GL USA (Inc.), Dublin, OH, USA and Troy Rovella and Peter Martin, PG&E, Walnut Creek, CA, USA


[13] Advancing ILI technology and pipeline risk management through advanced analytics of big data, by Geoff Hurd, Stuart Clouston, Jeff Sutherland, and John Elliott, BHGE, Calgary, AB, Canada

[34] Evaluation of limitations and applicability of stress and strain concentration factors for use in engineering critical assessments of dents, by Shanshan Wu, Dr Thomas Bubenik, Joseph Bratton, and David Kemp, DNV GL USA (Inc.), Dublin, OH, USA

[51] Nondestructive classification of LF, HF, and HF-normalized electric-resistance-welded (ERW) longitudinal seams, by Steven Palkovic, Parth Patel, Soheil Safari Loaliyan, Mohammad Islam, and Dr Simon Bellemare, MMT, Cambridge, MA, USA


[14] Enhanced utilization of ILI inertial measurement data, by Dane Burden, T.D. Williamson, Salt Lake City, UT, USA

[35] Nondestructive examination protocols for MAOP verification of station pipe, by Simon Lockyer-Bratton, Dr Peter Veloo Exponent, Los Angeles, CA, Mary Louie, Exponent, Menlo Park, CA, USA, Mark Ryan, Michael Rosenfeld, Kiefner & Associates, Columbus, OH, USA, and Troy Rovella, PG&E, Walnut Creek, CA, USA

[52] Bayesian inference approach to establish sample size for material verification, by Troy Rovella, Peter Martin, Masoud Moghtaderi-Zadeh PG&E, Walnut Creek, CA, USA, Joel Anderson, Enable Midstream, Oklahoma City, OK, USA, Kofi Inkabi, Exponent, Oakland, CA, USA, Vyaas Gururajan, USC, Los Angeles, CA, USA and Dr Peter Veloo, Exponent, Los Angeles, CA, USA




5.3 ILI verification

6.2 Engineering assessment (cont’d)

8.1 Cracks & seam welds


[15] Validation of computed tomography technology for pipeline inspection, by Mark Piazza, Colonial Pipeline Co,

Alpharetta, GA, USA, Timothy Burns, Shell Pipeline Co, Houston, TX, USA, James Medford, Inspection Associates, Inc., Cypress, TX, USA, and

Taylor Shie, Shell Pipeline Co,

Houston, TX, USA

[36] Leveraging ILI data to support ancillary asset integrity tasks, by Lisa Barkdull and LeeAnn Escobar, Quest Integrity, Boulder, CO, USA

[53] Improved system for the detection, sizing and prioritization of seam weld corrosion, by Matthew Romney, T.D. Williamson, Salt Lake City, UT, USA and J. Bruce Nestleroth, Kiefner and Associates, Columbus, OH, USA


[16] Interaction rule guidance for corrosion features reported by ILI, by Lucinda Smart, Kiefner &Associates, Inc., Ames, IO, USA, Yanping Li, Enbridge, Edmonton, AB, Canada

J. Bruce Nestleroth, Kiefner & Associates, Inc. Columbus, OH, USA, and Suzanne Ward, Enbridge, Edmonton, AB, Canada

[37] Reliability-based criteria for corrosion assessment, by

Riski Adianto, Maher Nessim, Dongliang Lu, Shahani Kariyawasam, and Terry Huang, C-FER Technologies, Edmonton, AB, Canada

[54] High-resolution inspections for crack detection: the next level of accuracy, by Rogelio Jesus Guajardo Rodriguez and Thomas Hennig, NDT Global GmbH & Co KG, Stutensee, Germany


[17] [INVITED] Development of an industry test facility and qualification process for ILI technology evaluation and enhancements – performance evaluation phase, by

Pablo Cazenave and Ming Gao, Blade Energy Partners, Houston, TX, USA and Hans Deeb and   Sean Black, PRCI, Houston, TX, USA

[38] Equivalent load fatigue – An efficient modification to the familiar Paris equation, by Stephen Wood and Alfonso Garcia, Enbridge, Edmonton, AB, Canada

[55] Gap analysis of crack integrity management for pipelines, by Dr Jing Ma, Kiefner and Associates, Columbus, OH, USA


[18] Location and validation of metal loss defects identified by ILI, by Dr Michael Beller, Rosen, Lingen, Germany, G.Reid,  Sonomatic, and Dr Roger King, International Corrosion Services, Manchester, UK

[39] Technical background of a simplified process for conducting ECA of indicated pipeline indentations with metal loss, by Fan Zhang, Michael J Rosenfield, Kiefner and Associates, Columbus, OH, USA

[56] Common pitfalls to avoid when managing seam-weld integrity, by Michael Turnquist, Quest Integrity, Boulder, CO, USA




5.4 ILI verification (cont’d)

9.1 Risk-based inspection

8.2 Cracks & seam welds (cont’d)


[19] Total quality API 1163 approach to ILI verification, by Chad Haegelin and Joel Lindstrom, Integrity Solutions Ltd, San Antonio, TX, USA

[40] Risk-based approach to inspection interval optimization, by David Joyal, Jana Corporation, Aurora, ON, Canada

[57] Investigation of crack assessment parameters for a hypothetical pipeline, by Tara McMahan, Eric Graft, and Dr Thomas Bubenik, DNV GL, Dublin, OH, USA


[20] Run comparison as a solution to incomplete ILI data and as an alternative to re-inspection of a challenging pipeline, by Kai Xin Toh, Quest Integrity, Cheras, Malaysia


10.1 Hydrostatic testing

[58] Screening for long seam anomalies in ERW pipe using ultrasonic crack ILI data: a method for pipeline operators to unlock the value of their data, by Bernardo Cuervo and Mark McQueen, G2 Integrated Solutions, Houston, TX, USA

[41] A practicum on pressure testing – compilation of best practices, by Sheri Baucom and Jerry Rau, RCP, Houston, TX, USA




Coffee (Marriott)


11.1 Repair

12.1 Leak Detection I

13.1 Mechanical damage


[21] Composite repairs – what does “permanent” mean?, by Casey Whalen, Milliken Pipe Wrap, Houston, TX, USA

[42] The challenge of implementing and maintaining CPM leak detection on gathering networks, by Peter Han, Atmos International, San Antonio, TX, USA

[59] Assessment of mechanical damage within dented pipe using multi-data ILI technology, by Luis Torres, Kaitlyn Korol, and Neil Hodson, Enbridge Pipelines, Edmonton, AB, Canada


[22] Full-scale finite element analysis and field success prove composite reinforcement is a viable repair for girth weld joint defects on vintage pipelines, by Buddy Powers, Tim Mally, and Mahdi Kiani, ClockSpring, Houston, TX, USA

[43] Pipeline leak detection using tracer compounds and sledding techniques, by Ian Harris, Praxair Services, Inc., Tucson, AZ, YSA

[60] An evaluation of instrumented indentation testing to estimate yield and tensile strength, by Dr Nicoli M Ames, Exponent, Denver, CO, USA, Mary Louie, Exponent, Menlo Park, CA, USA Dr Jeffrey A Kornuta, Exponent, Houston, TX, USA, Dr Peter Veloo, Exponent, Los Angeles, CA, USA, Troy Rovella, and Peter Martin PG&E, Walnut Creek, CA, USA


End of day

Friday 22 February


11.2 Repair (cont’d)

14.1 Data management

13.2 Mechanical damage (cont’d)


[23] Evaluating the performance of composite systems for reinforcing non-leaking crack-like defects in transmission pipelines, by Colton Sheets and Chantz Denowh, Stress Engineering Services, Houston, TX, USA

[44] Industrial revolution 4.0: disruptive to pipeline integrity management?, by Mohd Nazmi bin Mohd Ali Napiah, Petronas, Kula Lumpur, Malaysia

[61] Compositing multi-technology ILI surveys for the integrity management of mechanical damage, by Luis Torres, Catherine Rieck, and Collin Taylor, Enbridge Pipelines, Edmonton, AB, Canada 


[24] Predictive modeling for shrink sleeve failure using machine learning, by Matthew Brown, Lake Superior Consulting, Duluth, MN, USA

[45] The good the bad and the ugly: categorizing pipelines using big data techniques, by Roland Palmer-Jones and Michael Smith, Rosen UK, Newcastle upon Tyne, UK


[62] Gouge detection on dents below 1% depth with multiple data set technologies on an ILI tool, by Timothy Goller and Adrian Belanger, T.D. Williamson, Salt Lake City, UT, USA


[25] Steel sleeves: a new look at a widely-used repair method, by Dr Chris Alexander, ADV Integrity, Inc., Magnolia, TX, USA, Tommy Precht, Allan Edwards, Lake Charles, LA, USA, and Chip Edwards, Allan Edwards, Tulsa, OK, USA

[46] Enabling the digital pipeline, by Steve Banks, i2i Pipelines, Manchester, UK

[63] Detailed dent assessment: avoiding the pitfalls, by Aaron Lockey, Tim Turner, and Susannah Turner, Highgrade Associates, UK




12.2 Leak detection II

14.2 Data management (cont’d)

15.1 SCC


[26] Leak detection and prevention using free-floating in-line sensors, by John van Pol, INGU Solutions, Calgary, AB, USA

[47] The challenges of keeping integrity management systems relevant, by Sonny Llave, Pradeep Dhoorjaty, and Danny Golczynski, Wood Group, Houston, TX, USA

[64] The detection and sizing of circumferentially oriented stress corrosion cracking using axially oriented magnetic flux leakage inspection, by Ron Thompson, Ray Gardner, Katrina Dwyer, and James Hare, Novitech, Inc., Vaughan, ON, Canada


12.2 Leak detection II (cont’d)

14.2 Data management (cont’d)

15.1 SCC (cont’d)


[27] Development of a framework for evaluating and verifying external leak detection systems for pipelines, by Mathew Bussiere, C-FER Technologies (1999) Inc., Edmonton, AB, Canada

[48] Swimming in the data lake — efforts towards efficient management and processing of pipeline data for integrity management, by Michael Smith, Johnathan Martin, and Marcillo Torres, Rosen USA, Houston, TX, USA

[65] An approach for evaluating the susceptibility of a pipeline to circumferential SCC, by Jane Dawson and Ian Murray, BHGE, Cramlington, UK



[49] Leveraging machine learning techniques to improve corrosion risk prediction in pipelines, by Ramnath Easwar, Abhinav Priyadarshi, Andreas Gaarder, Jay Karen William, and Vijaytha Balaji, Wood Group, Houston, TX, USA

[66] Full-scale testing of SCC in high frequency-ERW pipe with comparisons of inspection techniques to actual flaw measurements, by Colton Sheets, Stress Engineering Services, Houston, TX, USA


End of conference





 Organized by:

Clarion Technical Conferences     Tiratsoo Technical

 Platinum Elite Sponsor


 Platinum Sponsors

Intero Integrity Services     Enduro

Gold Sponsor


 Silver Sponsors

   N-Spec    Q-Inline  TD Williamson       Circor Energy - Pipeline Engineering.png

Supported by:

Pipelines International     Journal of Pipeline Engineering     Pipeline & Gas Journal     the In Line Inspection Association     Oil & Gas Journal     PRCI     PPSA     North American Pipelines Inspectioneering Latincorr   


please check back

There is increasing pressure on the pipeline industry to be able to demonstrate that its asset management and engineering capability management are at a satisfactory level. This is needed to give policymakers, regulators and industry stakeholders confidence about the safety and environmental sustainability of petroleum pipelines.
Regulators, in particular, are seeking assurance from pipeline owner/operators that they have capable pipeline engineers designing, constructing, operating and maintaining petroleum pipelines. At present, there are no generally accepted approaches to recognizing and developing pipeline engineering capability.
The paper will discuss three levels of capability recognition as: (1) registration – as pipeline engineers (not just in mechanical, civil or chemical engineers (overall standing level) – (2) qualification (sub-discipline/job level) and (3) competency (task level). The most granular and useful of these is competency. This is because it is at the level that is most immediate: the task at hand.
Competency, the combination of knowledge and experience that leads to expertise, is increasingly seen as the best practice basis for learning, particularly for professionals. Significantly, once competencies have been defined in competency standards, they can become the building blocks used to define the requirements for both registration and qualification.
The Australian Pipelines and Gas Association (APGA) has developed a comprehensive competency system for both onshore and offshore sectors.  There are 226 onshore competency standards and 57 offshore competency standards describing, in a succinct format, what is required to be competent.
The succinct format of the competency standards avoids the pitfalls of many other systems of competency description, providing enough information to be clear about what is required without unnecessary complexity. In addition to the detailed competency standards, the competency system has tools, resources and a progressive rating scale that make competency standards accessible and easily used.  The competency system is characterized by such flexibility that to date APGA has identified 15 applications, all of which will add value to engineers and companies that employ them.
The paper will explain, in detail, APGA’s Pipeline Engineer Competency System, how it works and how it can provide the building blocks for a wide range of tasks that support the training, development and recognition of pipeline engineers’ capabilities, including defining the requirements for registration and qualification.
The paper will provide case studies, based around the APGA Competency System, showing how it can be used to create requirements for qualifications and registration and be used to design in-house training and development plans.

Pipeline standards and regulations require all staff working on pipelines to be both ‘competent’ and ‘qualified’, but there is little guidance on how organizations can demonstrate their staff are competent and qualified.
Any demonstration of competence must be objective; therefore, an individual’s curriculum vitae (CV) is not a convincing demonstration, as a CV is usually highly subjective and has no independent verification. Many organizations rely on individuals performing self-assessments,or rely on line management ensuring staff are competent at the tasks they perform, but usually there is no structured process for either specifying or assessing the competency. This informality is not a convincing demonstration.
Competencies need to be formally assessed (verified), to satisfy industry standards and regulations. The competencies are assessed using a ‘competency standard’, which details all the required skills, knowledge, and experience expected for that competency.
The assessment could be an examination, an interview, or a performance review. ‘Qualified’ means ‘an individual that has been evaluated’; therefore, individuals passing the assessment are qualified in that competency. An individual claiming a competency, without a verification through an assessment, is not qualified in that competency.
This paper gives an outline of competence assessments, and a process for competence assessment.

There is constant pressure on regulators to assure that pipelines do not harm people and the environment. The regulator holds the pipeline company accountable to have the appropriate integrity management plans in place to effectively manage the integrity of their pipeline system. However, since regulators do not operate the pipeline system and as a consequence may not have complete information about the system, this poses challenges to the regulator when assessing the safety of the pipeline.   The paper will examine pipeline integrity and ILI issues that regulators deal with that pose challenges to assuring pipelines are safe.  These issues include the competency of integrity and ILI personnel, lack of information on the pipeline system attributes, appropriateness of the integrity assessment methods, material quality issues, ILI tool and NDE uncertainty, and appropriate threat identification and meaningful risk assessments.  Similar to regulated companies the regulator is constantly dealing with the unknowns associated with the issues and must ultimately determine when the company has sufficiently demonstrated that the pipeline is safe.   As the Russian proverb states “Doveryai no proveryai” (“Trust but verify”) is a key component of a regulators oversight of companies integrity management programs.  The paper will conclude by discussing potential actions that can be taken by companies to provide the confidence to regulators that pipeline integrity is being effectively implemented and that the pipelines are safe.

TransCanada was faced with a significant challenge to inspect a 941 km NPS 48 pipeline. The options for the inline inspection (ILI) were multiple segments which would cause an increased cost with new pigging facilities required and a delay to the ILI schedule, or attempt to pig the longest natural gas pipeline section in North America. The extraordinary proposal would require a Smart Approach to preparing the pipeline for ILI,and assessing its readiness.   The program required long distance, high speed Smart Tools and a massive 48” combination Magnetic Flux Leakage (MFL) tool capable of negotiating the 941km line from Burstall, Saskatchewan to Ile des Chenes, Manitoba, Canada. 
Given the alternative of the installation of 3 additional launcher and receiver stations and the risk to overall project schedule from extended inspection operations, TransCanada took the bold decision to clean and inspect the pipeline in a single pass. This option created a new set of challenges to guarantee ‘first run success’ in one of the harshest pigging environments and in a line where the cleanliness condition was unknown. In any ILI program an understanding of the pipelines internal condition, geometry and cleanliness  is critical to success, more so for a 941km pipeline In particular unknown debris deposits existing in the pipeline  could be collected with the highly aggressive MFL tool brushes, could easily and very quickly have led to very significant debris build up during inspection that at best would likely cause degraded data leading to an unwanted re-run and at worst, the possibility of a stuck pig and subsequent retrieval program. Assessing readiness and assuring success, the project utilized a Debris Mapping Tool to measure and quantify debris levels in the pipeline.  The insight gathered by the tool provided a baseline upon which the likelihood of success could be measured.
In July 2017, a successful Debris Mapping and MFL inspection was completed. This paper discusses the main program risks, mitigation steps taken over and above a standard ILI run. Key considerations and actions taken relating to additional engineering and tool modifications to various components of the inspection vehicle itself will be discussed.
Lastly, insight will be given into an extensive smart cleaning program using a combination of mechanical cleaning associated and debris level assessment, specifically designed and tailored for the project to ensure that the pipeline was both ready for ILI and that cleaning had reached optimum for ILI so that full, high quality MFL data would be collected in a single run.

With over a decade and a half since The Pipeline Safety Improvement Act of 2002 was passed and tens of millions of man-hours and billions of dollars spent in support of this effort, it is useful to look back and see if there has been a detectable improvement in safety.  Intuitively, the expectation is that measurable safety improvements would reflect in the long-term incident statistics. In 2006 the Government Accountability Office concluded; "As the program matures, PHMSA's performance measures should allow the agency to quantitatively demonstrate the program's impact on the safety of pipelines."  This is especially true now as operators have completed their baseline, reassessment and in some cases, are even into their second reassessment of pipelines.  To test if a measurable difference in safety has been made, a rigorous statistical methodology will be employed to determine if there has been a detectable shift in incident rates and severity of incidents since implementation of the rule.  The analysis will examine four incident metrics; fatalities, injuries, Serious incidents, and Significant incidents for the fifteen years prior and following the implementation of the Integrity Management rule.

This paper will outline the validation and testing of an MFL based technology capable of withstanding high temperatures over long exposure times. The focus of the validation testing program was to test existing components for failure and optimize as needed, while also testing tool design improvements intended to reduce the impact of high temperatures on the tool itself.
The pipeline discussed in this paper transports heated bitumen. We consider the following boundary conditions as the core basis of the test:
- Temperature of 140 degrees C
- 17 hours minimum inspection time
- Bitumen product
The paper focuses on the different project development and test phases derived from gaps between standard tool specifications and operational conditions.
The paper covers four test phases that have been achieved to date:
- Feasibility Study and Small-scale Tests
- Production and Assembly Phase and Small-scale Tests
- Tool Qualification Testing Phase
- Full Scale High Temperature Testing of a Complete Tool set-up as Proof of Concept
This industry-first development of a high-temperature inspection tool and the various phases of testing performed will ensure that the MFL ILI tool is suitable to meet the defined inspection boundary conditions successfully, far beyond the limitations of other attempts previously available in the market.

For some pipeline with challenging low frequency electric resistance welded (LF-ERW) seams and arduous operational profiles, hydrostatic testing has remained the preferred integrity assessment approach.    But with the elimination of in-service ruptures, the operators of these pipelines seek an alternative seam assessment method that provides a sufficiently conservative integrity assessment without the potentially negative impacts of hydrostatic testing.  As in-line inspection (ILI) and field nondestructive evaluation (NDE) improve, anomalies can be better assessed.  With improved ILI assessments, pipelines that have been historically hydrostatically tested can now use ILI to ensure operational integrity.  The improved ILI technology assessed in this work is an enhanced ultrasonic crack ILI tool with higher circumferential resolution and finer axial sample intervals.  The key is combining the knowledge of the population of potential anomalies with improved assessment approaches.  Historic ILI data, including magnetic ILI data, can often help with identification of noncritical anomalies types.  In addition, the emerging full matrix capture (FMC) imaging method that quantifies the size, position, and orientation of seam weld anomalies was examined to better understand the ILI capabilities and limitations.  This paper discusses the work performed to ensure the efficacy of the improved ILI and NDE methods to accurately detect and quantify all anomalies that could possibly fail a hydrostatic test.  An early step in the process was removing three sections of pipe from service for technology calibration and assessment.  Each spool was examined with ILI technology in a pump-through facility, inspected using many NDE methods and then destructively tested.  These results were communicated to ILI analysts and used to calibrate and improve the interpretation of the inspection results.  Then the pipeline was inspected as part of the scheduled integrity assessment.  Using field evaluation of anomalies detected by ILI, pipes were selected for removal from service for pressure testing and destructive examination.  This paper presents the inspection and destructive testing results in addition to prognosis for the use of the ILI in lieu of hydrostatic testing for LF-ERW pipe.

The Pipeline Industry regulator intends to ensure that operators possess traceable, verifiable and complete (TVC) pipeline records as a basis for sound integrity and risk management when verifying the MAOP of their pipelines. This is being driven by PHMSA in the notice of proposed rulemaking (NPRM) 49 CFR 192.624. The NPRM is a major topic of conversation for the industry and although the exact content of the NPRM has yet to be defined, it is clear that operators will be tasked to confirm what materials are present in their pipelines and characterize their properties. Inline inspection (ILI) plays a crucial role in supporting both of these tasks. This paper will discuss the role of ILI in the MAOP verification process, especially related to material verification as envisaged by NPRM 49 CFR 192.607. Operators have already started to utilize advanced ILI services as part of their material verification process in combination with in-ditch verification activities.
The paper will then go on to look at the aggregated results of numerous recent inspections and the developing picture this data gives of the range of materials in the North American pipeline network, the frequency with which outliers are identified (sections of unknown pipe), and other trends such as numbers of diversions


Mechanical damage is one of the most common threats to pipeline integrity. Undetected, it can lead
to significant damage and subsequent failure.
Recent developments in advanced in-line inspection (Ill} technology using multiple sensors or
mult iple t ools have made it possible to reliably detect and size anomalies. However, it is not always
easy to characterize those anomalies or to distinguish a mechanical damage defect such as a gouge
from an interacting threat like a dent with corrosion.
In addition, innovations in Ill technology mean there are significantly more data streams to review.
Reviewing Ill data often involves expending signif icant effort over long periods of time while coming
up with the same result - in other words, poring over the data until one is "blue in the face"yet still
not being able to discriminate pipeline defects accurately. What becomes important, then, is the
"art'' of looking, the ability to see beyond the easily observable, to bring things out of the side lines
and into sharperfocus.
What the industry requires, then, is a different way of " looking." Algorithmic processes can be
defined to look for well-defined obj ects such as welds in Ill data; however, it is often necessary to
use a heuristic or not fully algorithmic process to look for less well-defined anomalies such as dents
with gouges.
Multiple datasets can be leveraged using innovative techniques, resulting in a process to
discriminate critical mechanical damage features from other anomaly types. Once discrimination is
complete, anomalies can then be assigned to "buckets" based upon characterized features and
prioritized forfurther investigation based upon three criteria: their effect on integrity, their
environment and the impact offurtherinvestigation. This paper will discuss obstacles related to
analyzing mechanical damage-associated Ill data; outline the new process; and highlight the power
of the new approach as applied to example features.

In-line inspections provide valuable information about reported anomalies on a pipeline.  By comparing results from successive inspections, a pipeline operator can identify locations where corrosion growth is occurring and, sometimes, estimate corrosion growth rates.  But matching reported anomalies from two inspections is difficult due to uncertainties and variabilities associated with each inspection.  Pipeline operators have no clear guidelines on how to match anomalies and extract good corrosion growth rate information.
Nonetheless, many pipeline operators use anomaly matching after successive inspections to estimate corrosion growth rates.  Anomaly matching is straightforward when dealing with isolated anomalies, but there are no clear guidelines when multiple anomalies from one inspection match multiple anomalies in the other.  Understanding these more complicated situations is key to understanding where and when corrosion is occurring because isolated anomaly matches do not capture the more severe corrosion on a line.
This paper compares pit matching methods (e.g., closest anomaly, deepest anomaly, largest overlap, etc.) and introduces the use of modified Hausdorff distances for more complex situations.  Unity plots for isolated anomalies are compared to plots for more complex situations, and comparisons are made between the growth rates obtained with each.  Finally, a case study summarizes how the results can be evaluated in practical terms.

While In-line Inspection Magnetic Flux Leakage (MFL) tools have been used for many years to successfully manage corrosion related threats, small pinhole-sized metal-loss anomalies remain a significant concern to pipeline operators. The physical dimensions of these anomalies combined with their proximity to and/or interaction with other nearby anomalies can challenge MFL’s detection and sizing capabilities. Other factors such as tool speed, cleanliness of the line, and incorrect assumptions have an impact as well, but the effect of these factors is not well known.
Previous work has presented an analysis of the effect that incorrect assumptions about wall thickness and anomaly radial position have on sizing accuracy. This paper investigates the effects of sensor lift-off, tool speed and presence of magnetic deposits. Magnetic modeling software was used to determine the MFL response of metal-loss anomalies, and the effect of these factors on the MFL response and reporting accuracy was investigated.  In addition to the modeling, comparisons of actual MFL data with field excavation results were studied, to understand the limitations of specific MFL technologies and to determine the cause of MFL depth-sizing errors as well as the potential size of those errors.

Every year we hear more and more about Big Data, how it's going to be the future, how machine learning is going to change what we do and are capable of. But, after all this time, who can answer the following simple questions as it applies to our industry: How do I get Big Data?  How will it make a difference to integrity management program efficiency and ultimately pipeline safety? And when will it be here?
This paper focuses on answering these simple questions by providing a context for existing, active systems that are successfully employed today in the ILI world and particularly their subsequent role in pipeline integrity decision making. And then how such computing technologies as 'Big Data', 'Machine Learning', 'AI', 'Cloud Computing' are evolving with the increasing expectation of getting more for the significant volumes of information and performance data available to the pipeline operator and ILI vendors. Key considerations necessary for establishing a robust and viable framework needed to manage and analyze the vast quantities of pipeline inspection data being collected every day as a means for performance and reliability improvement will be discussed, including current and historical digital technologies of both ILI and NDE dig excavation systems.  Furthermore, how, with an ability to analyze pipeline defect characteristics over a wide spectrum of pipelines and conditions, development processes for advanced characterization and accuracy improvements can be dramatically accelerated relative to historic methods.  Insights will be provided as to how advanced analytics are now being further enabled with the massive processing power available through these cloud-based systems. Lastly, when we combine these Big Data environments with the ongoing improvements in ILI regarding higher accuracy and more consistent field calibration data, we outline how that can open up the possibilities for operators to better understand their pipeline risks and how to prioritize most efficiently when it comes to managing repair and maintenance.

In-line Inspection (ILI) tools equipped with Inertial Measurement Units (IMU) provide operators with the ability to examine the precise centerline trajectory of their pipelines. High resolution IMUs use a combination of gyroscopes and accelerometers to compute the attitude of the ILI vehicle; by pairing the IMUs with other ILI sensors and periodic above-ground surveyed locations, the geographic location of the pipeline can be calculated.
The primary use of IMU data from ILI has been to determine High Consequence Area (HCA) class locations and provide additional information for precisely locating field excavation sites. This is accomplished by correlating the computed geographic location of a pipeline to the ILI identified features.
This paper discusses two additional uses for IMU data to assess pipeline threats and the benefits of correlating them to a Geographic Information System (GIS).
Pipelines are constructed as nominally straight sections of pipe with prescribed bends placed along the route to navigate the acquired rights-of-way (ROW). Bends can be manufactured to specific dimensions before being placed or constructed in the field to accommodate route topography. Unintended curvature may also exist in pipelines due to forced pipe end alignment, ground subsidence, improper bedding or other external stresses. Both field bending and unintended curvature induce material strain on the pipe and can pose significant integrity threats to girth welds and interacting anomalies. Utilizing the IMU pitch, azimuth and odometer distances curvature for the entire pipeline can be calculated. Curvature values can subsequently be converted into horizontal, vertical and total strain values and assigned a tension and compression orientation. The strain and orientation values can then be analyzed for unintended bending, and correlated to the ILI data to identify interacting threats. Using strain calculations, field bends can be examined to ensure their radius of curvature is within the limits of pipeline codes. Strain due to unintended curvature can be correlated to interacting ILI features prone to initiate under additional stress and GIS that could highlight areas susceptible to ground movement.
External stresses causing unintended curvature can remain static but geological movement is constant and ground movement can often be dynamic. While ILI intervals vary, subsequent inspections do occur and the value of additional IMU inspections appears to diminish once a baseline pipeline location is known. However, in addition to the GPS location of newly identified features, curvature data from both inspections can be compared and differences analyzed for pipeline movement. To facilitate comparison, both inspections must have been completed by a high resolution IMU and have matching odometer information, this is typically scaled or interpolated from a girth weld alignment. Both sets of curvature data can be overlaid and differences in horizontal or vertical strain calculated. Strain differences above the stated measurement accuracy for both inspections often indicate the pipeline has moved since the previous inspection. When above ground survey spacing is 3km, a single IMU inspection accuracy is generally ±1m in latitude, longitude and elevation. This is a 1:3000 accuracy for each inspection,and represents a possible 2m displacement measurement if the location of each inspection is directly compared. This displacement error is inadequate for assessing how much a pipeline has moved. By tying the two solutions together at tighter intervals this error can be significantly reduced and displacement errors of 10cm or less can be realized. Areas of movement can additionally be correlated to GIS data to highlight areas susceptible to ground movement.

Pipeline operators rely extensively on in-line inspection (ILI) systems and other forms of non-destructive examination (NDE) as the basis for meeting the continual assessment requirements to evaluate pipeline integrity. The verification of ILI prediction accuracy involves correlating ILI data to direct inspection/assessment information using traditional NDE methods such as Ultrasonic Testing (UT), Phased Array Ultrasonic Testing (PAUT), and Electromagnetic Testing (ET). Substantial effort has been taken to establish and consider measurement tolerances for ILI systems, but not as much attention has been placed on understanding the tolerances of NDE technologies. Through studies conducted by Pipeline Research Council International, Inc. (PRCI) these NDE methods have been shown to have inconsistencies and inaccuracies in sizing and characterizing pipe wall and weld seam crack-like anomalies. Effectively managing the uncertainty in NDE measurements and selection of the appropriate technologies for anomaly verification is as important to continuous improvement in pipeline integrity management programs as ILI tool tolerance. A primary goal for pipeline operators is to develop field-ready NDE methods for full volumetric characterization of pipeline anomalies. The inability of traditional NDE methods to accurately provide a three-dimensional (3D) image of a pipe wall features in the field leads to critical decisions being made based on imprecise tools, with layers of conservatism being included in the analysis for both NDE and ILI measurements. The lack of precision often leads to conservative and therefore excessive repair digs, unnecessary pipe replacements, and in some cases hydrostatic pressure testing of the pipeline to verify the system integrity. This paper presents the results of a research project that includes comparative analysis of seam anomaly characterization data from an ILI system, traditional NDE inspection, Computed Tomography (CT), and metallurgical results. CT is the only current method that has shown consistency in providing accurate 3D profile measurements of an anomaly comparable to destructive testing and direct measurement of anomaly characteristics. The CT results presented in this paper represent the potential for field ready inspection capabilities as the data were obtained from full circumference measurement of pipe samples rather than plate samples, which have typically been studied in prior analysis of CT methods. Obtaining accurate data on anomaly dimensions is critical to understanding and improving the application of ILI and NDE data to drive integrity decisions. Improved results in the field will lead to improved decision-making to protect the environment and public safety.

Corrosion anomalies which reduce the strength of the pipeline must be mitigated appropriately. When corrosion defects have varying morphologies it is not always simple to determine the point at which the corrosion region becomes a safety concern, particularly for complex corrosion areas where multiple corrosion anomalies may interact with one another. Therefore, understanding how various anomalies may interact is important to determining the overall remaining strength of a pipeline under pressure. Many criteria for this spacing and how to apply the rules are recommended in the literature and have been studied either as the focus or periphery by several more, but no single criterion is provided as regulation. The task is left to the pipeline operator to choose the interaction rule for what is defined as ‘closely spaced corrosion.’ The method by which the failure pressure is calculated should be considered as varying levels of conservatism are inherent in these assessments. Recommendations for interaction guidelines have been determined by either empirical or analytical approaches. The empirical approaches may be limited when an insufficient number and variety of pipes can be burst tested. Many analytical approaches are based upon relationships of remaining wall and simple corrosion morphologies which may not be applicable to real world corrosion. The source of the corrosion anomaly data is an important variable when selecting and applying interaction rules. In-line inspections (ILI) are the most common methods by which to obtain corrosion anomaly data, but each technology has an inherent measurement error and bias which should be considered. This paper will go into detail on each of the items discussed, present the current state of research into this subject in the industry, and will present a general recommendation for selection of an interaction criterion for corrosion features

The project “Development of an Industry Test Facility and Qualification Processes for in-line inspection (ILI) technology Evaluation and Enhancements” aims to expand knowledge of ILI technology performance and identify gaps where new technology is needed. Additionally, this project also aims to provide ILI technology developers, researchers and pipeline operators a continuing resource for accessing test samples with a range of pipeline integrity threats and vintages; and inline technology test facilities at the Technology Development Center (TDC) of Pipeline Research Council International, Inc. (PRCI), a PRCI* managed facility available for future industry and PHMSA research projects. An ILI pull test facility was designed and constructed as part of this project based on industry state-of-the-art and opportunities for capability improvement. The major ILI technology providers, together with pipeline operator team members, reviewed the TDC sample inventory and developed a series of ILI performance tests illustrating one of multiple possible research objectives, culminating in 16-inch and 24- inch nominal diameter test strings. The ILI technology providers proposed appropriate inspection tools based on the types of the integrity threats in the test strings, a series of pull tests of the provided ILI tools were performed, and the technology providers delivered reports of integrity anomaly location and dimensions for performance evaluation. Quantitative measures of detection and sizing performance were confidentially disclosed to the individual ILI technology providers. For instances where ILI predictions were outside of claimed performance, the vendors were given a limited sample * Formerly with PRCI. of actual defect data to enable re-analysis, thus demonstrating the potential for improved integrity assessment with validation measurements. In this paper, an evaluation of the ILI data obtained from repeated pull-through testing on the 16 and 24-inch pipeline strings at the TDC is performed. The resulting data was aligned, analyzed, and compared to truth data and the findings of the evaluation are presented.

Deciding the precise location of a defect is necessary for successful validation of metal loss defects identified by inline inspection.  Options for calculating distance along the pipeline are presented and their suitability discussed.  The quality of inspection required for validation is also discussed and an overview of the available techniques provided.

Historically, hazardous liquid pipeline operators have approached in-line inspection (ILI) validation months after the tool run, relying on verification digs and analysis of unity curves to validate tool accuracy and determine data adequacy.  However, with implementation of the new regulation in 49 CFR 195.591 (effective January 23, 2017), operators must comply with the requirements and recommendations of American Petroleum Institute Standard 1163 (API 1163), Inline Inspection Systems Qualification Standard.  Compliance with API 1163 necessitates a comprehensive written framework to guide operators through all elements of the standard, including identifying specific risks for investigation, selecting proper inspection technology, and collecting data at all stages of the ILI project to ensure operational validation and inspection results verification.  To aid operators in this effort, a holistic, comprehensive approach to ILI verification should be taken: one that begins during the assessment planning and tool selection phase and continues throughout all stages of the project to ensure each program element is satisfied, including ILI specifications, personnel qualifications, tool speed and sensor validation, results verification, and compliance documentation.  Use of this program can also alleviate the technical paralysis often associated with determining the number of verification digs and related unity curves.  This presentation will review a total quality management approach to assist pipeline operators in complying with API Std 1163.

A gas lift line with 14% restricted passage was inspected using Quest Integrity’s Ultrasonic Metal Loss-Geometry combination ILI tool. The pipeline was leaking, critically affecting production rate and inspection activity. With minor line modifications and repair, the ILI tool completed the 15km inspection in 10 hours. Due to sludge in the line from incomplete cleaning, the first pass captured only 70% data coverage. A 72-hour expedited preliminary report enabled the location and prioritization of the necessary repairs.
A run-comparison was initiated as an alternative to re-inspection and to supplement the integrity management process, without affecting production. For a comprehensive integrity/risk management, corrosion growth rates were established by comparing raw signals from the previous 2010 inspection and current UT inspection, and applied to anomalies in the obstructed area.
Completion of high-risk anomalies’ field verifications increased confidence and refined the run-comparison results. Interestingly, ROV survey pipe displacement results correlated with the UT ILI reported dents and field bends. Further work should be conducted to maximize the value of UT inspection in this area as it can provide valuable information of subsea pipe.
This inspection’s operational learnings were applied to two subsequent similar ILI inspections, collecting 99.5% data, with minimal effect on production

As composite repairs become a more common tool for integrity engineers and project managers, the question has evolved from “will composites work” to “how long will a composite work?” While many operators and composite manufacturers are in favor of “bury and forget” and call every repair permanent, there are many serious issues that need to be addressed. This paper seeks to highlight and clarify many of these concerns and explain current repair standards including general repair theory, composite fatigue, cyclic conditions and environmental impact with a focus on specific defect types such as dents and cracks. Composite repairs shouldn’t be designed for today, they need to be designed for predictability in the future.

Recent guidelines and laboratory tests have allowed assessments to be made of defective girth welds in transmission pipeline, permitting operators to choose remediation over removal. This study extends the limits of structural reinforcement using current weld inspection techniques to identify specific anomalies that can be addressed using an engineered solution.
Finite Element Analysis (FEA) was applied to study the effects of using a carbon fiber and epoxy composite material to enhance the structural integrity of a pipeline with defected girth weld joints in sub-zero conditions. FEA also was applied to applications of this composite material on full-scale pipe spools.
Linear FEA simulations were performed to evaluate the composite reinforcement of a crack in the girth weld. Experimental testing was carried out on 24-inch (600-mm) diameter pipe spools at -20ºC with a simulated through-wall crack using strategically located tack welds to hold the two spools together. Two configurations of the pipe were tested – leaking and non-leaking. The leaking configuration was tested using four repair configurations. In addition, under conditions provided by an interested end user, an engineered composite wrap design was installed and tested to the maximum pressure on a test spool designed to simulate a failed weld. Strain gauges were used to quantify the composite repair system’s effectiveness at low-temperature (-20ºC) and high-pressure (32 bar) conditions. Both FEA results and full-scale testing demonstrated that the composite repair system is capable of sealing and successfully reinforcing simulated through-wall cracks in vintage girth welds.
Based on the testing, an end user elected to implement the repair system for defected vintage girth welds. This paper includes a discussion of the successful field implementation and logistics for repairing the girth welds.

The reinforcement of cracks and crack-like flaws in transmission pipelines continues to be an area of significant interest for pipeline operators. In recent years, composite systems have been proposed as a potential reinforcement technique; however, few extensive studies have been conducted to characterize their performance when reinforcing cracks and crack-like flaws. Therefore, a study was undertaken to assess the performance of composite systems when reinforcing crack-like anomalies using modern, high-frequency electric resistance welded (HF-ERW) and vintage, low-frequency ERW pipe (LF-ERW).
To evaluate the performance of composite reinforcement systems, cyclic pressure, hydrostatic, and pressure-to-failure (burst) tests were completed using HF-ERW and LF-ERW pipe with pre-cracked electric discharge machined (EDM) notches to represent crack-like flaws. Multiple composite repair systems and pipe sizes were tested. The pre-cracked EDM notches were located in both the base pipe and the long seam weld at installation depths of up to 65% of the pipe's nominal wall thickness.
The cyclic pressure and burst testing demonstrated that performance of the crack-like defects improved when reinforced using the systems considered in the study. This was true for all sizes of modern and vintage pipe tested. Cyclic pressure testing showed that reinforced samples almost always exceeded the number of cycles to failure of unreinforced samples, and all reinforced burst test samples had failures in excess of 100% SMYS. Variations in the amount of reinforcement provided allowed the systems to be ranked and reviewed for future design considerations.

In 2007, external corrosion and linear indications were discovered underneath shrink sleeves installed on field girth welds located on a 10-mile section of Natural Gas transmission pipeline which was constructed in 1993.  Over the next 10 years, 1,615 of 1,971 shrink sleeves were removed.  Given the volume of this dataset, the operator established a goal to prioritize removal of remaining shrink sleeves in this segment, while establishing a predictive proactive model, to prioritize shrink sleeve removal across a system.
The system is expected to contain nearly 20,000 shrink sleeves.  To build this model, data was managed in a custom developed SQL database with Microsoft PowerBI, ESRI ArcGIS Server Portals, and machine learning algorithms applied for powerful analytics. We will define the lifecycle of a shrink sleeve goes through four stages:

In concept, Stages 3 and 4 can be detected through in-line inspection, whereas Stage 1 and 2 are believed to be predictable through modeling.  The fundamental result of this project delivered a data structure where advanced predictive analytics can be applied to develop models capable of predicting Stage 2, specifically.

Steel sleeves play a critical role in the pipeline rehabilitation program of most pipeline operators. Steel sleeves are used to repair a wide range of anomalies, including corrosion and dents. Three full-scale studies were conducted to evaluate the performance of steel sleeves used to reinforce corrosion and dent features subjected to cyclic pressure service. The first study, called the Dent Validation Collaborative Industry Program (DV-CIP), was sponsored by pipeline operators and repair companies. Strain gages were installed on two dents reinforced with steel sleeves; an incompressible filler material was used fill one of the steel sleeves, but not the other. The other two studies evaluated the performance of Type A and B steel sleeves used to reinforce corrosion and dents subjected to static burst and cyclic pressure conditions.
This paper will provide valuable information regarding the importance of filler materials and also quantifying the strain reduction provided by steel sleeves used to reinforce corrosion and dents subjected to static and cyclic pressures. Of particular importance are the empirically-determined fatigue lives that can be used to provide both liquid and gas transmission pipeline operators with estimated service lives for corrosion and dent features reinforced with steel sleeves.

ND gov. Burgum challenged the North Dakota oil industry to eliminate spills while doubling production to 2 million barrels a day. As a follow-up the iPIPE consortium was formed to scan worldwide for novel leak detection and prevention technologies to be tested by the operators taking part in the consortium.
This paper presents the results of a six-month testing period of one of the two selected technologies; Ingu's Pipers, an in-line screening tool for small diameter and unpiggable pipelines that identifies risks that threaten pipeline performance and safety with zero-downtime.
The paper will cover a comprehensive set of tests in metallic and non-metaliic lines transporting oil, gas or produced water.

External Leak Detection (ELD) is an emerging class of pipeline leak detection in which sensors deployed external to the pipeline detect release events by measuring specific quantities such as temperature, vibration, strain and presence of hydrocarbon vapors and/or liquids. Such systems are reportedly capable of detecting small releases that fall below the detection threshold of traditional leak detection systems (i.e. computation pipeline monitoring systems) and can therefore serve to complement existing leak detection systems by extending the overall detection range. There are fundamental differences between traditional pipeline leak detection and ELD and therefore a different approach is required for evaluating ELD systems.  To this end, an ELD evaluation framework has been developed. It is intended to be used by pipeline operators to identify and evaluate candidate ELD systems intended for possible deployment on onshore transmission pipelines, and to ultimately assist operators in making an informed technology selection. The developed framework provides a basis for verifying the performance of candidate ELD systems in the context of specific performance requirements unique to the pipeline for which ELD is being considered. It also provides a systematic and objective means for comparing multiple ELD systems.

On May 9, 2018, PHMSA released the draft version of a report titled “Pipeline Risk Modeling Overview of Methods and Tools for Improved Implementation”. This is a long (110-page) report that includes discussions and technical recommendations, including those from NTSB.
A focus seems to be on preserving the ASME B31.8S list of four types of risk assessment models: Qualitative, Relative/index, Quantitative, and Probabilistic. Each type is discussed and evaluated.  Beyond the choice of this risk modeling categorization, there is much to digest from this report.
In this PPIM paper, the authors distill the content of this long PHMSA document into critical take-aways from a risk assessment practitioner perspective.  This includes answers to basic questions:  What should a pipeline operator get from this guidance document?  Should an existing risk assessment methodology be modified, based on this report?  What is the best choice for pipeline risk assessment?

Pipeline risk assessment is a foundational component of effective pipeline integrity management. The Federal pipeline safety integrity management regulations require pipeline operators to use risk assessments.  Many operators adopted primarily qualitative and relative risk models to initially meet the need to prioritize baseline integrity assessments and identify pipeline threats.  However, the use of risk assessment required by the pipeline safety regulations goes beyond the prioritization of pipeline segments for baseline assessment and includes (but is not limited to):

As pipeline operators attempt to meet the above requirements and progressively adopt an operating strategy of continual risk reduction whilst minimizing total expenditures within safety, environmental, and reliability constraints, the need for a more quantitative approach to risk assessment is becoming increasingly evident.  This view is supported by a recent PHMSA review recommending pipeline operators develop and use risk models consistent with the need of supporting risk management decisions to reduce risks rather than focusing the choice of risk model on the perceived initial quality and completeness of the input data.  The use of quantitative risk models will enable superior understanding of the risks from pipeline systems and improve the critical safety information feeding into the overall integrity and risk management processes.
Whilst a wholescale move to quantitative risk modelling is likely to be a daunting task to many operators, it is possible to achieve a meaningful quantitative risk assessment for what are likely to be the key threats affecting pipeline integrity management without the need of a major investment to cover implementation of new software and associated databases.  Such a risk assessment can be achieved for many threats by using the quantitative data provided by an ILI survey of the subject pipeline augmented by relatively few other key data attributes. 
This paper describes an ILI based risk model developed initially for gas pipelines that provides a quantitative risk assessment of multiple pipeline threats (probability of failure and consequences of failure).  Results are presented to show the value of quantitative risk outputs and how they can be used to support decisions in the overall integrity and risk management processes

Roughly 10,000 "Baby Boomers" will turn 65 every day for the next decade.  This trend is particularly acute in Oil & Gas as boom and bust cycles continue to displace the workforce.  The pipeline industry can adapt to the changing workforce.  The desired result is reduced time to competency, saving money.  A key success factor is the ability to build up expert competence fast.
This paper presents a structured approach to knowledge risk management.  The approach is designed to complement an operator's API RP1173 PSM system.  The approach consists of three parts: a knowledge risk assessment, a gap analysis, and deployment of tools to fill identified gaps and develop continuous improvement.
Knowledge risk assessment is a qualitative process designed to identify risks related to knowledge retained by the organization or individuals.  The assessment examines knowledge critical for performance, the level of proficiency available/needed, the degree of documentation, and the extent critical knowledge is held by a small number of people.  These four dimensions provide a comprehensive profile of the way knowledge is currently managed and signposts the risks between current and future states.
The risk assessment produces a risk matrix and listing of specific knowledge areas where gaps exist or knowledge is at risk.  This knowledge may be explicit, implicit, or tacit.  A gap analysis is then conducted to identify key focus areas and specific knowledge to be addressed.  The final stage is application of tools to quantify the knowledge gaps and develop continuous improvement.  Tool output is used to define the scope and deliverables for future initiatives.

On May 9, 2018, the Department of Transportation’s Pipeline and Hazardous Materials Safety Administration (PHMSA) released a draft report in support of improving commonly used pipeline risk models. The report was the culmination of data collected during the Risk Model Working Group (RMWG) technical presentations and meetings held in 2016 and 2017. These meetings focused on defining, reviewing, and documenting industry best practices in implementing various forms of risk models to accurately assess the risk of a given segment of pipeline. The RMWG and other industry experts reviewed both qualitative and quantitative (probabilistic) risk models and concluded quantitative models may be the most versatile and capable to provide the best estimation of risk. However, some operators, in particular the smaller scale operators, do not have access to the appropriate amount of data or expertise necessary to fully utilize a quantitative model. This paper will compare and contrast the various modeling approaches, including qualitative (index-based), SME-based, and data driven quantitative models in an effort to assist operators in the selection of an appropriate model based on their capabilities and data availability. The models evaluated in this paper include index models, probabilistic models, qualitative models, and relative assessment models. The criteria used for the PHMSA evaluation included overall ease of use, data collection methods, interpretability of results, scalability, and ease of integration of new data.

Risk management of gas and hazardous liquid pipeline systems is a core element of integrity management regulations (49 CFR part 192, subpart O; 49 CFR 195.452) and an integral component of asset integrity management programs. The Pipeline and Hazardous Materials Safety Administration (PHMSA) recently published a Pipeline Risk Modeling Report supporting improvements in pipeline risk models and highlighting advantages of quantitative risk approaches. Implementation of advanced, quantitative risk methodologies allows for evaluation of integrity projects through the calculation of quantified risk reductions and return on investment (ROI) analysis. This paper will provide a framework for utilizing a quantitative risk program to inform the evaluation of asset integrity projects

A nondestructive assessment of electric-resistance-welded (ERW) seam types through in-situ inspection can provide valuable data for pipeline integrity programs. The durability of ERW seams is known to depend on pipe manufacturing practices that have evolved over decades.  However, for pipelines that have incomplete or missing material test reports (MTRs), there is no method to accurately identify the seam type and quality without the destructive removal of a section of the pipe wall at the seam for laboratory examination. This work presents a new methodology to accurately and nondestructively identify low frequency (LF), high frequency (HF) and high frequency normalized (HFN) ERW seams. The approach combines multiple nondestructive evaluation (NDE) techniques including a characterization of the macro-etched heat-affected-zone (HAZ) and assessment of hardness variation across the seam obtained through Hardness, Strength, and Ductility (HSD) tests.  These quantitative measurements are input into a classification model that is calibrated with an existing database of known ERW pipe joints to automatically classify the ERW seam type. Case studies on in-service ERW pipe joints from in-ditch assessments are also provided.  Future work will extend this multi-variable model to include characteristics of the steel quality and grade to establish correlations with an index of seam toughness.

The pipeline industry continues to strive for improvements when evaluating the severity of dents. As one current example, API is developing a Recommended Practice “Assessment and Management of Dents in Pipelines” with one of the objectives to provide guidance on engineering critical assessments of dents. Previous research has provided methodologies to assess the fatigue life of a dent using stress concentration factors. Recognizing that the existing models have limitations, questions have been raised regarding the use of stress or strain concentration factors in dent assessments. Stress concentration factors have shown applicability when the response of the dent is linear with pressure fluctuations while strain concentration factors may be more relevant when the response of the dent is non-linear.
This paper examines the differences between stress and strain concentration factors using Finite Element Analysis (FEA) for various dent shapes and internal pressures.  The modelling assumes elastic-plastic material properties and large displacements to accurately capture the denting and rerounding process and during subsequent (shakedown) pressure cycling. The resulting stress and strain concentration factors are examined and compared as a function of internal pressure for each dent shape scenario.  Finally, and to provide context, the impact on fatigue life is discussed.
This study aims to better understand the applicability and limitation of stress and strain concentration factors when performing engineering critical assessments of dents.

The Pacific Gas and Electric Company’s (PG&E) facility integrity management program (FIMP) is developing a program to verify the maximum allowable operating pressure (MAOP) of pipes in gas transmission stations. The first phase of PG&E’s program leverages state-of-the-art nondestructive examination (NDE) technologies to identify, detect, and size manufacturing and construction defects in the body, long seams and girth welds of station pipes. The distribution of manufacturing and construction defects for a given pipe is taken – along with inputs for material properties and historical operating condition data – to calculate the burst pressure and remaining life using a probabilistic engineering critical assessment (ECA) framework. In 2018, PG&E implemented a pilot program for a number of measurement and control gas transmission stations. Some of the challenges faced during the pilot program have required the development of a more rigorous pre-field NDE protocol, which establishes thresholds for anomalies that require depth sizing and the minimum defect size that shall be detected. These protocols leverage the extensive multi-year records research effort PG&E is undertaking on all station features, historical manufacturing practices, and pipe fracture toughness databases. The established protocols create a practicable field NDE based MAOP verification program for assets that cannot be in-line inspected.

Pipeline integrity management focuses on the engineering critical assessment of analyzed data; allowing for action in a timely and prudent manner. Analyzed data is most often presented in a tabular output such as Excel spreadsheets or database files. Analyzed data can also take the form of multi-axis grids, finite element models or 3D copy renderings.  With advancements in the presentation of analyzed data, engineering critical assessment continues to improve with focus on specific target anomaly challenges, advanced numerical/statistical analyses and innovative integration of multiple analyzed datasets. The assessment of in-line inspection data has evolved from the qualitative interpretation of a “log reader,” to the implementation of emerging sciences and innovative application of proven methodology.
However, actionable information can also be obtained from a raw data review of the recorded data. While human interpretation is limited in the ability to quantify engineering critical parameters, this paper will discuss two case studies where raw data review directly provided information such that asset integrity actions could occur in a timely and prudent manner.
Case studies will be presented to illustrate the importance of critical engineering assessments of inspection data.

In the current era where pipeline safety is of paramount interest, optimization of corrosion mitigation for vintage pipelines has a significant effect on the industry’s management systems and related costs. To that end, reliability-based Limit State Design and assessment (LSD) corrosion criteria have been developed for onshore pipeline as part of a joint industry project. These criteria have the unique characteristic of being calibrated to meet specific reliability targets within a certain tolerance but having a deterministic format that does not require probabilistic calculations for application.
The assessment criteria cover both burst and leak failure modes. Their formulations are characterized by three elements: the equations used to calculate the characteristic demand (e.g. operating pressure) and capacity (e.g. burst pressure resistance at a corrosion feature); the characteristic values of the key input parameters for these formulas (such as yield strength, pressure and feature depth); and the safety factors defining the characteristic demand as a ratio of characteristic capacity.
The performance of the LSD criteria, defined as the number of corrosion repairs required, was compared to those required by the CSA Z662 deterministic assessment criteria and the full probabilistic criteria developed by TransCanada Pipeline Ltd. (TCPL) to ensure that the criteria lead to practical solutions for real cases. The comparison was conducted using a comprehensive set of TCPL pipeline cases that cover a wide range of pipeline diameters (NPS 6 to 42), hoop stress-SMYS ratios (0.4 to 0.8) and corrosion severities (feature density from 0.625 to 6508 per km). The results show that the LSD criteria perform similarly to the TCPL reliability-based criteria, and that both are generally less conservative than the CSA deterministic approach.
This comparison demonstrates that the LSD criteria provide a simple and deterministic procedure that capitalizes on the benefits of reliability analysis in eliminating unnecessary conservatism and focusing on the repairs required to achieve consistent safety levels for all cases. Thus, these criteria will enable operators to optimize the cost by achieving maximum risk reduction for the dollar spent.

Pressure cycle fatigue analysis following a fracture mechanics approach, such as the Paris Equation, can be a time consuming and computationally intensive endeavor.  Depending on the resolution and span of your pressure spectrum (typically ∆5psi and 3 months), hours or even days of computation time is required to complete a fatigue assessment on an ILI data set, especially for large crack populations (>20,000).  Pipeline Operators will not sacrifice precision, and therefore will not opt for the simpler S-N based fatigue relations offered by the Miners rule. The Equivalent Load method addresses the efficiency issue of the Paris Equation by tackling the bottleneck step– which is the calculation of the Load-Integral.  The Load-Integral requires cycle-by-cycle integration through the rainflow counted pressure series to properly define the incremental crack advancement and stress intensity factor at the crack tip.  However, the Load-Integral of one flaw can be related to another with correction factors that account for differences in cyclical stress.  Following this technique, full integration of the Load Integral is not necessary for every flaw that is part of the same pressure series (pump to pump segment), which cuts down the computation time from days to minutes.  This paper discusses the derivation of the Equivalent Load correction factors, and the validation study.  The Equivalent Load remaining lives were compared to the BS7910 Annex M Fatigue model for six crack In-line-inspection data sets from different liquid transmission pipelines which varied in vintage, diameter, and historical cycling severity.  The comparison analysis demonstrates the Equivalent Load Method generates similar remaining lives with an average difference of 2% from BS7910, but at a dramatic fraction of the time.

Dent with metal loss (DwML) features detected by in line inspection (ILI) may be caused by contact by excavating equipment or by contact with rocks.  The damage from these two sources differs notably.  Currently,  PHMSA treats DwML damage located in high consequence areas  as severe conditions demanding immediate response.  PHMSA proposes to extend the response requirement to medium consequence areas and allow pipeline operators to apply an engineering critical assessment (ECA) and postpone response if certain structural integrity criteria are met.  The current paper describes the detailed technical background that can support a simplified ECA process capable of distinguishing the differing effects of metal loss as gouges or corrosion.  For metal loss determined to be gouges, the process provides criteria to classify the feature by severity levels with appropriate response schedules.

Many different inspection and maintenance activities are conducted on gas transmission and gas distribution pipelines – valve inspections, in-line inspections, CP system inspections, odorant monitoring, etc. – representing significant operational expenditures for pipeline operators. With sound risk models, risk-based optimization approaches can be applied to these integrity management activities to provide optimized planning strategies which balance their risk and cost. The result of implementing these risk-based inspection intervals has been measurable savings consistently ranging between 20 – 40% with no impact to the overall system risk. In this paper, examples are provided for risk-based optimization of leak survey programs, odorization system inspections and valve inspections. Through development of models to characterize the risk benefit and costs of inspections, a set of the optimum inspection intervals for minimizing the risk-cost combination is determined. Use of the optimum intervals is demonstrated, including the impact on OPEX expenditures. A metric is introduced to evaluate the efficiency of different inspection activities, allowing for OPEX dollars to be allocated to activities which provide the most risk reduction benefit for the amount spent. The general risk-based framework for determining optimal inspection intervals is suitable for any periodic gas pipeline inspection or maintenance activity.

Pressure testing is one of the oldest methods of ensuring integrity of pipelines and other vessels. Because of this, there exists a massive amount of information on the topic in written and verbal form, much of which is outdated, misunderstood, inconsistent with other information sources and in some cases incorrect. Additionally, it’s often found that the pressure testing procedure(s) incorporated into internal manuals of gas and liquid operators is one of the oldest procedures and therefore potentially outdated and not in keeping with newer information and technological advances, specifically with regard to data-collection instrumentation, certification tools, and proper usage of spike testing. The paper will present the observations of the authors regarding the present state of pressure testing for hazardous liquid and natural gas pipelines. It will provide a guide to pressure testing best practices, by reviewing the sometimes inconsistent, outdated and technically deficient procedures currently in use.
This paper will:

  1. provide a comprehensive guide to pressure testing best practices drawing the distinction between gas and liquid code,
  2. provide resolution and clarity where industry approved documents conflict, and
  3. be a tool that operators can utilize to review and update their internal procedures

America is at the forefront of the shale oil revolution, where a matrix of shallow wells is drilled across an area to produce oil from shale formations.  Many pipeline networks have sprung up to deliver crude oil and produced water from numerous wells to central processing and distribution facilities.  Gathering networks share the same risks of leakage as other hydrocarbon carrying pipelines, and federal regulation CFR 195.444 requires leak detection systems on the pipelines that pass through High Consequence Areas (HCAs).  Computational pipeline monitoring (CPM) leak detection systems typically installed on a central computer, continually monitor pipelines using data from instrumentation installed at strategic points on the pipelines.  This presentation will address many of the challenges in implementing CPM leak detection on crude oil and produced water gathering networks, including selection of the right technology based on instrumentation, communications, and the transient nature of these pipelines.
In addition to implementation, the presentation addresses the entire lifetime of the leak detection system, including testing, performance, maintenance and network expansion.  The physical layout of gathering networks change far more frequently than traditional transmission pipelines due to the dynamic nature of drilling and exploiting shale fields.  The constant addition to the number of injection points complicates the effort to maintain optimal perform in the leak detection system, especially on systems that already have dozens of injections.
The presentation draws on the experience of implementing leak detection on dozens of gathering networks in Texas, North Dakota, Saskatchewan, Alberta, and Manitoba and includes graphs showing real pipeline data and leak detection responses.
The presentation is intended for control room operators, project managers, control room managers and application engineers


please check back

The 'traditional' way of managing pipeline integrity, at least based on PETRONAS' experiences, relies on governing manuals, procedures and guidelines; competent people; effective systems - database, integrity/risk assessments, maintenance management, reliability/integrity performance software/dashboard etc. software; and various technologies for inspection, monitoring, maintenance and repair. The CORE of the whole process is COMPETENT PEOPLE as all sub-processes or activities require full intervention of managers, engineers, technicians and operators, at different stages of the activities. The 'traditional' way can be considered effective since most, if not all, of pipeline operators subscribing to the above framework. In recent two to three years, we've heard the buzz of '4th Industrial Revolution' or 'Industrial Revolution 4.0 (IR4)'. What is that all about? It is about digitilization, 'Big data', machine learning and artificial intelligence. In PETRONAS, there are on-going initiatives at petrochemical plants, gas processing plants, upstream's well engineering and retail business in pursuing the digitilization inline with the IR4. The aim of the initiatives is basically to improve the efficiency of critical processes thus at the end will reduce overall cost and improve production margin and revenue/profitability. It is easier to implement and justify for the plants, well and retail as they're 'directly' have impact on the bottom-line. But what about the oil and gas pipelines? This paper will present the initiative of PETRONAS' Pipeline fraternity consisting of pipeline experts, managers and engineers in the development of intelligent and integrated pipeline integrity management system. The system is designed to have minimal PEOPLE intervention as it will utilize artificial intelligence/machine learning. It is envisage that by having the system, critical decision making pertaining to integrity and reliability of pipeline system could be done in 'split seconds' extending the asset life and eliminating unwanted incidents i.e. leak/rupture.

What is bad? What is good? When we review a set of in-line inspection data, we get a feel for the condition of a pipeline. If the condition is good, then the future integrity management should be relatively straightforward and inexpensive. If the condition is bad, then there will be more work to do and we can expect significant costs for rehabilitation.
As the first step towards a true system for the predictive maintenance for pipelines, we have started a process of organizing the in-line inspection of data sets. For each pipeline, we have defined some key descriptors such as product, operating stress, age, coating types, etc. Then we calculated a few simple metrics such as the number of features per kilometer. We can then compare the condition of specific pipelines with large populations of pipelines and categorize the condition in relative terms. By comparing populations of different ages, we can make predictions regarding the likely future condition.
Much of our regulation in the industry is based on very uncertain or poorly understood science, and we compensate for this by being conservative. Analytics on large volumes of data will give us a more reliable and cost-effective way to determine whether a pipeline is 'good' or 'bad' and respond appropriately. Many purists would call this an 'ugly' method, compared to the elegance of fracture mechanics. But we think it's the future.
The result is an extremely powerful tool for planning future long-term integrity management activity. This paper describes the development of this application of data analytics and illustrates some fascinating examples and insight gained along the way. Issues encountered with different metrics are identified and key representative metrics proposed.

Over the past decade the Energy Industry has been going through a digital transformation that is changing the way the Industry does business in the new era of lower Oil prices. The advances in sensor technology, big data analytics and automation, combining with the new age of IIOT (Industrial Internet of Things) has created a new landscape of information, resource optimization and competitiveness. The digitization of assets is enhancing production, reducing unplanned shutdowns and increasing quality, safety and efficiency at all levels within an organization. Data and the actionable insights it can give is at the forefront of this new digital revolution.
While the benefits of sensor and digital technologies are spreading across the Energy Industry the pipeline sector has been a slow adopter. The industry lags behind the drilling and production sectors that are starting to see the monetarized benefits of adopting a digital technologies.  It could be argued that pipeline infrastructure is the most critical part of any asset but currently benefits the least from the benefits of digital technologies. This lack up uptake can be attributed to the extraordinary challenges presented buy digitizing a pipeline network. Sensorizing pipelines is more challenging than other assets, as they are geographical dispersed, more often in difficult and remote locations, and there is a significant legacy network that is inaccessible (buried or subsea).
Refinery pipelines are easily accessible above ground and have access to power and communication networks so will be the easiest assets to sensorize and monitor. However, It is the upstream networks, both onshore and subsea, carrying and gathering the unprocessed product from the wells that are most at risk from aggressive and dynamic failure mechanisms and would benefit the most from big data and predictive maintenance strategies. This paper will explore how the Energy industry can start bringing the benefits of digital technologies to these pipeline networks and look at some of the building blocks that maybe necessary to achieve this. Ideas and experiences on the simplicity of smart pigs, frequency of deployment, big data and automation as well as the IIOT will all be investigated.

This paper covers details of two project cases involving a vast gas network and a vital cross-border transmission system.  The first involves a journey over the last 15 years, as the network operator had development and expansion plans in mind.  A key component of that plan is to have a means to monitor the operational and physical integrity of that ever-expanding network.  This presentation will cover the process and the journey of that, how they’ve expanded and the methods employed to test and keep their systems relevant.  The second involves the challenges in cross-border systems, improving the economics based on design and subsequent operations.  This high-profile case covers the innovations considered in maintaining integrity and how operations has progressed to date.

As the pipeline industry ages, the quantity and complexity of pipeline data are both rapidly increasing. As a result, those individuals responsible for managing pipeline integrity are likely to find it increasingly difficult to make decisions with confidence and competence. To Support pipeline operators in this endeavor, ROSEN has created the integrity managment platform NIMA. NIMA combines sophisticated data management functionality with a base library of integrity management processes.
This paper explores the challenge of data management and processing through a case study on a network of over 80 pipelines. The pipeline data (e.g. in-line inspection, survey data and environmental data) are integrated into a single cloud-based application, allowing for efficient, automated fitness-for-service and corrosion growth analysis using Python. This provides a rapid means of assessing the immediate future integrity of each individual pipeline according to best practice.
With integrity assessment results structured and stored in a consistent manner, data analytics can be used to extract further value from the data. 'Descriptive Analytics' techniques (e.g. machine learning) can be used to create generalized models for unseen data.
It is shown that these techniques can support operators in making quick, reliable and actionable decisions without drowning in the data lake. The work contributes to the wider efforts of the pipeline industry to fully realize the benefits of 'big data'.

Managing corrosion in hydrocarbon production systems without frequent direct techniques such as ILI pigging has been a challenge in the Oil & Gas industry for years. Traditionally, corrosion estimation is based on numerous industry-accepted corrosion models which are reliant on input parameters such as pressure, temperature, CO2/H2S/water content, flowrates, etc. While these models are quite fundamental and dependent on the actual chemistry of the fluid; however, at times these models underperform when compared with actual field data.
With increasing focus on measuring and storing vast amounts of data generated, the use of Machine Learning in Industry has been gaining more traction by the day due to its mathematically elaborate framework in determining critical parameters and estimating the vital variables.  Machine Learning (ML) techniques have been used to predict pipeline condition and leakage in the past.
This paper presents the results of a study of using cutting-edge ML techniques on 200+ flowlines in the prediction of corrosion risks in handling multiphase flow. Very good results are observed in predicting the associated corrosion risk and is a valuable tool in aiding the pipeline integrity assessment with increased accuracy and validation.  This paper is designed to share those valuable insights and lessons learned.

The Pipeline and Hazardous Materials Safety Administration (PHMSA) has proposed new requirements for verification of pipeline materials where traceable, verifiable, and complete records do not exist. This verification requires knowledge of material properties, including strength and chemical composition, and PG&E is evaluating the use of nondestructive evaluation (NDE) methodologies for determination of these properties.  A widely used NDE method for determining the chemical composition of pipe materials in the field (in situ) is spark optical emission spectroscopy (OES).  However, those results can be compromised by contamination, improper experimental setup, and/or environmental conditions.  This work evaluates the alternate approach of using lab-based (ex situ) elemental analysis on metal filings collected nondestructively from the pipe surface.  During the testing program, metal filings collected from over forty pipes and fittings were analyzed by atomic absorption spectroscopy (AAS) and inductively coupled plasma (ICP) methods. These results were compared with those from spark OES performed both in situ on the pipe OD, and ex situ on coupons removed (destructively) from the pipe body. This paper presents findings to date and discusses areas of opportunity for improved understanding of the analysis of chemical composition of pipelines from filings and discusses the potential significance of differences in the results from different analysis methods.

Fracture models typically assume an ideal crack profile, such as a semi-elliptical shape.  Actual crack-like anomalies in pipelines usually have an irregular profile, however.  Common practice is to approximate such profiles with a semi-elliptical crack whose length is equal to the total flaw length and whose depth corresponds to the maximum depth of the flaw.  This approach is very conservative and can result in significant underestimates of burst pressure.  Some pipeline fracture models account for the crack profile with variations of the B31G effective area method.  which is suitable for metal loss flaws that fail by ductile rupture, but it is not appropriate for cracks that fail in a brittle manner.
This paper describes recent improvements to the PRCI MAT-8 fracture model that account for arbitrary crack profiles.  An extensive finite element study was undertaken to model numerous non-ideal longitudinal crack profiles in pipe joints.  The output from these analyses led to a procedure to convert an arbitrary crack profile into an equivalent semi-elliptical crack.  The depth of the equivalent crack is equal to the maximum flaw depth, but the length is typically much less than the total flaw length.  For example, given a 16-inch long anomaly with significant variations in depth, the effective flaw length may be in the range of 1 to 2 inches.  This modification results in burst pressure estimates that are more accurate and less conservative than the traditional approach.

Within Part 192, if tensile properties are unknown, an operator may perform destructive tensile testing. The sampling requirements for this testing are established within Part 192 Appendix B. While the final language is in flux, in 2016 PHMSA proposed adding federal rules requiring field verification and testing to establish material properties for insufficiently documented pipe segments. PHMSA has also proposed rules that would create an allowance for operators to utilize in-situ NDE technologies in lieu of destructive testing to verify tensile properties. Development of a material verification program needs to balance the uncertainty of any destructive or NDE method to verify material properties, with the cost of performing the required sampling. The authors of this study contend that a Bayesian approach allows for the establishment of such a technique. Bayesian inference allows the updating of a prior probability as more data is collected to arrive to a posterior probability. This approach can be utilized to establish a practicable sampling program for both the minimum number of joints within a population and the minimum number of tests required per joint.

Selective seam weld corrosion (SSWC) is localized corrosion of the weld bondline of pipe.  Typically found in pre-1970s pipelines manufactured using low-frequency electric resistance welding or electric flash welding, SSWC is affected primarily by the degree of exposure to corrosive conditions (i.e., poor or absent coating and ineffective cathodic protection) and by the nature of non-metallic inclusions in the bondline region.  Therefore, it can also be found in older vintage HF-ERW pipe manufactured pipe skelp produced by ingot casting method.  SSWC leads to the development of a long, wedge-shaped groove within an area of corrosion crossing the long seam (CCLS). This can make pipelines more susceptible to leaks and other failures, including those related to overpressure events. In the United States, PHMSA 49 CFR §195.452 (h)(4)(iii)(H) mandates that an operator must schedule evaluation and remediation of corrosion of or along a longitudinal seam weld within 180 days of discovery of the condition.  So, although the failure mode due to SSWC is low, regulations drives early identification and repair or replacement of the affected pipeline section.
A recent Pipeline and Hazardous Material Safety Administration (PHMSA) project, Research Announcement #DTPH56-13-RA-000001 Topic 9 – Technology – “Improve and Develop ILI Tools to Locate, Size, and Quantify Complex/Interacting Metal Loss Features and Dents”, solicited technology development for improving anomaly detection on carbon steel transmission and distribution pipelines. Specifically, PHMSA sought to improve sizing capability for complex corrosion and interacting defects along the long seam through the development and validation of new ILI tools.  T.D. Williamson partnered with Kiefner and Associates on a project that resulted in advancing ILI systems and analysis techniques. The intended outcomes of this work — improved detection, sizing and prioritization of SSWC — would enable pipeline operators to prioritize SSWC within the required 180-day timeframe.
This presentation will outline results from the SSWC research pertaining to the PHMSA R&D project, discuss how the SSWC process has been operationalized from an ILI perspective, and show validated field results from the SSWC management process.

The ultrasonic crack technology is based on Pulse-Echo transducers for liquid lines. This technique uses a piezo electric transducer generating a 45° shear wave. For the past 20 years, operators have seen only minor developments and improvements from the technology. The most significant improvement in the last years has been moving from bucket sizing to absolute depth sizing. Pulse-Echo limitations have remained for some defect types which are mainly caused due to geometric limitations, e.g. hook cracks where the response of the ultrasonic echo has a direct relationship with the geometry of the flaw.
Since the introduction of absolute depth sizing, the Pulse-Echo technique has had a limit of 0.160in as maximum depth for a feature; at lower depths, the depth accuracy has been fixed to +/- 0.040in. Such limitations are not sufficient for operators today and has resulted in a need to push ILI providers to improve. By using high-resolution robots, specialized measurement, and sizing methodologies, recent developments from NDT Global have been able to address these two limitations and open the door for further improvements.
With this paper, the authors wills discuss the technological limitations from Pulse-Echo along with the process including research using sophisticated modeling and simulation environments, small scale laboratory testing, and large-scale testing in NDT Global's testing facilities; all in accordance with API1163 to address the technique's limitations. Furthermore, the authors will share results from commercial high-resolution (UCx) inspections for detection, identification and sizing.
Finally, the authors will present two scenarios where high-resolution inspections have helped to address interacting cracks and deformation features as well as data patterns concerning proper identification of hook cracks.

The pipeline industry has put forth tremendous efforts to mitigate crack-related integrity concerns. However, further investigations are needed to fill the existing gaps. In this paper, a few topics are selected for discussion based on their relevance and lack of adequate solutions, including seam type identification by visual and NDE methods, SCC growth rate and mechanism, selective seam weld corrosion properties and interpretation. The contents presented in this paper will provide guidance and up-to-date knowledge of these defect mitigations

This paper identifies several common pitfalls that operators must avoid when assessing seam-weld integrity or implementing an effective seam-weld integrity management plan (IMP). The lessons learned presented in this paper will educate operators on how to employ best practices when managing seam-weld integrity.
The common seam-weld assessment pitfalls addressed in this paper are:

The information presented in this paper is intended to educate operators on areas for improvement. Operators should be mindful of these lessons learned and ensure that their current approach for maintaining seam weld integrity takes this information into account.

Selecting the appropriate assessment method for a pipeline system requires an understanding of the pipeline segment, the potential threats to the pipeline segment, and the various assessment methods and technologies available along with their performance capabilities.  Each assessment method has its own set of advantages and disadvantages depending on the pipeline and hreat being assessed.
In most cases, the assessment of a pieline segment using one method is sufficient for the management of that threat or threats.  However, for high risk pipelines (driven by likelihood and consequence, or both), this paper explores the ptentential benefits of leveraging different assessment methods by examining a number of scenarios with a specific focus on the management of cracks within a pipeline segment.  It looks at the benefits of mulitple assessment methods employed at the same intervals but in varying order.  It also investigates the impact of in-line inspection performance, excavation criteria, and spike test pressures.

High Frequency Electric Resistance Welded pipe (HF-ERW) has been around for more than 40 years with a reputation of high toughness and fewer seam weld defects than Low Frequency (LF) ERW pre-1970s. However, both High frequency ERW pre-1980s and low frequency ERW pre-1970s are susceptible, under the right conditions, to develop hook cracks at seam welds due to a high content of sulfur in the pre 1980s steel. In addition, LF-ERW welded long seams have experienced in-service failures due to lack of fusion. These issues are well known and understood and have been a component of pipeline integrity management for a number of years.
This paper describes a supplemental screening process that operators can perform as part of their due diligence. The process uses ultrasonic crack detection data and the vendor display software to identify and prioritize potential longitudinal seam weld anomalies, specifically focusing on lack of fusion and hook cracks that may be associated with ERW long seam welds. The result of the process delivers a ranking of the anomalies to help operator prioritize Integrity Management efforts and resources

Within the Oil and Gas pipeline industry, mechanical damage has been identified as one of the most significant threats due to its complexity and unpredictability. In most situations, mechanical damage in pipeline systems results in coating removal, metal removal, cracking, or a combination thereof. In addition, mechanical damage may introduce a dent in the pipe which can limit the ability of In-Line Inspection (ILI) tools to detect and properly characterize features within the limits of the dented area.
Mechanical damage has been of great interest for pipeline operators, ILI companies and regulatory bodies. In 2013, the Pipeline and Hazardous Materials Safety Administration (PHMSA) funded a research project, Topic 9 – Mechanical Damage, with the goal of improving the characterization of defects from ILI data to locate, size and quantify complex or interacting features with dents on carbon steel transmission pipelines. The PHMSA project utilized a multi-data ILI technology to evaluate the characterization of metal loss associated with a dent. Moreover, they assessed the performance of a classifier that distinguishes dents with gouges from dents with corrosion. Through ILI pull through testing on manufactured mechanical damage anomalies, the projects showed that the classifier of such technology was able to discriminate gouges from corrosion features what were associated with dents with 95% effectiveness. In this paper the performance evaluation of the multi-data ILI technology used in the PHMSA project was completed under actual operating conditions and through field non-destructive examination (NDE) of reported features. Also, the multi-data ILI system was compared to other ILI technologies that are commonly used in the management of mechanical damage. The results of this work were integrated into a prioritization process to define an important element for mechanical damage: the ability to distinguish a dent with gouging from other dents such as plains dents and dents with coincident metal loss.

The Pipeline and Hazardous Materials Safety Administration (PHMSA) has proposed new requirements for the regulations governing gas transmission pipelines. The proposed changes will require operators to demonstrate that material documentation records for line pipe and fittings are traceable, verifiable, and complete (TVC). In the absence of TVC records, operators will be permitted to use nondestructive evaluation (NDE) techniques to verify material properties. However, PHMSA recommends that NDE technologies are independently validated, and the data interpreted to conservatively account for uncertainties. Pertinent material properties for pipeline materials include yield and ultimate tensile strength (UTS), fracture toughness, and chemical composition. This paper presents an evaluation of instrumented indentation testing (IIT) for measurement of yield and UTS on pipe features in situ. During this evaluation, over 100 different pipeline features were tested by IIT, and the results compared with data from destructive tensile testing. Almost half of the features were installed before 1970, thus making them particularly relevant to the types of features that will be most affected by the proposed rule change. The testing program includes evaluation of the sources of uncertainty in IIT measurements, and investigation of alternative algorithms for converting the raw IIT data into yield and tensile strength.

The capability for assessment of mechanical damage heavily depends on the information provided by In-Line Inspection surveys. Caliper and Geometry tools are commonly used for classification, sizing and detection of deformation features; however, this ILI technology does not fully characterize mechanical damage. In order to characterize mechanical damage, information provided by other ILI technologies has to be incorporated into the decision-making process.
This paper presents a methodology that allows the compositing of multi-technology ILI surveys into a single and enhanced “Master Geometry Inventory” ILI data set. This work also shows how the composite survey has been used to characterize deformation features that are suspected to be mechanical damage, systematically identify new mechanical damage features as well as facilitating the Threat Integration of secondary features (Crack-like features, stress corrosion cracking (SCC) features, and Metal loss feature) from multiple datasets. The multi-technology data set was used to develop a prioritization approach for suspected mechanical damage features for repair based on ILI surveys, repeatability and threat integration. ILI data of two different pipeline systems were used to validate the algorithm. The composite multi-technology ILI data includes: high resolution caliper, ultrasonic crack detection, ultrasonic wall measurement, axial magnetic flux leakages, circumferential magnetic flux leakage and electromagnetic crack detection technologies. This paper also outlines the overall mechanical damage characterization approach, its automation, and also presents how the enhanced ILI data correlated with non-destructive examination (NDE) results. Underlying the approach are the demonstrated benefits and advantages to the management and storage of ILI data in a single repository, providing the accessibility to the data required for an automated solution.
The leveraging and integration of historical ILI data sets from other technology types is a significant advancement from the traditional analysis based on a singular ILI run data set from the most recent ILI tool run. By maintaining a definitive master list of geometry features with enhanced feature characterizations, a more accurate and comprehensive assessment can be conducted and geometric features can be better managed and tracked.

U.S. federal regulation 49 CFR 192/195 requires repair of dents greater than 6% of the pipeline's outer diameter, as well as any dent with metal loss. Pipeline failures have occurred from dents that pass these regulations. For example, dents with depths shallower than 3% may result in pipeline failure due to fatigue. Conversely, dents interacting with coincidental metal loss have been known to survive for extended service.
Technology that records multiple data sets using a single in-line inspection device has shown capability in detecting low level mechanical damage interacting with dents less than 1%, with the results verified by dig feedback. The technology - which uses an oblique high field magnetizer, an axial high field magnetizer, an axial low field magnetizer, and a high-resolution geometry module - has also discriminated metal loss signatures indicative of gouging from corrosion and identified dents that have been rerounded from those that are constrained. These factors are all important to integrity decision making.
This paper evaluates the multiple data set tool's detection and sizing accuracy against the in-the-ditch NDE results. Specifically, it compares the capabilities of the tool versus NDE on dents below 1% depth with associated gouging or metal loss, and on constrained and rerounded dents.

Pipelines frequently contain large numbers of dents. In the last 10 years the number of shallow dents reported by in-line inspections has increased, as tools have become more sensitive. These dents can fail at the time they are formed, or later under fatigue loading. However, the great majority of dents do not fail. How do we determine which dents need to be repaired?
For some pipeline defects, such as corrosion, we have accurate assessment methods based on simple parameters. Similarly simple methods have historically been used to assess dents, based solely on dent depth. Indeed, the regulations in some parts of the world (including North America) mandate this approach. However, the reliability of these methods is poor, so large safety factors are used and assessments can be excessively conservative.
More recently, dent assessment methods based on strain and finite element analysis have been adopted by many pipeline operators. These methods can be more accurate, but they must be applied correctly to avoid non-conservative results. This paper discusses several potential pitfalls in applying these methods, including some that have appeared in published work. The important factors which need to be considered in a dent assessment are discussed, along with the relative conservatism of different assessment approaches.

Stress corrosion cracking (SCC) is a problematic phenomenon in line pipe as detection by many MFL inline inspection devices is not possible or has such high thresholds of detection it is insufficient to adequately identify threats before the next inspection cycle. In this paper we investigate the use of ultra-high frequency MFL sampling using Hall effect sensors to detect and characterize circumferentially oriented stress corrosion cracking (C-SCC). In pull tests, an axial flux direction magnetizer is used with 2.1 mm (0.083 in.) circumferential sensor spacing and sampling rates up to 4,000 Hz. Minimum detection criteria are estimated using excavated line pipe containing C-SCC, supported by laboratory testing and measurements. This inspection tool configuration proved capable of detecting in a small sample set circumferentially oriented SCC as small as 0.5 inches (12.5 mm) in length and 10% deep. Sizing tests were conducted yielding 80% depth certainty tolerances consistently less than ±20% of wall thickness and 80% circumferential width tolerances consistently less than ±0.6 inches (15 mm).

Circumferentially orientated stress corrosion cracking (referred to as CSCC or transverse SCC) is a variation of transgranular SCC.  When CSCC occurs the principle stress acting on the crack is typically an axial bending stress.  This stress is different from the normal circumferential hoop stress due to internal pressure and may be due to ground instabilities beneath the pipeline or ground movement.  Construction practices can also induce axial stresses in the pipe; this may occur where the pipe was forced into alignment for welding at tie-in locations, e.g., at crossings or field bends or at other locations subjected to high external loads, e.g., road crossings.  The parameters related to CSCC are much the same as with the low pH form of SCC with exception of the source of the principle stress.  High hoop stress due to internal pressure is not a prerequisite for CSCC but the presence of axial stress is.
The availability and use of In-line inspection (ILI) tools to reliably detect CSCC is nowhere near as advanced or industry accepted as it is for axial cracking.  Hence, for most operators an alternative approach for assessing the susceptibility of a pipeline to this threat and prioritizing pipeline segments for evaluation is required.
This paper describes an assessment protocol for assessing and prioritizing the segments of a pipeline for susceptibility to CSCC.  Our approach has been based on experiences of working with several pipeline operators who have found CSCC on their networks.  The protocol consists of a set of ILI detectable conditions augmented by other known parameters that together would indicate a potential CSCC threat.  The assessment protocol identifies the pipeline segments with susceptibility to CSCC based on the number of threat indicators that occur in the same location.  The threat indicators, taken in isolation, may not all present a susceptibility to CSCC, however, where two or more of these indicators occur coincidently the susceptibility of CSCC will be greater.  The presence of multiple indicators is used to prioritize the pipeline segments with the highest CSCC threats.    Based on the combination of certain CSCC threat indicators the pipeline segments are prioritized for further review or field investigation. Examples of the application of the CSCC threat assessment are provided in the paper.

The program described in this paper utilized full-scale testing and metallurgical analysis to compare predicted failure pressures of stress corrosion cracking (SCC) colonies using both in-line inspection (ILI) and in-the-ditch inspection techniques to the actual performance results of full-scale cyclic pressure and burst testing. Testing was completed using 10.75-inch x 0.188-inch, Gr. X52 high-frequency electric resistance welded (HF-ERW) pipe that had been removed from service. Imaging software was used to generate profiles of the SCC flaws from laboratory break-open samples. This provided a more accurate representation of the length and depth of individual cracks within the SCC.
Prior to pressure testing, the provided pipe samples were non-destructively inspected using ILI, phased-array ultrasonic (PAUT), and IWEX techniques. All techniques identified significant SCC with reported flaw depths up to 88% of the pipe wall thickness. It was found that depths reported by PAUT were typically within +/- 20% of the depths measured during metallurgical examination. The minimum failure pressure was observed to be at an internal pressure equal to 103% SMYS. The maximum failure pressure occurred at an internal pressure equal to 154% SMYS. None of the samples tested failed below failure pressures predicted by the PAUT vendors. Additionally, the failure pressures were compared to failure predictions calculated using flaw depth profiles from the post-test metallurgical analysis.