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

February 17-18 February 19-21 February 18-20


Tuesday, February 18
17:00-19:00 Exhibition welcome reception sponsored by ROSEN Group


Wednesday, February 19
  1.0 Plenary opening session 
8:00 Opening remarks
8:15 [1] Utilization of vendor auditing for continuous improvement of ILI system performance
Juan Martinez, Karl Augustus, Justin Harkrader Colonial Pipeline Co., Alpharetta, GA; Tara Podnar McMahan, John Godfrey, DNV GL, Dublin, OH, USA
8:45 Young Pipeline Professionals Annual Recognition Award presentations
9:00 [2] Beyond compliance...
Jan Fowijn and Chris Yoxall, ROSEN USA, Houston, TX, USA
9:30 [3] The genesis and applications of API Standard 1163, 'Inline Inspection Systems Qualification Standard'
Bryan Melan, Tide Water Integrity Services LLC, Georgetown, TX, USA
10:00 Refreshment break sponsored by Halfwave
11:00 [4] A process for hydrostatic test data acquisition and validation
Gary Zunkel, Bluefin, Ames, IA, USA
11:30 [5] A review of nondestructive technologies and statistics-driven approaches for estimating toughness
Dr John Kiefner, Michael Rosenfeld, Adam Steiner, RSI Pipeline Solutions, New Albany, OH; Troy Rovella, PG&E, San Ramon, CA; and Dr Peter Veloo, Exponent, Los Angeles, CA, USA
12:00 [6] Pig and dig: how much is enough, how much is too much?
Joel Anderson, Enable Midstream Partners, Oklahoma City, OK; W. Kent Muhlbauer, WKM Consultancy, Austin, TX; Justin Raimondi, Mistras-Integrity Plus, Fort Collins, CO, USA
12:30 Lunch sponsored by Enduro
  2.1 Cracks  3.1 ILI applications  4.1 Machine learning 
14:00 [7] Black swans, red he rrings and sacred cows: the zoology of pipeline cracks
Dr Ted Anderson, TL Anderson Consulting, Longmont, CO, USA
[30] Calculation of a laser-scan-like 3D defect profile from conventional MFL data
Johannes Palmer, Chris Yoxall, ROSEN USA, Houston, TX, USA; Andrey Danilov, ROSEN Group, Berlin, Germany
[53] Risk‐based decision‐making supported by machine learning
Michael Gloven, Expert Infrastructure Solutions (EIS), Denver, CO, USA
14:30 [8] Next generation ILI crack inspection service – an operator vendor collaboration for a 26-inch pipeline
Dr Thomas Hennig, Thorsten Sickinger, Dr Michael Haas, Rogelio Guajardo, NDT Global, Stutensee, Germany; Steven Bott, Rob MacKenzie, Mike Hill, Owen Burdek, Enbridge Pipelines, Edmonton, AB, Canada
[31] A case study – benefit of pipeline specific sizing for engineering criticality assessment
Genevieve Stilwell, Enbridge, Houston, TX, USA; Lyndon Lamborn, Steven Bott, Enbridge, Edmonton, AB, Canada; Stephan Tappert, Holger Charbon, Baker Hughes, Stutensee, Baden-Württemberg, Germany
[54] The current progeny of inline inspection machine learning
Dane Burden, Paul Dalfonso, TD Williamson, Salt Lake City, UT, USA; Adrian Belanger, TD Williamson, Houston, TX, USA
15:00 [9] Measuring fatigue crack growth rates of pipeline steels and their implications to fatigue analysis
Sergio Limon, Jacob Bradford, Elevara Partners, Salt Lake City, UT; Alex Litvinov, Jim Harter, LexTech, Centerville, OH, USA
[32] Replacing hydrotesting of low frequency ERW pipe with an enhanced ILI solution – Eclipse
Dr Thomas Hennig, Rogelio Guajardo, NDT Global, Stutensee, Germany; Marcus Le Roy, Marathon Pipelines, Findlay, OH; Santiago Urrea, NDT Global Inc, Houston, TX, USA
[55] Machine learning to support risk and integrity management
Christopher De Leon and Michael Smith, ROSEN USA, Houston, TX, USA
15:30 Refreshment break sponsored by Halfwave
  2.2 Cracks (cont'd) 3.2 ILI applications (cont'd)  5.1 Data management & leveraging 
16:30 [10] Generation and monitoring of synthetic crack-like features in pipeline materials using cyclic pressure loading
Dr Chris Alexander, Jon Rickert, ADV Integrity, Inc., Magnolia, TX; Rhett Dotson, Felipe Freitas, Simon Slater, Christopher De Leon, ROSEN USA, Houston, TX, USA
[33] Improving remediation programs with risk-based analyses of inline inspection data
Benjamin Hanna, Dr Thomas Bubenik, William Harper, DNV GL, Columbus, OH, USA
[56] Data management strategies for an efficient implementation of Industry 4.0 integrity assessments
Aidan Charlton and John Healy, Penspen Ltd, Newcastle, UK, Gustavo Romero, Penspen, Latin America
17:00 [11] Understanding the detection capabilities of an ultrasonic crack ILI robot in a dent
Rogelio Guajardo Rodriguez, Dr Thomas Hennig, NDT Global, Karlsruhe, Germany; Carlota Mendez, Beatriz Tarramera, NDT GDAC Spain S.L., Barcelona, Spain
[34] Closed form probabilistic method for conducting pipeline remaining life assessment from ILI data
Jocelyn Nelson, ExxonMobil Upstream Research Company, Spring, TX; Preston Smith, Pamela Tanner, ExxonMobil Upstream Oil and Gas Company, Spring, TX, USA; Abhed V., ExxonMobil Upstream Integrated Solutions Company, Bangalore, India
[57] Pipeline GIS and mega rule part 1: material property verification and the system of record
Christopher De Leon, ROSEN USA, Houston, TX, USA; Simon Slater, ROSEN USA, Columbus, OH, USA
17:00-19:00 Exhibition reception sponsored by Intero
17:30 Conference day 1 concludes


Thursday, February 20
  2.3 Cracks (cont'd) 6.1 ILI analysis 7.1 Practical pigging
8:00 [12] Burst testing of pipes containing stress corrosion cracking
Simon Slater, Chris Davies, ROSEN USA, Columbus, OH; Dr Chris Alexander, Chantz Denowh, ADV Integrity, Magnolia, TX; Todd Post, Consumers Energy, Jackson, MI, USA
[35] RunComs and randomness
Joel Anderson, Enable Midstream Partners, Oklahoma City, OK, USA
[58] Fast-track development and deployment of high-temperature ILI tools
Nader Alhamalawi and Basil Hostage, 3P Services, Wietmarschen / Lohne, Germany
8:30 [13] A validation study of computed tomography inspection technology using full-scale test articles with crack-like features
Dr Chris Alexander, Jon Rickert, ADV Integrity, Inc., Magnolia, TX; James Medford, Inspection Associates, Inc., Magnolia, TX, USA
[36] Maximizing the accuracy of MFL pipeline inspection
Ben Scott, Baker Hughes, Cramlington, UK
[59] Behind closed doors; pipeline closures and traps revisited
Dr Mike Kirkwood, T.D. Williamson, Dubai, UAE, Alan Morton, T.D. Williamson, Tulsa, OK, USA
9:00 [14] Pipeline pressure analysis in the frequency domain
Michael Rosenfeld and Dr Benjamin Zand, RSI Pipeline Solutions LLC, New Albany, OH, USA
[37] Individual anomaly sizing certainty on MFL Data
Christoph Hermes, Niklas Wilming, Viktor Reimer, Markus Ginten, ROSEN Group, Lingen, Germany
[60] Pipeline change of service – from dirty to clean in 5 easy steps
Mark Gourley, Baker Hughes, Houston, TX, and Brett Roper, Energy Transfer, Tulsa, OK, USA
9:30 Refreshment break sponsored by Halfwave
  8.1 Circumerential cracks 6.2 ILI analysis (cont'd) 7.2 Practical pigging (cont'd)
10:30 [15] A risk-based approach to circumferential stress corrosion cracking
Mark Wright, ROSEN USA, Houston, TX, USA
[38] ILI run-to-run comparison and corrosion growth screening case study
Matt Ellinger, Eric Graf, Stacy Gibson, DNV GL, Dublin, OH; Pam Moreno, DNV GL, Katy, TX, USA
[61] Ready, aim, pig!
Dr Mike Kirkwood, T.D. Williamson, Dubai, UAE and Alan Morton, T.D. Williamson, Tulsa, OK, USA
11:00 [16] Detection of non-axial stress corrosion cracking (SCC) using MFL technology
Matthew Romney, Dane Burden, T.D. Williamson, Salt Lake City, UT, USA
[39] Leveraging ILI comparative analysis to accurately determine corrosion growth in pipelines
Lisa Barkdull, Matthew Lewis and Josiah SooTot, Quest Integrity, Houston, TX, USA
[62] Stuck pig recovery
Gary Anderson and Rolf Gunnar Lie, T.D. Williamson, Whitley Bay, UK
11:30 [17] A case study in the detection and sizing of circumferential stress corrosion cracking
Ron Thompson, James Hare, Novitech Inc, Vaughan, ON, Canada; Ray Gardner, Katrina Dwyer, Xcel Energy, Denver, CO, USA
[40] A fistful of data
Jonathan Martin, Roland Palmer-Jones and Michael Smith, ROSEN Group, Newcastle, UK
[63] Cost-effective ultrasonic inspection of large diameter pipelines: technology update and case study
Ries Augustijn, Intero Integrity Services, Tricht, The Netherlands
12:00 Lunch
  9.1 ILI verification 10.1 Codes and regulations 11.1 NDE (other than ILI verification)
13:30 [18] Corrosion under insulation in pipelines, avoiding pitfalls when inspecting with MFL or ultrasonic ILI tools
Bernardo Cuervo and Mark McQueen, G2 Integrated Solutions, Houston, TX, USA
[41] The Gas Mega Rule: are you ready?
Dr Mike Kirkwood, T.D. Williamson, Dubai, UAE and Jeff Wiese, TRC, Reston, VA, USA
[64] Enhancements to a prototype grade calculation algorithm through data analytics
Eduardo Munoz, Dr Arash Kamari, Kiefner and Associates, Houston, TX; Troy Rovella, PG&E, San Ramon, CA; Dr Peter Veloo, Exponent, Los Angeles, CA, USA
14:00 [19] Considerations for validating an ILI technology: controlled implementation of a new MFL product
Brett Conrad, TC Energy, Calgary, AB, Canada and McKenzie Kissel, Onstream Pipeline Inspection, Calgary, AB, Canada
[42] 192 final rule - impact on crack fatigue growth and pressure test reassessment intervals
Matt Ellinger, Shanshan Wu, Dr Tom Bubenik, Zach Booth, DNV GL, Dublin, OH, USA
[65] Integrated material property verification - combining state of the art ILI and in-ditch testing
Simon Slater, ROSEN USA, Columbus, OH, USA, Christopher De Leon, ROSEN USA, Houston, TX, USA; Dr Simon Bellmare, Dr Steven Palkovic, Massachusetts Materials Technologies, Waltham, MA, USA
14:30 [20] Improving inline inspection performance with dig feedback
Tod Barker, Adrian Belanger, T.D. Williamson, Salt Lake City, UT, USA
[43] Case study on next generation capability
Daryl Brister, Shea Capability & Compliance Solutions, Houston, TX, USA
[66] Influence of line-pipe steel microstructure on NDE yield strength predictive capabilities
Dr Nathan Switzner, Dr Solver Thorsson, Dr Jeffrey Kornuta, Dr Peter Veloo, Exponent, Los Angeles, CA; Dr Peter Martin, Troy Rovella, PG&E, San Ramon, CA; Michael Rosenfeld, RSI Pipeline Solutions, Columbus, OH, USA
15:00 Refreshment break sponsored by Halfwave
  9.2 ILI verification (cont'd) 12.1 Unpiggable pipelines 11.2 NDE (other than ILI verification) (cont'd)
15:30 [21] Validating an ILI run with and without data loss
Ahmad Al Saif, Ayman Janbi Saudi Aramco, Dammam, Saudi Arabia
[44] Inspection of small-diameter, difficult-to-pig pipelines using MFL technology
Ken Maxfield, KMax Inspection, Salt Lake City, UT, USA
[67] Nondestructive testing of pipeline materials: further evaluation of portable OES, XRF, LIBS, and filings to estimate chemical composition
Dr Monty Liong, Dr Nathan Switzner, Dr Peter Veloo, Exponent, Los Angeles, CA; Melissa Gould, DNV GL, Yorba Linda, CA; Dr Peter Martin, Troy Rovella, PG&E, San Ramon, CA, USA
16:00 [22] Full-scale destructive testing and metallurgical evaluation to quantify inspection performance and validate remaining life predictions
Michael Turnquist, Quest Integrity, Houston, TX, USA
[45] Interpretation of local inspection data for unpiggable pipelines – understanding the meaning of nothing
Susannah Turner, Fraser Gray, Highgrade Associates Ltd, Newcastle, UK
[68] A statistical approach to material verification of expected grade through opportunistic field measurements
Dr Simon Bellemare, Dr Steven Palkovic, Parth Patel, Soheil Safari, Massachusetts Materials Technologies, Waltham, MA, USA
16:30 [23] Accounting for a half-century of pipeline materials to enhance AXISS™ ILI axial strain measurement accuracy
Mohamed El Seify, Baker Hughes, Calgary, AB, Canada; Raymond Kare, Sylvain Cornu, Eddyfi Technologies, Milton Keynes, UK
[46] Do we require gauge pigs for difficult pipelines? (To gauge or not to gauge?)
Geert Bontekoe, Quest Integrity, Houston, TX, USA
[69] Nondestructive evaluation of strength, toughness and residual stress
Dongil Kwon, Seunghun Choi, Seoul National University, Seoul, South Korea; Dongseong Ro, Frontics America, Schaumburg, IL, USA; Kwangho Kim, Frontics, Seoul, South Korea
17:00 Conference day 2 concludes


Friday, February 21
  13.1 Repair 12.2 Unpiggable pipelines (cont'd) 14.1 Engineering assessment
8:00 [24] Full encirclement engineered laminated steel sleeve system: engineering development and test results
Shawn Laughlin, Pipe Spring LLC, The Woodlands, TX, USA
[47] Tank farm pipeline inspection program using robotics
Greg Herbstritt, Juan Martinez, Todd McClellan, Colonial Pipeline Co., Alpharetta, GA, USA; Paul Laursen, Pipetel Technologies, Toronto, ON, Canada
[70] Using deep learning to identify the severity of pipeline dents
Ishita Chakraborty and Brent Vyvial, Stress Engineering Services, Houston, TX, USA
8:30 [25] An in-service welding and hot tapping competency framework
Tran Mah-Paulson, T.D. Williamson, Edmonton, AB, Canada
[48] Terminal pipework – spot checking or checking for spots?
Danny Molyneux, Quest Integrity, Aberdeen, UK
[71] Changes to in-line inspections practices to support the eca option for extending response time
Bruce Nestleroth, Kiefner Applus, Columbus, OH and Michael Rosenfeld, RSI Pipeline Solutions, New Albany, OH, USA
9:00 [26] Statistical analysis of dig operations leading to productive repairs
Jordan Dubuc, Dr Yevgeniy Petrov, Tim Edward, Michael Murray, OneBridge Solutions Inc., Boise, ID, USA
[49] Statistical methods for planning and evaluation of inspection on unpiggable pipelines
Mark Stone, Sonomatic, Inc., Katy, TX, USA
[72] Quantifying the effects of interaction criteria on ILI defect assessments
Dr Tom Bubenik and Steven Polasik, DNV GL, Dublin, OH, USA; Jason Cramer, Dominion Energy Ohio, North Canton, OH, USA
9:30 Refreshment break
  13.2 Repair (cont'd) 12.3 Unpiggable pipelines (cont'd) 14.2 Engineering assessment (cont'd)
10:00 [27] Use of Full-scale Testing as a Means for Managing Pipeline Integrity
Dr Chris Alexander, Chantz Denowh, ADV Integrity, Inc., Waller, TX
[50] A comparison between Pipers® and MFL pipeline data
Zachary Shand, Anouk van Pol, John van Pol, Ingu Solutions Inc., Calgary, AB, Canada; Shelise Berteig, Whitecap Resources Inc., Calgary, AB, Canada
[73] Improved methods for sizing metal loss in dents for ECA
Rhett Dotson, Fernando Curiel, Luis Sacramento, ROSEN USA, Houston, TX, USA; Jacob Duska, Zach Locks, TC Energy, Charleston, WV, USA
10:30 [28] Life extension of pipelines with crack-like defects using composite repairs
Dr Paul Hill, Team Industrial Services Inc, Kendal, UK; Dr Troy Swankie, DNV GL, Loughborough, UK, Tim Webley, Cadent Gas Ltd, Birmingham, UK
[51] Pipeline integrity management of unpiggable pipelines – a case history
Dr Arash Ilbagi, Ahmad Saab, Frank Gareau, Acuren, Calgary, AB, Canada
[74] Analysis of the properties of vintage girth welds
Timothy Rudd, Valero Logistics UK Ltd, Milford Haven, UK; Kalen Jansen, ATCO Pipelines & Liquids, Edmonton, AB, Canada; Kirsty McDermott, National Grid, Warwick, UK; Dr Andrew Cosham, Ninth Planet Engineering Ltd, Newcastle, UK
11:00 [29] Analyzing the use of standard inspection tools on composite repairs
Aleese Post, Casey Whalen, ClockSpring|NRI, Houston, TX; Colton Sheets, Stress Engineering Services, Houston, TX, USA
[52] Ensuring effective compliance of external corrosion direct assessment
Rod Rheaume, HMI Technical Solutions, LLC, Johnston, RI, USA, Alfred Giordano, HMI Technical Solutions, Houston, TX and Michael (Alex) Bryant, TC Energy, Charleston, WV, USA
[75] Advancements in records verification, class and consequence tools for the new gas and liquids rules
Mauricio Palomino, G2 Integrated Solutions, Houston, TX, USA
11:30 Conference concludes


 Platinum Elite Sponsor


 Platinum Sponsors

Intero Integrity Services     Enduro   TD Williamson

Gold Sponsor


 Silver Sponsors

   N-Spec     Q-Inline      Circor Energy - Pipeline Engineering.png Argus

 Organized by:

Clarion Technical Conferences     Great Southern Press

Supported by:

   North American Pipelines  Pipeline & Gas Journal       Inspectioneering     LatincorrPipelines International     Oil and Gas Journal


A decline in first run success rate of a pipeline operator’s integrity management program triggered an initiative to improve the execution and performance of its in-line inspection (ILI) program.  As a result of the initiative, a two-step program was implemented to target continuous improvement, and ensure clear communication of its expectations through a comprehensive revision to its ILI specification.  Within the revised ILI specification, roles, responsibilities, reporting, and performance expectations were clearly defined.  It also specified a consistent set of performance metrics for each ILI survey performed.  The second step was to meet with ILI vendors for the purpose of a comprehensive review of their performance, processes, procedures, and quality programs using a systematic plan.  The purpose of the review was to create opportunities for collaborative continuous improvement.  Direct feedback and improvement opportunities were identified and discussed.  As a result of the two-step program, the pipeline operator is experiencing improved first run success rates as well as other positive results.  This paper will detail the key elements of the revised ILI specification designed to drive continual improvement and outline the systematic process used to review and provide feedback on ILI vendor performance. Positive results experienced thus far will be summarized.


It is clear that we need to invest in energy efficiency and renewables. At the same time, most predictions show that oil and natural gas are an important part of our energy future. In the U.S. oil and natural gas production is soaring and there is no credible way to transport oil and natural gas without pipelines. We think to believe that we do all we can to protect communities and the environment with safe pipeline operations: beyond compliance. Was it only just that simple. The energy debate raises deep controversies. For those in the industry it may often seem that regardless of improvements in integrity management practices and pipeline safety, public resistance to pipelines and anxieties of various kinds only seem to intensify. Knowledge and experience operate within various social frameworks, including the nature and degree of people's trust in scientific experts and authority. This talk gives an insight of theories of public opinion formation and change. Understanding the public’s concerns at a fundamental level is useful for all stakeholders and the pipeline industry at large in developing and communicating the pipeline narrative.


In the late 1990’s the natural gas and hazardous liquid pipeline operators recognized the need for some level of industry standardization for the use of ILI systems.  This need was driven by the increasing popularity of ILI as a proactive integrity assurance tool.  United States (US) Government regulators also soon became increasingly interested in pipeline integrity assurance for public safety as a result of major pipeline accidents in Bellingham, Washington (1998) and Carlsbad, New Mexico (2000).  In Europe, the Cullen Report of the 1988 Piper Alpha Accident had already started a movement towards regulation of proactive pipeline operations and maintenance. 

This paper has been created to understand and document the development of API 1163, “In-line Inspection Systems Qualification Standard, and the intent of its development. Particular attention is paid to SYSTEM OPERATIONAL VERIFICATION and SYSTEM RESULTS VALIDATION.  These definitions were initially confusing to some users and have been further clarified in the most recent editions of the standard, including a swapping of the terms “validation” and “verification”.

API 1163, as is most industry standards, was not intended to be established as a mandatory industry code but was written with the understanding that regulatory bodies could make it a requirement in the future.  In fact, it was incorporated into the US Department of Transportation CFR49 Part 195 in early 2017.


This paper identifies issues with hydrotest data acquisition practices and provides process recommendations to improve the quality and accuracy of hydrostatic test results.  Procedures for acceptance of hydrotest results are common in the industry, including calculations for resolving changes in pressure in response to changes in temperature of the test medium. These procedures assume temperature data accurately reflects the test water within the pipe.  Instrumentation selection, location, and installation methods can significantly affect the accuracy of data.  Since the temperature probe is typically not in direct contact with the water inside the pipe, measurements can be influenced by the environment around the temperature recorder.  Accuracy and resolution in typical analog temperature recorders is not sufficient for calculations to resolve small pressure changes resulting from temperature changes in the test medium.  Electronic recorders have better accuracy and resolution but without proper installation the data still may not be representative of the test medium.

This paper provides recommendations for proper selection and installation of instrumentation used in hydrostatic testing to improve data accuracy.  A process is presented to validate the data collected during testing by evaluating all test information for inconsistencies prior to conducting calculations for pressure to temperature relationships.  


PHMSA’s modifications of the federal rules governing pipelines transporting natural gas call for MAOP verification of pipelines whose current MAOP was established through §192.619(c). Engineering Critical Assessment (ECA) approaches to MAOP verification have been added as part of the modifications to federal rules. ECA combines flaw detection data, material properties, historical pressure cycling data, fracture mechanics, and fatigue modeling. Material properties for the ECA include toughness. Toughness characterizes the resistance of a pipeline joint, or feature, to crack extension. Its characteristics can be a function of location in a joint, e.g., long-seam, heat affected zone, and primary or base metal. It is also strongly dependent on temperature. An additional variable can be material inhomogeneity, that is, toughness can be a function of orientation. Finally, the uncertainty in laboratory standard tests for toughness can have a high degree of correlation to test temperature and sample location. In this presentation the authors discuss their review of the state-of-the-art of nondestructive testing technologies proposed to estimate toughness of pipeline materials. These NDE technologies include external and internal (ILI) application. A proposed validation methodology and a methodology to determine the required precision and accuracy of toughness NDE technologies will be discussed. 


The multiple levels of variability in inputs such as pipe properties, defect dimensions, and corrosion rate lead to a large amount of uncertainty in remediation decisions. The problem is, in the face of this uncertainty, how to determine when “enough is enough” in your dig program - what to leave in the ground until the next reassessment. 

While there are equations (B31G, RSTRENG, etc.) that give an estimate of the remaining strength of a pipe given a certain wall loss, these equations are often conservative estimates for a known anomaly size. Furthermore, the uncertainty in the dimensions and corrosion rate add more unknowns. Even if the reported depth and length are assumed to be without error, there are many other degrees of uncertainty from other inputs to the remaining strength equation that it can’t be solved directly with a reasonable degree of confidence. 

Operators often either ignore this uncertainty or resort to making “worst case” assumptions. Ignoring uncertainty is hardly defensible and “worst case”, while conservative, can indirectly increase your risk by misdirecting the use of finite resources.

In this paper we propose a Monte Carlo solution for dealing with the uncertainty surrounding key integrity management decisions. While these methods might seem exotic and complex, they are not as difficult as they may seem and facilitate integrity decisions that are both defensible and conservative.


In-service ruptures due to crack-like anomalies tend to be “black swans” in that they are rare events that require multiple factors to coincide.  Factors that govern rupture include the dimensions of the flaw and the material toughness at the flaw tip.  Moreover, fabrication imperfections such as hook cracks and lack-of-fusion flaws tend to behave as notch-like features rather than sharp planar cracks that are aligned with the hoop stress.  Over time, however, pressure cycling leads to the initiation of sharp fatigue cracks at notch-like features, resulting in a significant reduction in burst pressure.  Therefore, the number and severity of pressure cycles that a flaw experiences over its lifetime is another important factor in rupture probability.

A common practice for evaluating the crack threat is to perform burst testing on pipe joints in which significant anomalies have been identified by inspection.  However, this practice is ineffective relative to its cost.  Black swans correspond to the extreme tail of the bell curve, while cut-outs harvested during digs tend to sample the middle of the bell curve.  In other words, the likelihood of sampling a black swan with a cut-out is extremely low. Sizing uncertainty in both ILI and in-ditch NDE makes it difficult to identify the pipe joints with the largest flaws.  Moreover, flaw size is merely one of many factors that govern the risk of in-service rupture.  Burst test data on a small number of cut-outs provides very little information on the integrity of the remaining joints in the pipeline of interest.

This paper offers alternatives to the traditional approach to cut-outs and burst testing.  Cut-outs can provide valuable data, but only if controlled experiments that isolate key variables are performed.  Moreover, a risked-based analysis that considers multiple variables provides a more effective means to prioritize digs and identify cut-outs.


Historically, hydrotesting is the most accepted methodology to detect critical flaws in a pipeline system. A release during this test removes a critical flaw from the system. Pipelines are taken out of operation and potentially damaged from large pressure cycles, so additional infrastructure is required for water handling and disposal.  As such, effective and efficient in-line inspection systems are preferred by many operators. 

In 2014 Enbridge published a request for proposal to develop and provide a solution on a specific type of long seam cracks in a 26 inches pipeline. During the early phases of the project NDT Global analyzed in detail Enbridge's requirements, including the specific challenges, spool type, seam characteristics etc. and provided different proposals to Enbridge. In 2016 both parties signed a development contract to develop and build a 26" Next Generation Crack Detection Robot. 

On an abstract level, the different activities during the development can be summarized as follows:

  1. Gap Analysis
  2. Literature research & technology screening
  3. Development of mechanic and electronic concepts
  4. Prototyping and component testing
  5. Development of entire robot, test and validation
  6. API 1163 conform validation 
  7. Initial application in real pipeline with validation digs, 

The robot was utilized successfully and collected 45TB  data. Processing, analysis and reporting was performed within pre-agreed time frames. Initial field findings show high correlation of ILI and real flaws and proof the stated accuracy of the new service.

The robot can be utilized in the best configuration depending on threat, pipeline conditions, inspection speed and/or medium characteristics.

Special modes for base material and seam or even special seam type modes are possible to setup. This flexibility allows the operator to collect the best data for each situation. Feeding the information into the integrity system allows Enbridge to maintain safety of the asset.


For energy pipelines that experience cyclic pressures, a fatigue analysis is an important component of a integrity management program. Fatigue is the progressive cumulative damage and is known to initiate and propagate cracks to failure in pipelines due to internal cyclic pressures that are less than the Maximum Operating Pressure (MOP). While it is challenging to establish the appropriate conditions and predict the time for cracks to initiate from discontinuities in pipelines, their growth rate and behavior can be reasonably predicted by applying a fatigue crack growth model supported by experimental fatigue crack growth rate testing.

The choice of a fatigue crack growth model can significantly affect the final fatigue life prediction of pipelines with cracks and the determination of the next integrity assessment interval. Fatigue Crack Growth (FCG) rate models are meaningfully expressed as a function of the Stress Intensity Factor range ∆K and the rate of crack growth as a function of the number of fatigue cycles da/dN.  FCG rate models are generally derived by fitting experimental fatigue crack growth rate data which are strongly influenced by the test environment, temperature, frequency, and applied stress range. Unfortunately, there is limited experimental FCG rate data for air conditions and even less data are available for corrosive environments applicable to API 5L line pipe steels.  This paper will present a complied list of published fatigue crack growth rate testing data applicable to pipelines and will describe the benefits and limitations when performing FCG rate testing of pipelines in accordance with ASTM E 647. Care should be exercised when using published fatigue crack growth rate data obtained from steels and environmental conditions similar to those of the pipeline being evaluated because a mismatch can result in unreliable fatigue life predictions or overly conservative results.


Crack management has become a major focus for many gas and liquid transmission pipeline operators. Failures associated with crack-like features have been a concern for both operators and regulators. As a result, pipeline operators are excavating large numbers of features for not only in-line inspection validation purposes, but also to make repairs. Additionally, in-line inspection technologies have advanced significantly in recent years and are identifying an increasing number of features with greater levels of accuracy.

Because of high levels of conservatism associated today’s assessment methods, pipeline operators are spending a significant amount of capital excavating crack-like features. There is a need for improved assessment methods that integrates testing simulated / synthetic crack-like features. This paper will provide details on a study funded by ROSEN to systematically generate crack-like features in pipeline materials with the application of cyclic internal pressure loading. Synthetic crack-like features were generated using electronic discharge machining (EDM) to form notches in pipe materials. Initial stages of the program involved sectioning features to quantify crack growth levels. Once a systematic process had been validated, testing involved cyclic pressure fatigue to failure and burst testing. Programs such as the one presented in this paper are useful for both generating features in pipeline materials and also quantifying behavior considering cyclic pressure and burst loading. 


One of the major threats to a pipeline's integrity are combined features, cracks associated with dents being one of them. The mechanical design of the Ultrasonic Crack (UC) ILI robot consists of a sensor plate with a fixed incidence angle depending on the coupling medium. This plate is attached to the skids which are in contact with the internal wall of the pipeline. When the ILI robot crosses a dent, the incidence angle is not optimal, therefore, the detection of any interacting feature with a dent is compromised. 

In this paper, the authors will describe the systematic approach, consisting of four phases, in order to understand the detection capabilities of an UC ILI robot while crossing a dent.  

  1. Mechanical Design 
  2. Simulation Campaign 
  3. Pump Test Artificial Features
  4. Pump Test Real Features 

In addition, the authors defined variables such as dent geometry, location of the feature within the dent area, and depth of the feature to categorize the likelihood of an analyst identifying an interacting feature due to the liftoff caused by the dents and the amplitude drop caused by the wrong incidence angle. 

The summary of the research will include the boundary conditions of the detection and identification capabilities of cracks within dents for an UC ILI robot. 


Managing the assorted threats in pipelines is an evolving process. The inspection technologies, material testing methods and models used to assess feature severity improve over time. With respect to the assessment of crack-like features, the most commonly used approach is some form of analytical calculation using the inputs of feature, size, loading conditions and material properties. The inputs selected will almost certainly include some level of conservatism. In addition, there are many things embedded in our approach that tend to differ from what will be the case in “real-life”.  The ability exists to supplement feature assessment with larger-scale testing of actual features removed from service. This enables a demonstration of what the actual burst pressure would be in a real-life scenario while helping to understand the failure mechanism and type. The results from larger-scale tests can be used to bolster understanding and assist in making sound integrity decisions. They can be used in addition to the predicted burst pressures to deliver more representative results of the full material section under the same loading conditions as features would experience. Large-scale testing to burst also enables a critical review of the reported feature dimensions from ILI or other NDT to reality. In many ways, this large-scale testing is an evolution of the laboratory testing. The aim of this paper is to present a summary of the large-scale testing performed by Consumers Energy on pipes containing SCC that have been removed from service. The learnings from these tests are proving invaluable in terms of understanding the threat and how to manage the results from feature assessment calculations. This proactive approach of testing pipes removed from service helps Consumers to manage their pipeline and offers tremendous learning to the industry.  


The aging pipeline infrastructure in the United States has created the need for advanced engineering assessment methods. An essential critical part of this process involves the use of inspection technologies, primarily in-line inspection (ILI) and in-the-ditch technologies. To validate ILI technologies, it is essential to have advanced Non-destructive Examination (NDE) inspection technologies to verify ILI measurement accuracy.  Current NDE methods are available for inspecting deformations and corrosion, however, inspection of cracks and seam anomalies remains a significant challenge. Computed Tomography (CT) is an advanced inspection technology that delivers greater accuracy than conventional NDE inspection techniques for these challenging features.

This paper will present details on a comprehensive study designed to validate the CT technology by evaluating the growth of various crack-like features subjected to cyclic pressure. Simulated crack-like features were generated using electronic discharge machining (EDM) to form notches in pipe materials. The simulated defects form cracks after a certain number of pressure cycles and eventually cause pipe failure. After a designated number of pressure cycles were applied, cracks were inspected using the CT technology and compared to results based on measurements from actual sectioned features.

The elements of this program will contribute significantly to advancing inspection capabilities for the pipeline industry. The results of this study can provide a basis for establishing industry calibration and reference standards against which ILI technologies can be evaluated.  A critical limitation in the pipeline industry is the absence of reference standards based on real-world samples and the necessity to destructively test samples used in technology evaluations. The successful implementation of CT addresses current inspection limitations and is poised to make significant advances in validating ILI measurements. 


Pipeline operating pressure data is routinely analyzed in the time domain using “rainflow” cycle-counting to characterize the stochastic pressure signal in terms of the frequency, magnitude, and sequence of pressure cycles of identified magnitudes.  The results are typically used in one of three ways: (a) comparison against template histograms denoted by severity class (per TTO-5); (b) estimation of an equivalent number of uniform-magnitude cycles according to a cumulative damage rule (e.g. the Palmgren-Miner Rule) which is then applied to a fatigue life prediction model; or (c) discrete calculation of incremental crack growth produced by each stress cycle in sequence until the defect is predicted to fail.

Technique (a) is useful for prioritization but is non-predictive; technique (b) is convenient and lends itself to probabilistic models but cumulative damage rules have been shown to be subject to large and unpredictable error; technique (c) is computationally too burdensome for convenient probabilistic applications.

This paper describes the analysis of operating pressure data in the frequency domain in the form of a power spectral density (PSD) analysis.  The PSD analysis of pressure data produces clear distinctions between differing pressure histories that are not readily apparent in simple histograms.  The PSD profile can be analyzed to develop corrections to the Palmgren-Miner equivalent number of pressure cycles.  As a result, a simple uniform-cycle estimate of fatigue life from an integration of the Paris Law can be performed to produce a deterministic or a probabilistic forecast with improved accuracy and computational efficiency.


Inspection technology and assessment methods are currently insufficient to manage the increasingly common threat of circumferential stress corrosion cracking (CSCC) in pipelines. This paper will examine a range of novel techniques, that build into a risk based approach, which can be used to predict and identify locations susceptible to CSCC. This data driven approach uses a range of inline inspection and other data sets to build insights into the condition of the pipeline and the potential causes of CSCC. A selection of real world examples will be presented to demonstrate some of the possibilities and explain residual uncertainties. The data will then be combined into a risk based approach to demonstrate how the methodology supports integrity management decisions.  


Many factors affect how and when line pipe will experience a pipeline integrity threat, including material, vintage, environment and loading conditions.  An integrity threat of particular interest is stress corrosion cracking (SCC). SCC is a type of environmental assisted cracking (EAC) that can occur in line pipe under a very specific set of conditions.  First, the material must be conducive to corrosion, which in general is true for common grades of line pipe.  Second, the material must be subjected to a corrosive environment. And finally, the material must be subjected to tensile stresses.

The most common appearance of SCC results in colonies of cracks parallel to the axis of the pipe.  This is expected, as the principal stress due to internal pressure results in a hoop stress.  The in-line inspection (ILI) industry has responded to this threat with crack detection technologies that are specifically designed to detect and size axially oriented SCC and other crack features. 

However, additional pipe loads associated with external forces, such as line movement, can introduce additional stresses, resulting in non-axially oriented SCC.  Under these circumstances, the crack detection technologies specially designed for axially oriented cracking will not be able to accurately detect and size these features.

In response to this gap, research has focused on leveraging existing technologies that can infer the presence of non-axial SCC. Existing technologies can detect the presence of corrosion, determine the additional strain caused by pipe curvature and, given the right conditions, Magnetic Flux Leakage (MFL) technologies have demonstrated the capability to indicate regions with crack-like flaws. 

This presentation will review several case studies where T.D. Williamson has leveraged the data from the Multiple Data Set (MDS) technology to find regions where non-axial SCC might occur.


Circumferential stress corrosion cracking (C-SCC) can occur in an operational gas or liquid pipeline when the tensile axial or longitudinal stress exceeds hoop stress. Possible sources of axial stress include but are not limited to bending stresses, soil movement, construction practices, and thermal expansion. When this situation is coincident with a corrosive environment, say where the pipeline coating is damaged or cathodic protection is compromised, C-SCC can occur. Since C-SCC is not readily detectable using many ILI technologies, current practices for dealing with C-SCC take a direct assessment approach, using protocol driven priorities for excavation where conditions are conducive to C-SCC.

In this paper we expand on experiences stemming from investigations in direct detection and sizing of circumferential stress corrosion cracking (C-SCC) using axially oriented magnetic flux leakage (MFL) with high sampling rates. Using in-house developed methods in tandem with criteria summarised in the literature, data from inline inspections including axial MFL (AMFL), transverse or circumferential MFL (CMFL), caliper geometry, and inertial measurement unit (IMU) were searched and screened for possible C-SCC. Integrity excavation findings resulting from this reporting were then compared with reporting to gauge accuracy in both identification and characterisation of C-SCC. These integrity excavations were in a natural gas pipeline located in mountainous topography, lending to a higher probability of C-SCC. Results are summarised detailing reporting identification and sizing accuracy, with a synopsis of next-steps for this ongoing research project.


Thermal insulation over coating is effective...most of the time, but on occasion when the coating is compromised, a threat to the pipelines integrity can be realized due to corrosion under insulation (CUI). After a serious incident on the coast of Santa Barbara, CA in 2015, pipeline operators are required by law to act and address the threat of CUI with more frequent assessment intervals, confirmations, and a more stringent repair criterion (PHMSA’s advisory bulletin ADB-2016-04). To assist in determining the existence and the extent of this threat, in-line inspection (ILI) is recommended, but what is the best ILI technology, tool type, and resolution for detecting CUI? Two case studies for guidance on selecting the proper ILI tool and data integration will be presented, the findings not only are applicable to CUI, but also to other forms of localized corrosion, such as microbiologically influenced corrosion (MIC). The results will surprise you! 


There is a large network of pipelines that transfer vast volumes of hydrocarbons in a safe and cost-effective manner.  To prevent failure(s) or leak(s), pipeline operators utilize maintenance programs to regularly inspect these pipelines for integrity threats such as corrosion.  In-line inspection (ILI) technologies have been developed over the last 30 years and have become a pivotal tool in operator’s integrity management programs (IMPs). An inspection technology will be accompanied with detection and sizing performance specifications, which allows pipeline operators to understand the inspection capabilities and limitations of that technology.  There are several ways to develop a performance specification for an inspection technology, one method is through artificial defects and simulated speed tests (Arti Bhatia, 2004) (K Reber, 2006), however, the challenge for pipeline operators is the fact that a new inspection technology performance specification is not validated in the pipeline operating environment. The risks for an operator using an unvalidated technology include tool performance below the expected specification affecting the safety of the asset, mechanical damage to the tool or pipeline system, or receiving ILI results that are not currently accounted for in the operator’s IMP.  This paper highlights the methodology that TC Energy (TCE) has developed for validating a new ILI technology and the collaborative effort completed between TCE and Onstream to implement the TriStream magnetic flux leakage (MFL) technology. The process started with the vendor disclosing the new MFL platform performance specification with supporting documentation to demonstrate the readiness of the product. The operator’s validation process resulted in requiring additional considerations and supporting details, which the vendor supplied. Upon successful completion & review of the documentation, the product was subjected to performance review through in-situ field trial.  


An important part of any pipeline integrity management program is performing pipeline excavations. These excavations are more effective when all the data available for each is captured, for example, ambient and soil temperature, humidity, soil pH, pipe to soil potentials, coating condition etc. as well as the defect dimensions. Dig results are also valuable to in-line inspection (ILI) companies to improve sizing model accuracy, to educate analysts in defect characterization, and the development of new technical capabilities.

This paper presents case studies demonstrating how valid dig program feedback can help improve ILI results using reported features that are both within specification and features determined to be outside specification.  The paper also presents how ILI vendors are using dig feedback data and discusses what data is important for ILI reporting.  The paper concludes with how ILI performance is improved when combined with results from a good dig feedback program.

Key words: In-line Inspection, ILI, Dig Feedback, Dig Verification, Improved Defect Sizing, Defect Sizing Specification, Excavations


In-Line Inspection (ILI) technology is the considered one of the safest and most efficient inspection methods to inspect hydrocarbon pipelines. The retrieved data shall be validated and verified upon successful completion of the inspection. This paper is intended to introduce a new approach to validate the ILI run based on a statistical analysis comparing the new ILI run with a previous ILI run of the same pipeline by leveraging RMS model to quantify the similarity between the datasets. API-1163 and CEPA offer consistency criteria as a validation methodology for a new ILI run. Also, this paper will demonstrate a new scoring criteria for accepting Magnetic Flux Leakage (MFL) runs with partial data loss as number of MFL runs experience unexpected data loss, which might affect the minimum reporting threshold of the tool. The approach will help pipeline operators to identify the criticality of the missed data via a detailed comparison with the previous MFL run for the same pipeline and detailed analysis of the behavior of the tool during the run. The scoring criteria is aligned with the Pipeline Operators Forum (POF) requirements for data loss. Multiple case studies extracted from actual data will be presented throughout the paper. 


This paper presents a case study consisting of the execution and validation of remaining life predictions following a UTCD (ILI) of a liquid pipeline in a high-consequence area. The initial ILI identified several crack-like features in an area that would be difficult to perform excavations. Initial remaining life predictions demonstrated that these features had sufficient remaining life such that remediation was not required before the date of the next planned inspection. However, comparison of the ILI results to in-ditch NDE results indicated that the ILI did not meet its feature sizing specification.

The remaining life predictions were validated by executing the following tasks:

These results were used to update the initial remaining life predictions and further demonstrate that remediation of the difficult to excavate features was not necessary.


The evolution of steel manufacturing processes and the introduction of new alloying elements has resulted in improved strength and quality of pipeline steels. These developments resulted in changes to the magnetic properties of the steel and subsequently have impacted on the response of electromagnetic sensors used in ILI inspections. 

Many research programs have been conducted to investigate the effects that these changes have had since the 1980s, particularly on magnetic and electromagnetic methods such as the MFL technique as a widely used ILI inspection method. 

AXISS utilizes an electromagnetic measurement technique of which recent developments have progressed to enhance the AXISS system predictive accuracy by following a path of understanding, and quantifying, the effects of over half century of developments in the pipe steel manufacturing process. AXISS sensors were used to assess and characterise the magnetic response across an extended range of pipe material chemistries and mechanical properties. This paper presents the lessons learnt from a study results as from a large set of pipe samples recovered from field inspections, and benefits having an extended reference database of information that contains pipe chemical and mechanical characteristics. 


A full encirclement engineered steel laminate sleeve system has been developed and optimized for pipeline integrity management applications.The Leewis Augmentation Analysis (LAA), an engineering critical assessment (ECA) which follows the traditional Barlow and ASME B31G equations and methods serves as a basis to determine the pressure containing capability of repaired pipe has been created.This ECA incorporates the technical developments and progress in assessment methods over the last several decades and provides pipeline operates with improved functionality. The LAA is presented and discussed. Successful full scale ASME PCC-2 style burst and compliance tests are presented and reviewed. Highly instrumented full scale testing has also been complete to obtain the effective modulus of elasticity of the system as well as a measurement of the rate at which internal pressure is shared by the repair system. The initial stress strain response has shown that at 50 micro-strain within the unrepaired base pipe the system has accepted load. Long term creep and cyclical testing of the steel laminate is presented.  10 million cycles at 50% of ultimate lap shear stress of the adhesive has been obtained, the testing is reviewed. These engineering parameters serve as the basis for mitigation of strain based concerns.  


People, process, technology. Having strong processes and technology are great for any company; however how do you ensure people are competent when it comes to performing an in-service welding and hot tapping project? What are the common gaps when it comes to performing a repair live welding project from engineering, management, operations to field staff?

API 2201 Safe Hot Tapping Practices in the Petroleum and Petrochemical Industries describes the concept of competent vs qualified persons. A qualified person implies that a person has knowledge or education beyond or different from a competent person. The qualified person would possess a recognizable degree, certificate, or professional standing. A competent person has the capability of identifying existing and predictable hazards in the surroundings or working conditions which are hazardous or dangerous to personnel.

Construction codes, like ASME B31.4 or ASME B31.8, or the code of federal regulations may have maintenance and repair in-service welding requirements for installing hot tap fittings, repair sleeves, supports, or other components to a live hazardous pipeline system, although these codes only state that people performing the work are competent, and not how to ensure people are competent.

This presentation will recommend a new competency framework for an in-service welding organization, discuss the application of other oil and gas maintenance and repair codes and standards to in-service welding on pipeline systems, and address critical in-service welding quality and integrity management system requirements. 


Inline inspection data from several runs spanning many years is available for individual pipeline segments, but compilation of this data into a comprehensive picture of pipeline integrity necessarily relies on computational tools. A critical advantage of modern data storage, analysis and visualization techniques is the relative ease of performing statistical assessments of integrity operations. Data from a single user of OneBridge Solution’s software may comprise over 1,000 ILI runs, hundreds of pipe segments, several million aligned anomalies, and thousands of repair records. Automated alignment of ILI data allows a single physical anomaly to be reliably tracked through many years of repeated measurements of growth and correlated repair records which also factor in PODS asset data. 

We present a study of cases where ILI anomaly measurements warranted a dig operation in which repair actions were either performed or found to be unnecessary. The fraction of dig operations leading to a productive repair varies with the condition triggering the dig and discretionary choices about dig condition parameters. We explore the relationship between these parameters, ILI measurements, dig-to-repair ratios and the impact to operational spend.


Pipeline operators and supporting service organizations use various means for establishing the mechanical integrity of pipeline systems. Typically, in-line inspection technologies provide the first source of data from which critical integrity management decisions are made. These resulting inspection data are processed and utilized in a variety of ways, including finite element analysis, fracture mechanics, and risk-based software packages that all involve some form of numerical modeling. While there is no doubt that numerical modeling plays a critical and essential role in managing pipeline integrity, the increased use of full-scale testing could greatly enhance industry’s ability to evaluate the threats associated with anomalies of various forms.

The fundamental goal of any full-scale test is to simulate real-world pipeline conditions and establish a true limit state condition by applying simulated loads that could lead to failure. These include various means of loading including bending, axial tension / compression, cyclic pressure testing, and burst testing. Further, full-scale testing is an ideal means for validating repair technologies, including composite repair systems and steel sleeves. This paper provides several case studies on the full-scale assessment of dents, crack-like features in seam welds, and simulated bending associated with geohazard loads, as well as the assessment of repair technologies including composites and steel sleeves. Also included is a discussion on data acquisition systems and monitoring devices, as well as a section on safety and calculation of burst energies.

The goal of this paper is to provide readers with a better understanding on the benefits associated with full-scale testing, while at the same time presenting various options available for testing that include the fabrication of specialized fixtures and equipment to achieve desired loading conditions.


Composite repairs are used commonly to repair wall thinning in pipelines but there is growing interest in using them to extend the service life of pipelines containing crack-like defects; this paper reports an example of such a repair.  A large colony of circumferentially oriented crack-like defects was found on the extrados of a field bend in an above ground section of an 18” diameter pipeline during scheduled maintenance works.  The investigation concluded that the defects were likely to have been caused during construction of the pipeline.  A fracture mechanics assessment was undertaken to estimate a hypothetical worst-case defect size to have just survived the hydrotest on commissioning of the pipeline.  A remaining life assessment was then undertaken based on known pipeline pressure loading and historical temperature profiles.  It was concluded that crack growth could have occurred and that remedial work was therefore required.  A carbon fibre composite repair system that had been tested previously to verify its ability to extend the fatigue lives of crack-like defects was selected.  Structural monitoring was installed to the pipeline to measure strains on the bend both beneath and remote from the defects and the repair.  These data were used to generate a base line against which the benefit of the repair could be quantified.  A composite repair was then installed over the defects and the strains in the line and the repair were monitored over a 12 month period.  The reduction in strain due to application of the composite repair aligned with predictions and enabled the remaining life of the repaired section of the pipeline to be extended such that it no longer presented a threat to the design life of the pipeline.   


As Oil & Gas technology continues to advance, manufacturer’s must make sure that composite repairs do not become a hinderance to commonly available technology. One such concern is the detectability of composite repairs by In-Line inspection tools in carbon steel pipelines. As composite materials are not inherently detectable, there is a concern that an owner/operator may accidently dig up a damaged area only to discover it has been previously repaired. This paper discusses the detectability of magnetic In-Line Inspection markers installed within a composite repair system using a Magnetic Flux Leakage (MFL) inspection tool. The testing consisted of 15 different configurations using five composite repair materials along a 17 foot section of pipe. In most configurations, the composite repairs were detected by the MFL inspection tool showing reliable detectability for in-field use. Another common issue is in regards to holiday inspection devices used over composite repairs. It seemed that in many instances, properly installed composite repairs in the field would trigger an incorrect response. For this test 11 different configurations were installed on a stretch of pipe, several including ILI markers. Once cured, the samples were inspected using a SPY Model 780 Portable DC holiday detector at two voltages, 1.2 Kilovolts (KV) and 2.7 KV. Detection at the lower voltage provided mixed results with some repairs showing good insulation while others always seemed to trigger the tool. 


3D defect profile as input for more-accurate defect assessment and supplemental the safe pressure prediction

Reliable prediction of safe operating pressure is increasingly required in pipelines. Complex-shaped volumetric metal loss is a challenge in this regard. Methodological ILI restrictions and principal reporting simplification reduce the precision of ILI results and output tables.

Today, phenomenological MFL data parameterization is rearranging complex metal loss structures and often results in reduced accuracy. Nevertheless, the established maximum depth boxes do not necessarily prevent cases where MFL data interpretation is insufficiently conservative.

The presented groundbreaking MFL data evaluation technique directly calculates the accurate 3D metal loss geometry as an alternative to the currently practiced boxing. The resolution of the innovative result opens a new dimension for MFL. Auxiliary components of the typical MFL indirect interpretation method lose importance with this new system. Experience-based practices using human expertise, artificial intelligence or elaborate sizing models play a role still, but the conclusive accuracy of the calculated metal loss profile becomes significantly independent from the variability of these experience-based practices. 

This technology enables new services for difficult-to-access areas, areas with complex corrosion, or instances in which a laser map cannot be obtained, resulting in a more accurate, reliable and detailed corrosion growth assessment. This paper will detail concrete case examples from blind tests with high-resolution laser maps and will compare the new approach with results from the traditional MFL evaluation. Performance and reliability analyses help rationalize this innovative technique and demonstrate the contribution to more reliable management of metal loss threats.

As part of describing this new evaluation technique, theoretical and practical advantages will be discussed of combining axial and transverse MFL measurements for detection and sizing of metal loss anomalies.


Please check back


Pipeline operators use different approaches, often in combination to ensure the safe operation of an asset. Historically, hydrotesting is an accepted methodology for assessing critical flaws in a pipeline system, basically by stressing the pipeline above the standard operating limits. By design, a release during this test removes a critical flaw from the system. There are significant draw backs to this type of assessment. Such drawbacks include high costs of implementation, feature growth, previously blunt defects sharpening and system downtime. 

Driven by the operator to investigate alternative approaches, a consortium of 4 parties formed to develop and validate an alternative and enhanced solution.

The research and execution of the project was structured in four (4) different phases.

This paper is a summary of the research and work for this ILI In Lieu of Hydrotest initiative. 


Inline inspections (ILIs) provide valuable data that are used in integrity management programs to identify potentially problematic sections of a pipeline. The technology has inherent uncertainties that must be accounted for when integrity assessments are performed. Many pipeline operators use conservative deterministic methods of calculating and using corrosion growth rates to determine failure thresholds. ILI response programs and reassessment intervals are then constructed from these analyses, which are based on anomaly RPRs, burst pressures, remaining lives, or similar failure criterion. Including a risk-based probabilistic aspect to ILI response programs can enhance a failure prediction analysis. Remediating only the anomalies that pose a risk to safe operation for a specified timeframe may not be ideal if an extended reassessment interval is desired. By ranking and prioritizing the ILI reported anomalies in terms of which pose the highest risk to safe operation, a dig program can be created for each risk level deemed acceptable. This process depends strongly on the accuracy of the ILI tool. By using a Probability of Exceedance (POE) method in concert with the tool inaccuracies (in addition to variability in growth rates, probability of detection (POD), etc.), the likelihood of an anomaly to reach a failure threshold can be calculated. These probabilities can then be statistically summed up over a joint, portion of line, or entire pipeline. Typical ILI inspection data of a hypothetical pipeline were used to show how the addition of the POE method to a deterministic response program can be beneficial in managing risk on a pipeline; the reassessment interval length can drastically increase when remediation is focused on the high-risk anomalies. 


In Line Inspection (ILI) is a commonly used technique to assess the integrity of a pipeline.  When data from more than one ILI are available, the data can be compared to estimate the corrosion growth rate (CGR), which can then be used to estimate the remaining life of the pipeline – the number of years remaining until the metal loss exceeds a specified limit.  Deterministic methods (those that consider inputs as discrete values) are commonly used for remaining life assessment.  Rather than use average values, the inputs are adjusted to introduce conservatism.  Adjustments can include adding ILI tool tolerance to the assumed feature depth or applying the maximum estimated CGR from the ILI run comparison.

This paper will describe an approach to conducting remaining life analysis that treats the key input parameters, corrosion growth rate and feature depth, as distributions rather than discrete values.  The Closed Form Probabilistic Method for Remaining Life Assessment (CFPM RLA) approach takes CGR distribution descriptive statistics, feature depth sizing accuracy descriptive statistics for the ILI tool, pipeline physical properties and operating pressures to calculate the Probability of Exceedance (PoE) over time for two limit states: 80% wall loss (“leak”) and Modified B31G (“overload”). The method allows the Operator to explore how PoE increases with time, which assists the Operator in making risk based decisions on timing of re-inspection, repairs, or other mitigation efforts.  Case studies using the CFPM RLA approach will be shared.


As the pipeline industry develops a more extensive history of multiple inline inspections on the same pipelines, run comparisons are becoming an increasing common tool as well.  Run comparisons (RunComs) are a method for attempting to determine the corrosion rate for a pipeline by comparing the feature lists from two successive ILI runs, programmatically matching as many anomalies as possible and dividing the difference in reported depth by the time between runs to arrive at an assumed corrosion rate.  The one factor that is well known but typically ignored is that neither of the ILI runs is a perfect measurement or a very precise one at that.  The depth tolerance specification for an MFL tool allows for 20% of the measurements to have an error that exceeds 10% of the wall thickness.  Further complicating this process is that when several measurements are used in a calculation such as a RunCom, these errors compound to create an even wider band of uncertainty.  This is a familiar concept in any physical science and is known as error propagation.  Given the compounding of these successive errors in a calculation, it is necessary that these uncertainties be accounted for when decisions about the integrity of a pipeline are being considered based on a RunCom.  To judge the validity of any measurement or calculation, the end-user must have a means of determining the uncertainty involved.  Without accounting for this, a random error can easily be confused as “growth” of an anomaly between runs leading to the unnecessary expenditure of resources.

This paper will present the fundamentals of measurement theory and conclude with a discussion of the types of measurement error, how to properly account for error propagation and its application to run comparisons for more informed integrity decisions. 


There have been significant advances in magnetic flux leakage (MFL) in-line inspection (ILI) technologies in recent years. These have led to improvements in Probability of Detection (POD), Probability of Identification (POI) and Probability of Sizing (POS).

Whilst often the main focus of these advancements is the inspection vehicle itself, the end product of an inline inspection service is reliable and accurate data. This end product is influenced by various technological factors which include: recognition and detection algorithms; complex sizing models; robust and rigorous processes and highly trained and skilled data analysts.

This paper explores all the main factors that contribute to delivering the reliable and accurate inspection reports that pipeline operators demand today. This review will be supported by extensive comparison of ‘as reported’ data vs ‘in ditch’ findings. This is particularly valuable for operators of inaccessible pipelines, where proving ILI performance is at least challenging, and often not possible.


In corrosion management, the majority of pipeline operators utilize in-line magnetic flux leakage (MFL) inspection as the primary source of information for integrity decision-making. In-line inspection (ILI) service companies provide anomaly dimensions (length, width, depth) and general sizing accuracy tables that specify the expected uncertainties for each POF anomaly dimension class. These tables represent “population specifications” and, as such, provide tolerances that are met at a certain level of statistical confidence, typically stated at 80% or 90%. Due to various influencing factors, the true tolerance of an individual anomaly may well be lower – or occasionally higher – than the listed value. Integrity assessments, however, apply the population specification to all the individual anomalies called by ILI. This approach can lead to suboptimal decisions: indications selected for repair or replacement in reality may not require any action, while other anomalies not considered in the action plan in reality should be addressed.

As an improved method, this paper proposes an individual anomaly sizing certainty by using posterior distributions over length, width and depth and its propagation to the pressure assessment to enhance the corrosion assessment. Using the latest artificial intelligence methodology, namely Invertible Neural Networks (INN), the paper demonstrates the feasibility and validity of the individual sizing certainty estimation by comparison to verification results.


In-line inspection (ILI) integrity assessments provide pipeline operators with information that helps assess the current state of pipeline integrity. The information garnered from a single ILI is valuable towards ensuring safe operations of pipeline systems. However, ILI results can be further enhanced when subsequent inspections are available for the purposes of performing a run-to-run comparison. Harmonizing ILI data can provide powerful insights into identifying locations of active corrosion. To that end, the information can be used to establish corrosion growth rates along the length of the pipeline segment, and in establishing defensible integrity reassessment intervals. This paper will go through a case study in which ILI bias analysis, pit-to-pit matching, statistical data analysis, raw signal review, and corrosion growth rates will be evaluated.  


Pipeline integrity management focuses on the importance of sound engineering assessments of inline inspection (ILI) data to determine accurate corrosion growth rates (CGRs).  An accurate CGR is vital to ensure the safety of the pipeline and optimize the timing of remediation. Comparative analysis of ILI data for the purpose of ascertaining CGRs also provides operators with the data to establish the timing of the next inspection. 

Corrosion growth models comparing subsequent inspections can vary across the pipeline industry.  Models can be complex in nature, requiring specialized expertise, or simplistic, allowing for ease and accuracy of implementation. A corrosion model that is often used is the Half-Life model. In this model, assumptions are made whereby the corrosion has grown linearly from half the time the pipeline was put in service. Alternatively, a Full-Life model assumes that corrosion has been active since the first day the pipeline was placed into service. Another common practice is for pipeline operators to compare subsequent ILI feature spreadsheets to obtain a CGR for reported features. This model often assumes that corrosion features grew linearly without accelerated corrosion growth mechanisms. ILI comparative analysis with raw data review has increasingly been leveraged in recent years to more accurately compute CGRs. The raw data provides additional value and insight to spreadsheet comparisons. 

In this paper, studies conducted to compare these different CGR models are presented whereby the predicted depths for known anomalies are compared with the actual depths of the features, determined from in the ditch measurements.

Case Study 1: Anomalies’ Predicted Depth Utilizing Half-life and Full-life CGR versus Field-Found Depths

Case Study 2: Anomalies’ Predicted Depth Utilizing ILI Feature Spreadsheets CGR (without Comparative Analysis Raw Data Review) versus Field-Found Depths

Case Study 3: Anomalies’ Predicted Depth Utilizing Leveraging ILI Raw Data Review CGR versus Field-Found Depths


In-line inspection (ILI) data tells us about the condition of a pipeline. Integrity engineers often make ‘expert’ judgements and mentally classify the condition as ‘good’ or ‘bad’, based on the type, number and severity of features reported. If the condition seems good, then we expect relatively straightforward and inexpensive integrity management. If the condition seems bad, then we brace ourselves for more work and high costs. Are these expert judgements of ‘good’ and ‘bad’ valid, and can they be made more rigorous?

To answer this question, we began by exploring ILI results for over 1000 pipelines, using descriptive analytics techniques. For each pipeline, we defined two simple metrics to characterize their condition – the number of corrosion features per kilometre, and the probability of exceedance. By comparing these metrics on a global scale we proposed some definitions of ‘good’ and ‘bad’. That preliminary work was published at PPIM in 2019. To further evaluate these definitions and extend the usefulness we have more recently expanded the database to over 5000 pipelines and also considered additional condition metrics such as corroded area. In this paper we present results from this extended database and investigate changes when compared with the 1000 pipeline database 


The "Gas Mega Rule" – the US regulators revisions to 49 CFR Parts 191 and 192 was issued in September 2019 and has major repercussions for gas pipeline operations. A major challenge that operators will now face the requirement for material verification, reconfirmation of the maximum allowable operating pressure (MAOP) and better records/data management for gas transmission pipelines constructed before 1970.

In many cases operators may have such traceable, verifiable and complete (TVC) records but if such a document and records search, review, and verification cannot be satisfactorily completed, the operator cannot rely on this method for calculating MAOP and must instead rely on another methods. The most onerous would be a hydrotest, but a more efficient route is to fill the gap in the records. How it this best approached?

The process advocated is a stepwise approach to compliance. The first step is to assess what records are already available, second is to identify gaps, and third develop a plan to acquire the missing records using a combination of pipeline preparation, in-line inspection (ILI) for identifying material binning at a joint level and in-field verification using positive material identification. The full process will enable material TVC records to be produced which can then be used to confirm or re-confirm the MAOP/MOP. A digitized system will make it much easier to store, search and cross reference not only to meet the rule but for the assurance of safety for future operations. This also forms an important part of the deliverable of this unique solution.

This paper presents the requirements of the rule, the alternatives, and the approach that will provide operators with a compliant, safe and economically viable end to end solution. Examples will be presented highlighting the value and cost benefit. 


The Pipeline and Hazardous Materials Safety Administration (PHMSA) issued a Final Rule on October 1, 2019 that is expected to impact the pipeline industry’s approach to crack fatigue analyses.  The Final Rule defines pipe toughness values to be utilized when analyzing crack anomalies that are subjected to fatigue growth for instances in which actual toughness values are not available.  Pipeline operators often conduct these types of analyses to establish reassessment intervals for pipeline assets in their system.  Thus, the impacts of this Final Rule are felt by all pipeline operators who own assets in which cracking is considered a threat.  

The goal of this paper is to utilize results from a case study to quantify the effects that this Final Rule is expected to have.  This case study will be focused on pressure test scenarios in which the defined toughness values are utilized in establishing the just-surviving and critical flaw sizes that ultimately dictate maximum reassessment intervals. 

The paper will conclude by identifying 1) when is it appropriate to utilize alternative means of establishing pipe toughness values, 2) what avenues must operators take in order to utilize alternative pipe toughness values, 3) and how to establish whether traceable, verifiable, and complete records exist to justify use of alternative pipe toughness values.


We’ll address the findings from a 3.5 year program working with a pipeline service provider. The initial issue was a rising TRIR rate (total recordable injury rate) that crossed the 1.0 TRIR threshold that most pipeline companies require of contactors. The study shows how instead of spending money on “better” training, their management agreed to create a high reliability organization (HRO) safety culture change to build a “DNA” of safety competency within its frontline leaders and workers. 

The skills gap in our next generation workforce is well documented. But what may be less widely known is the impact safety competency has on new workers, when our highly capable and seasoned “boomer” generation is no longer working alongside them. This lack of one-on-one support can, and does, result in more workplace incidents. The problem is compounded when leaders prioritize results over reinforcing safe work practices and processes. As a result, younger, less experienced workers feel they must get the job done without really understanding how to mitigate risk.

The study shows how this company has started to bridge their safety gaps.

This study describes how a pipeline services company reinvigorated its safety culture, improved customer safety and ensured integrity of work through various channels, including organizational change management, leadership and workforce training, supporting leaders for success and instituting safe work management practices. The study focuses on specific steps in five key areas: 

  1. Determining why incident rates increased.
  2. Analyzing the workplace changes that affected a previously successful system. 
  3. Engaging leaders and workforce to reestablish a culture of safety.
  4. Creating a HRO around safety, technical skills to demonstrate correct performance.
  5. Building competence for the future.

Although this study details one company’s ongoing journey to manage its global workforce, the efforts are transferable to other organizations. 


The purpose of this paper is to describe a new family of combination Magnetic Flux Leakage

(MFL)/Deformation inspection tools used to inspect small-diameter, previously unpiggable, pipelines.

Small diameter pipelines have been historically very difficult to inspect with In-line Inspection (ILI) tools. Small diameter pipelines are challenging because they leave ILI system designers little space to fit the required system components of an ILI tool. In addition, many small diameter pipelines were designed and built without any consideration for ILI tool passage. For example, many of them have tight radius bend fittings and no launchers or receivers installed on the pipeline. 

Recent advancements in microprocessor computational power, memory density, sensor technology, engineering design/modelling software, and rare earth magnetic technology have allowed an inspection system to be developed to inspect these small diameter pipelines.

This paper will describe the design parameters used to develop this new system. Several case studies will be presented showing real-world application of this new system. 


There are many pipelines which cannot be inspected along their entire length by in line inspection. The integrity management of these unpiggable lines relies on a combination of threat assessment modelling, based on available design and operational data, as well as local inspection at accessible locations. This paper considers the planning and interpretation of targeted, local inspections for management of the threat of internal corrosion, in unpiggable production and export systems. 

The amount of data relating to the corroded condition which is collected by an inspection depends on the detection capability of the tool used and the extent of the pipeline which is inspected. The number and size of anomalies which are found by the inspection will also depend on the actual, underlying condition of the pipeline. 

The work presented considers the extent of inspection required such that sufficient data would expect to be collected for post-survey analysis. The paper discusses the nature of conservatism in the analysis: an integrity or risk assessment based on modelled data will use a conservative approach for anomaly size or corrosion rate, to model a worst-credible condition. However, this approach has the potential to be non-conservative for deciding the scope and extent of a local inspection because it would predict a larger number of more severe anomalies, which would be more easily detectable. 

The study also considers what conclusions may be reasonably drawn in the event that no anomalies are reported by the local inspections (the meaning of nothing). If the inspection returns no results (detects nothing), is it because there is no corrosion, because the inspection extent was insufficient, or because the tool was not suitable? The paper investigates a Bayesian approach to understanding the confidence with which the detection of no anomalies by a local inspection can support integrity assessment.  


For piggable pipelines the standard pigging sequence normally consists out of: Cleaning, gauge pigging, caliper pigging, and finally, the inline inspection (ILI) tool. This has worked well and is often recommended and required by the pipeline owners. But can the same sequence be used for difficult-to-pig pipelines and should the pipeline owner require a gauge pig run?

The ability of gauge pigs to mimic the navigation capabilities of the ILI tools in difficult pipelines is very limited. Sourcing a gauge pig that can navigate lines with features like 1D bends, back-to-back bends, mitre bends or un-barred tee’s at 6 o’clock, can be a bigger challenge than sourcing an ILI tool that can navigate the pipeline.

In most instances, when dealing with ILI tools that are designed for difficult pipelines, the damage is only a sign of the limitations of gauge pigging. Using a smart, progressive-cleaning sequence executed by an experienced operator can negate the need to use gauge pigs. When combined with an ILI tool that can collect geometry and wall-loss data simultaneously while also minimizing the risk of damaging the pipeline or itself, then both the gauge pig and the caliper run can be eliminated from the pigging sequence. Low pressure / low seal ILI tools can be deployed in pipelines with minimal risk of becoming lodged into an obstruction and can be retrieved by reversing the flow.

During the presentation several examples of gauge plates with indications will be presented from pipelines that have been inspected without any severe complications. In addition, examples of ILI runs whereby the ILI tool stopped in the pipeline due to an obstruction, the flow was reversed and the remainder of the pipeline was inspected from the other side without the ILI tool nor the pipeline sustaining damage will also be presented. 


Colonial Pipeline has been implementing a comprehensive program to proactively maintain the integrity of their pipelines in their tank farms across the United States. As part of this effort, Colonial Pipeline systematically conducted robotic pipeline inspection in order to accurately assess the integrity of their pipes.

The primary objective of the robotic inline inspection was to acquire accurate data on the integrity conditions of Colonial Pipeline’s pipes in tank farms, which typically run from a tank to a manifold.

Colonial Pipeline operates numerous tank farms across a variety of states. Tank farm pipes typically carry products including diesel fuel and various grades of gasoline. These pipes in tank farms are not suitable for traditional inline inspection. When considering an optimal method, not only is it paramount to choose an inspection method that yields comprehensive and accurate data, but also one that minimizes the amount of time required to complete the inspection and generate corresponding reports.

In 2015, Colonial Pipeline utilized Pipetel’s Explorer robot for pipeline inspection for the first time to inspect a 24 inch pipe in a tank farm. Since then, more pipes ranging from 24” through 36” in diameter have been inspected.

This paper will primarily discuss the tank farm pipeline integrity program as well as the operations and preparation required for conducting a robotic pipeline inspection for a tank farm. We will also discuss the results obtained from robotic inline inspection including the accuracy of the data collected on internal and external corrosion as well as third party damages. Finally, we will analyze the impact of the data collected 


In many coastal regions, marine terminals are gateways through which both crude oil and refined products must pass to ultimately reach their consumers.  Large storage tanks temporarily hold these liquids after they are delivered to the terminals, before they are collected.  Complex systems of pipework connect the tanks often featuring buried and/or elevated sections, thermal lagging, multiple tight bends including miter bends and jetties to allow product transport to vessels berthed in deep water.

Assessing the integrity of terminals lines by manual spot-checking is common; tracking wall thickness at a fixed point over a period of years and monitoring for any degradation.  This is a labor-intensive and time-consuming process when there are so many pipes present and any protective paint or thermal insulation needs to be removed for each spot.  “Inaccessible” sections such as buried or elevated pipework are sometimes overlooked.  Readings are taken at the most convenient locations as opposed to the points where corrosion is most likely, so the data provides only a best-case assessment.

By comparison, inline inspection is rapid, easy and provides a picture of integrity across the whole pipe, ensuring the biggest risk areas are identified and quantified.  For facilities where a leak could see:

the environmental, economic and reputational costs of a terminal line failure surely merit a holistic inspection regime.

This paper compares the process and outcome of a common spot-checking approach to a detailed inline inspection survey at marine terminals.


The use of in-line intelligent pigs is usually the preferred method of inspection to determine the condition of pipelines since they offer near full coverage. A large proportion of installed pipelines are, however, such that pigging is not possible. Alternative means of inspection are therefore necessary for these lines.  Identifying suitable inspection techniques and strategies is becoming an increasing concern as pipelines age and, in many cases, are in operation beyond their originally anticipated design lifetimes. 

This paper gives an overview of methods for planning and evaluation of externally applied inspection of unpiggable pipelines where the aim is to validate integrity. It highlights how the degradation scenario should be considered and the planning stage. Three different categories are described.

  1. Situations where degradation with the potential to threated integrity is not expected and the aim of the inspection is to validate absence of degradation. In these situations a low coverage sample inspection, with sensitive techniques, which finds no evidence of degradation, even at a low level, is used to allow statistical estimates of the probabilities of degradation exceeding limiting conditions.
  2. Situations where some degradation is expected and it is sufficiently homogeneous such that a sample inspection can provide a basis for statistical estimates of the overall condition.
  3. Situations in which the degradation may be isolated and unpredictable in location, such that a high coverage screening inspection is the only feasible method to identify if any degradation present has the potential to threaten integrity. 

The paper summarises the different statistical methods used at the planning and evaluation stages for each of the above. Case studies, covering practical application, for range of different scenarios are provided. These highlight the links between the planning, implementation and evaluation phases.


Pipers® are an easy to deploy, miniaturized screening tool particularly suited for small diameter and difficult to inspect pipelines. Using a combination of sensors, Pipers® provide an assessment of the overall pipeline condition by identifying deposits, leaks, volumetric metal loss, and other anomalies that threaten pipeline performance and safety. Next to that, Pipers® can detect pipeline features such as joints, bends, and schedule changes.

This paper will provide a comparison of Pipers® data to traditional in-line inspection MFL data, discuss where they are complimentary and describe specific use cases for the Pipers® in combination with an MFL tool or as stand-alone screening. 

The comparison is based on screenings with Pipers® in a pipeline prior to running an MFL tool, and after running an MFL tool, and the results will be compared to that of the MFL tool specifically addressing the distance accuracy, defect and anomaly detection and overall pipeline condition assessment.  

This provides insights in the use of the Pipers® as a high-frequency screening tool, used in-between MFL runs and as a stand-alone tool for pipelines that cannot be inspected with traditional MFL tools, to optimize pipeline inspection programs, reduce pipeline downtime, and increase profits for pipeline operators. 


Two unpiggable pipelines were installed in a common trench in 1983. One was a 4-inch low vapour pressure liquid hydrocarbon pipeline and the other was a 34-inch natural gas pipeline. Both pipelines were externally tape coated and experienced external corrosion challenges. This paper will address the integrity challenges and solutions that were implemented for these two pipelines. 

Early random excavation and inspection of the 34-inch pipeline indicated that the pipeline was in relatively good condition. Over-the-pipeline surveys were then completed to identify focal points for inspection for both pipelines. A leak from the 4-inch pipeline leak triggered an in-line tethered survey that was performed from the failure location. This paper will compare over-the-line survey data with in-line inspection data and direct inspection data after the pipelines were excavated. Both pipelines were observed to experience the same mode of degradation. The consequence assessment for the 34-inch pipeline was significantly different than the liquid hydrocarbon pipeline. This paper will also discuss risk methodology differences and challenges for the two pipelines.


Assessments on un-piggable lines have proven to be a challenge for most operators.  In-line inspection tools have become the go-to method for most pipeline assessments, given the comprehensibility of the survey. For un-piggable lines, External Corrosion Direct Assessment (ECDA) surveys are one of the most effective means to perform an assessment. It is imperative that operators ensure the ECDA process can collect the required data to ensure a complete, thorough assessment of the line with out the use of ILI data to corroborate the findings.  The performance of an ECDA survey must meet applicable standards and regulations requiring a robust procedure which incorporates all data, ensures proper characterization of indications, and can incorporate all phases of the ECDA process to develop a full picture of the pipeline.  This presentation will explain pitfalls and places where operators may not have a fully developed procedure for ECDA and provide solutions for operators to ensure full compliance during execution of an ECDA. The goal of the presentation is to educate operators on the regulations and codes compared to the testing methods available to satisfy them.      


As the pipeline industry continues to move towards achieving zero pipeline incidents and underlying integrity data becomes more available and accessible, machine learning is emerging as a valuable practice to support the determination and validation of risk beliefs, measurement of root cause data and optimization of mitigation decision-making.

This paper presents the fundamental elements of machine learning as applied to linear and networked pipeline assets. Machine learning is simply a process to reveal useful patterns in data thru common methods found in linear algebra, descriptive and inferential statistics, and calculus. Underlying threat susceptibility and severity models are learned and validated through actual observations. The models then support the assessment of un‐piggable pipelines, dig prioritization programs, inference of missing data, prioritization of data collection activities, analysis of interactive threats, optimization of inspection intervals and selection of mitigative actions. More importantly, data-driven learned models are explicitly validated through actual observations, an often-overlooked concept in existing risk practices.


Industries have seen explosive growth of machine learning, and for good reason. Learning models drive state of the art performance in a variety of tasks. The In-Line Inspection (ILI) industry is no exception and applications of machine learning techniques have provided promising results for a wide range of needs. For any model to be successful, detailed attention is necessary at all stages of the process, as small nuances often manifest misleading results. This paper describes the development of simple, singular, and ensembled supervised machine learning algorithms designed to predict metal loss depth with an emphasis on nuances that could easily be overlooked. 

Supervised learning models use a mapped set of observations to truth to develop a mathematical function that can be applied to new examples. The input parameters, referred to as training data, are a vector of engineered features relating to the desired output. The quantity, quality and representation of the training data directly influence the performance of future predictions.

The data curation process is key in generating quality training data. It requires the collection, analysis and labeling of precisely correlated data between two measurement methods, such as ILI data and NDE data. Precisely matching the measurements from ILI and NDE can be difficult as they contain one-to-many and direct-to-indirect relationships. Significant attention must also be given to the distribution of features being represented in the training data. Uniform distributions of the desired prediction range e.g. 1%-100% depth and input features e.g. signal length, width and amplitude are challenging as the permutations can become extremely large. 

Model performance metrics require scrutiny as residual errors, highly correlated hold-out validation data and iterative development cycle bias lead to misleading statements about future prediction accuracy. Well distributed, blind-data, held outside the model development process, is key to accurately assessing future prediction accuracy.


This presentation focuses on leveraging historical in-line inspection (ILI) data to better understand the likelihood of failure associated with pipelines that have not yet been inspected. This helps to support the prioritization of asset integrity assessments such as Direct Assessment (DA) or investments to modify pipelines to make them piggable. There is significant scope for enhancing this process compared to current practice. 

Over the past 30 years huge volumes of ILI data have been collected, such that many uninspected pipelines can usually be compared with numerous similar inspected pipelines. This leads us naturally towards machine learning as an complementary monitoring solution for unpiggable, difficult to inspect, and low risk pipelines. Machine learning techniques help us to create accurate predictive models from relevant historical data. These models can then be used to support practical integrity management decisions.  

Using a pilot data warehouse containing ILI data for over 5,000 pipelines, we show how machine learning techniques can provide a route to understanding external corrosion behavior in uninspected pipelines.


Operators have a choice of a vast range of specialized integrity assessment software available, often this has required investment of significant resources and full value may not have been realized. Is it therefore even reasonable to expect that existing software is phased out before the planned amortization date or point of obsolescence in favor of a single solution?

The authors believe not; this however highlights the existing problem of compartmentalization within integrity management. An operator, inspection company or engineering consultancy may have implemented the leading software for each specific threat to integrity, but in many cases data transfer is limited to human intervention, often using spreadsheets. The risk caused by human induced errors, and of specific threats being overlooked is increased if integrity assessments are performed in non-interconnected silos.
Cloud infrastructure, artificial intelligence, proliferation or sensing devices, improvements in remote sensing technology, faster processors, access to skills and organizational cultural changes offer opportunities for the pipeline industry to better manage risk to pipelines. In a digital era why would a pipeline operator accept the results of an assessment based on data that can commonly be 6 months older or more? We know we can access live data from sensors such as corrosion probes, pressure and other operating parameters, so why would the assessment continue to be referenced to a single point in time? 

A key question might be; is digitalization enough, or should industry be aiming higher and be implementing an Industry 4.0 compatible strategy.

The authors aim to answer these questions, and demonstrate how improved data management, and improved formats for data transfer can allow implementation of Industry 4.0 pipeline integrity management which reduces compartmentalization and allows improved deliverables such as live integrity assessments and reporting, which overall can help operators reduce the risk of future pipeline failure. 


Following the introduction of the amended gas regulations, operators of gas transmission pipelines will be required to conduct a plethora of assessment and verification activities to achieve MAOP reconfirmation.  This will require the collection, organization, and analysis of a staggering amount of data. The effective execution of these activates will require a structured, methodical, and repeatable process and to document any reconfirmation activities and results.  This paper will provide an overview of the reconfirmation process, and how pipeline data systems can be used provide a platform to manage the data acquisition activities, the structure to support analysis of the acquired data, and verifying the results of this analysis within the System of Record (the pipeline GIS).  Examples using real data will be presented that demonstrate the process of material properties verification, MAOP verification, and how Pipeline GIS Data Models (PODS 7) can be leveraged to store data through the reconfirmation process. 


The paper describes the development and deployment of In-Line Inspection (ILI) tools for high temperature pipeline applications up to 125°C.  Initial target was a 16” pipeline belonging to an oil and gas operator in Indonesia.  Many design challenges needed to be first investigated and testing performed before being incorporated into the tools.  Testing, tool preparation, and specific project engineering was realized within a short time, due to then client’s operational requirements. 

These challenges included material selection and the electronic systems.  Suitable polyurethane material was identified and tested to be capable not only to withstand the high temperature and maintain good sealing and driving properties throughout the line.  Since the tool is MFL (Magnetic Flux Leakage), the magnetization levels achievable at the elevated temperatures were evaluated.  Other potentially temperature sensitive components such as cables and sensors were also proven in the testing.

Availability of such high temperature capable tools allows the operator, and potentially other operators, much more flexibility and freedom to plan and execute essential in line inspections.


Pipeline doors or “closures” are commonplace in the pipeline industry, providing access to the pipeline as well as to high-pressure equipment associated with the pipeline such as filters, separators, strainers, etc. Despite their prevalence, the importance of closures to the safe and efficient operation of a pipeline system is often overlooked. Recent changes in closure definitions and terminology warrant a review of the systems, applicable standards, designs and considerations related to choosing a closure for a desired purpose.

Closure can be defined differently, one definition, for example, is a pressure-containing component used to blank off an opening nozzle on a vessel or end of pipeline which could mean simply a bolted blind flange, a T-bolt cap. Others define a “quick-opening” closure as a pressure-containing component used for repeated access to the interior of a piping system. So clearly there are several ways the current codes can be interpreted, but what does it all mean?

In addition to the changes in definitions and terminology on the closures, during the design of the traps, many different codes and standards may be applied. Both the product that will be transported and where the equipment will be located can impact on the materials and the design of the traps. This can include the transitions from one design code to another, commonly referred to as breaks in the specifications, or ‘spec breaks.”  

This paper will focus on quick-actuating and quick-opening closures, presenting a history of pipeline closures from the early development to recent innovations. Also, the paper will address the issue of spec breaks and how designers and owners can benefit from the right choices for safe, cost effective and code compliant launch and receive facilities.  


Midstream infrastructure is the lifeblood of the oil and gas industry that allows products to easily and safely move around the country.  In cases where there is not a suitable pipeline available for a specific product, or economics prevent the construction of a new pipeline, then owners and operators can be faced with a difficult problem.  One of the solutions to this problem is to convert an existing pipeline product to the new desired product.

History has shown that pipeline conversions can present their own set of unique challenges that require careful consideration and a special set of expertise. The existing product must be removed and the pipeline prepared for any necessary mechanical work, the pipeline cleaned to prevent contamination of the new product, and the pipeline commissioned into new service while meeting product specifications on the first fill.  All of this must be done while considering the financial impact to the Operator while the line is out of service during the conversion process.

By applying some basic rules, Baker Hughes Process & Pipeline Services and Energy Transfer Partners worked together to successfully convert the 527-mile x 12-inch White Cliffs pipeline from crude oil to Y-Grade NGL service in 2019.  This was a heavy paraffin crude oil pipeline that was successfully converted to a high specification/clarity product pipeline and brought into service on the first fill by following five key steps.

The five key steps: Energy Transfer Partners engaged Baker Hughes early in the project, the project goals and requirements were clearly identified, communication lines were kept open, contingencies were planned for, and a plan to measure/validate success was implemented.  

This paper aims to provide details and outline a proven process that owners and operators can implement in order to successfully execute pipeline conversion projects.


From the use in ancient Rome to the early use in the oil industry, pigs still play an important role in the operations and maintenance of pipelines. Pigs play a vital part of the pipeline's life cycle from the commissioning phase, where specific pigs are used to prepare the pipeline for service, then into operation, where pigs provide a variety of purposes, through decommissioning, when the pigs are used to de-inventory and clean a pipeline as it is retired.

Given the important phases that a pipeline sees in its life, pigs play very different roles and given there are such a variety of pigs now on the market, it is important to make the right choice. Running a pig is often a major disruption to normal operations and hence it is imperative that the right pig is selected for the desired impact. When prepping a new line for service the pigs may be a combination of foams and multi-purpose, modular type pigs, while an operational pipeline may warrant more aggressive tools with heavy duty metal fingers to reach into corrosion pits.

A striking observation is the need to change the pigs used as the pipeline moves through its life cycle. Pigs used for commissioning may not be the right tools to remove debris during operation and certainly not for decommissioning. This paper explores the pipeline life cycle and the pigs that would be best applied to the need at the point in time. Also, the thorny issue of how clean is clean will be addressed as well as how often to re-run tools to get the maximum impact. 


Part of integrity management, pigging is performed as a standard, regular operational activity throughout the pipeline network’s lifecycle. Pipeline operators have been running pigs successfully for years. However, every so often, a problem occurs and a pig becomes stuck, stalled or damaged in the pipeline.

When another service provider’s bi-directional pig became caught up in a production tee during routine operational pigging, the operator of a 240-km, 28-inch gas export pipeline in southeast Asia contracted T.D. Williamson (TDW) to recover it.

Fortunately, the pig had stopped half way into the tee and was not blocking production flow entirely. However, the concern was that the pig could move further into the line and completely obstruct production flow, leading to shutdown of production gas.

After considering multiple engineering options for recovering the bi-directional pig, the operator, together with TDW decided to use mechanical means. TDW designed, manufactured a bespoke recovery tool then tested it in a mock-up of the launcher and tee that replicated the stuck pig scenario. This enabled a successful pig recovery operation offshore. 

This paper describes the steps taken to execute the pig recovery, including planning, site visit, engineering, tool manufacturing, testing and execution.  


Traditionally, Gulf of Mexico (GoM) Outer Continental Shelf (OCS) owners and operators have sought alternate methods to assess the integrity of their gathering lines than inline inspection.  In April 2010, the Deepwater Horizon explosion, and subsequent blowout, brought significant scrutiny from both state and federal regulatory agencies.

A customer approached Intero Integrity Services in 2019, in order to evaluate feasibility of inline inspections of its liquid gathering system in the GoM.  The feasibility included three areas:  1) Technical ability to inspect; 2) Cost of inspection; and 3) Value of reported results.

From a technical standpoint, the customer conditions challenged every aspect of standard inline inspections:  Never pigged for maintenance, nor integrity assessment; 1500 psi static pressure; limited platform workspace; remote location (90+ miles offshore); inability to track during the inspection (2,000’ water depth); limited pipeline availability based on production schedule; 8” nominal x 0.812" wall thickness X 25 mile pipeline; back to back “jumpers” to loop pipeline; and pumped in sea water.

From a cost perspective, the customer had limited assessment options.  The field was originally developed in 1998, and has operated in multi-phase production for 20+ years.  If pipeline replacement is required, the field will shut-in based on ROI compared to the current production curve.  External assessment methods are limited (ROVs and divers) and not comprehensive.  Other than inline inspection, hydrotest was the only option.  Again, not valuable information for the cost.

With the 0.812" wall thickness, the only inline inspection solution was ultrasonic (UT) technology.  Magnetic (MFL) tools would not be able to saturate the pipe wall to provide accurate measurements.  Intero has successfully inspected similar pipelines in the North Sea for decades, so not only the UT inspection technology, but also the historical database of inspection measurements from Intero made a perfect solution for the customer. 


PHMSA has modified the rules governing natural gas transmission pipelines. Before these changes, Operators re-established, or verified, pipeline material properties through laboratory testing of cut-outs. PHMSA’s 49 CFR 192 modifications allows Operators to utilize state-of-the-art nondestructive testing technologies to establish ultimate tensile and yield strength. Pacific Gas and Electric Company’s position on materials and MAOP verification is to estimate a feature’s most probable grade in order to use the resulting specified minimum yield strength for design pressure verification. Since 2016, PG&E has been collaborating with Kiefner and Associates, Inc. (Kiefner) towards this goal. Presentations on this work were made at the previous three PPIM conferences and this paper presents significant algorithmic enhancements relying on Kiefner’s data analytics capabilities. The grade calculation algorithm takes in-situ nondestructive testing results for strength and chemical composition as inputs to predict most probable grade. The training and development of this algorithm relies on Kiefner’s in-house material testing database of approximately 1500 unique laboratory samples. Properties of the samples in this database include chemical composition, mechanical properties, vintage, and grade data. This paper presents efforts towards enhancing the algorithm’s training, validation, and test sets. More specifically, we will discuss our approach to identifying data bias, filling database gaps, detection of outliers, and database improvement by over-sampling. We will also demonstrate the importance of data visualization as an integral part of database reliability and quality assessment. This work has led to quantifiable improvements in robustness of the grade prediction algorithm and increases confidence in its predictions. 


Driven by safety and added regulatory requirements, operators are working diligently to characterize the material properties of pipe within their assets. Opportunities to remove pipe from the ground and perform traditional mechanical testing in a laboratory are limited. Fortunately, operators are able to use a combination of enhanced ILI and state-of-the-art in-ditch testing to characterize material properties. This paper presents the most robust process available to ensure accurate data is collected – from the segments where it is required and with statistical relevance. ILI establishes a baseline that is used to determine where excavations are required and to align opportunistic digs with a unique knowledge of what is in the pipeline, aligned with operator records. In-ditch testing is used to gather accurate and complete information from the excavation; it enhances the verification process and increases the confidence that the material properties are adequately characterized. The product is compliant with traceable, verifiable and complete (TVC) pipeline material properties that can be used for effective integrity programs and to satisfy regulatory requirements.  


Recent changes to federal rules governing the operation of natural gas pipelines allow operators to use nondestructive examination (NDE) technologies for materials verification. Since 2013, PG&E has been evaluating instrumented indentation testing (IIT) to estimate yield strength (YS) using a qualification sample set >100 line-pipe features. Comparisons of NDE YS estimates with tensile test YS results have revealed differing trends based on pipe vintage and manufacturing process. E.g., two steel samples may exhibit similar results for YS from tensile testing but different YS estimates from IIT. IIT algorithms rely on empirical relationships that are based on the strain-hardening exponent to estimate YS. These empirical relationships may underrepresent the influence of microstructure. 

Pipeline manufacturing standards, e.g., API 5L, do not control microstructure. Therefore, manufacturers produce line-pipe to meet composition and mechanical properties specifications, regarding microstructure as a byproduct of the steelmaking and pipemaking processes. Microstructure, however, provides key insight about the deformation behavior and strengthening mechanisms of the steel, which then govern the strain-hardening rate. Metallography and microscopy were employed to reveal the influence of microstructure on NDE YS. Preliminary results of this analysis indicate that certain clusters of similar microstructures affect the agreement between NDE YS and tensile test YS.


PHMSA’s modifications of the federal rules governing pipelines calls for the verification of pipeline material properties. The material properties considered include chemical composition. At the 2019 PPIM conference, PG&E reported on the performance of nondestructive examination (NDE) to estimate pipeline steel chemical composition. NDE chemical composition data are a critical input to PG&E’s grade estimation algorithms. PG&E utilizes probabilistic grade estimations for its MAOP reconfirmation program. In 2019, we presented the performance of portable spark optical emission spectroscopy (POES) and filings analysis when compared against destructive laboratory composition results. Differences were highlighted that warranted further investigations into possible chemical composition through-wall segregation for vintage pipeline steels. We will report on our methodology, preliminary results, and limitations utilizing electron probe microanalysis, laser ablation inductively coupled plasma mass spectrometry, and laser-induced breakdown spectroscopy (LIBS) to examine the influence of through-wall segregation on differences between NDE and laboratory destructive composition results. Additionally, we expanded on our original investigation of NDE technologies by evaluating two new NDE methods, X-Ray fluorescence (XRF) and laser-induced breakdown spectroscopy (LIBS). Thus, the relative performance of four NDE methodologies will be presented, i.e., filings, POES, XRF, and LIBS.


Material verification is the process of measuring material property data on in-service pipelines when existing records are not traceable, verifiable, and complete (TVC). The new regulations for gas transmission pipelines include §192.607 which provides requirements for a material verification procedure. This work describes a statistical approach to meet these provisions by achieving a 95% confidence level on material sampling and conservatively accounting for measurement uncertainty when performing a nondestructive evaluation (NDE) of material strength properties. This analysis allows for the determination of whether a pipeline segment exceeds the expected grade, is more conservative than the expected grade, or requires additional testing and the use of additional techniques to make a final determination. Strength measurements collected during integrity excavations on more than 100 pipe joints with several pipeline operators are used to present the implications of applying this process to a diverse network of pipeline segments.


The mechanical properties of the materials involved are fundamental parameters in evaluating reliability. Conventional techniques for measuring these parameters cannot be used in-field because of such limitations as time-consuming procedures and destructive sample preparation.

The Instrumented Indentation Technique (IIT) can be a reasonable method for evaluating the mechanical properties of in-field facilities because it requires neither specific sample dimensions nor specimen fracture. IIT is also introduced in ISO as ISO TR 29381. This document mainly introduces the evaluation of tensile properties using IIT. Besides, American Welding Society (AWS) introduced IIT in its handbook for the measurement of residual stress.
We introduce a method for evaluating uniaxial tensile properties using IIT by a representative stress and strain approach. We have successfully estimated (within 10% of the reference value) the yield strength and ultimate tensile strength of materials by analyzing the elastic/plastic indentation factors.

In addition, we have found a close relation between some indentation factors and toughness. We propose an algorithm for evaluating toughness that uses a relationship between toughness and indentation factors such as the slope of the indentation curve and the integrated area under the indentation curve, we propose algorithm for evaluating toughness. 

To estimate residual stress, we analyze stress-free and stressed-state indentation curves to obtain the load difference at a given depth. From this result, we can evaluate the quantitative residual stress of the target region. Recently, we have suggested a novel way to estimate the load-depth curve of the stress-free using the indentation parameters that are invariant to the residual stress.

IIT and its models have been accepted in various industries for non-destructive testing (NDT) of gas transmission pipes for the strength to meet the new 192.607 regulation. Industry has been gathering the data for over five years using special permits from PHMSA under Operator Qualification (OQ). 



With the advent of machine learning, data based models can be used to increase efficiency and reduce cost for the characterization of various anomalies in pipelines. In this work, artificial intelligence is used to classify pipeline dents directly from the in-line inspection (ILI) data according to their risk categories. A deep neural network model is built with available ILI data, and the resulting machine learning model requires only the ILI data as an input to classify dents in different risk categories. Using a machine learning based model eliminates the need for conducting detailed engineering analysis to determine the effects of dents on the integrity of the pipeline. Concepts from computer vision are used to build the deep neural network using the available data. The deep neural network model is then trained on a sub set of the available ILI data and the model is tested for accuracy on a previously unseen set of the available data. The developed model predicts risk factors associated with a dent with 95% accuracy for a previously unseen data set 


The expected changes to pipeline regulations include provision for applying engineering critical assessment (ECA) of anomalies that would normally need investigation within a short period of time to extend the response time potentially to a monitored condition.  The caveat for the ECA analysis is that the analysis must be completed and documented within a period allowed for by the regulation, for example 10 days for immediate anomalies.  Having data and analysis on hand for an ECA is a key to meeting the deadlines as required by regulation.  This paper will provide some examples and scenarios where changes to the traditional approach to in-line inspection (ILI) can maintain the same level integrity while using ECA to place anomalies on a more reasonable schedule.  Examples are presented of approaches for addressing dents with metal loss, corrosion of or along a longitudinal seam, and stress-corrosion cracking.  Steps that can be taken in advance to prepare for rapid assessment within certain specified timeframes are discussed, as well as steps that can be taken that would improve the chances of achieving an extended response time. 


In-line Inspection (ILI) providers use defect interaction criteria to cluster individual ILI boxes, but the extent to which interaction criteria affect the number and severity of reported anomalies is not well understood.  Interaction criteria are used to cluster (group) individual anomalies (boxes) before calculating burst pressures and applying safety levels.  Different interaction criteria can add unnecessary digs to an operator's ILI response program without an appreciable increase in safety.  

This paper analyzes the effects of common interaction criteria on the resulting list of actionable anomalies. The effects of clustering with and without shallow corrosion is also considered.  The interaction criteria considered include those in industry references and range from “no interaction” to interaction at distances up to six times the wall thickness and larger. Using larger than necessary interaction criteria or including shallow corrosion in the clustering operation can artificially impact the calculated severity of reported clusters. The interaction criteria used should reflect when two or more anomalies structurally interact to reduce the calculate burst pressures. 

To illustrate the impact of different interaction criteria, the paper presents case studies, comparing the most severe anomalies as reported by the ILI after clustering has been applied. The case studies highlight the extent to which the interaction criteria affect real-world excavation and remediation decisions, with the ultimate goal of ensuring safety and conservative decision making at a reasonable cost.


Dents interacting with metal loss remain as a significant challenge to operators. Existing regulations require that dents with metal loss be treated as immediate repairs or 60-day repairs, resulting in costly excavations for many operators. At the time when these regulations were written, it was not clear whether in-line inspection technologies could discriminate the nature of the metal loss (i.e. corrosion or mechanical damage) or provide accurate sizing. Furthermore, advanced analysis techniques such as finite element analysis were limited, and fitness-for-service evaluations were not common. While the technological hurdles regarding evaluating interacting dent and metal loss features have been overcome, sensor lift-off remains a challenging issue for magnetic flux leakage (MFL) inspection tools, as sizing accuracy degrades at larger lift-off distances. Until recently, the sensor lift-off issue limited the ability to perform engineering critical assessments (ECA)  because the metal loss in dent features could not be confidently sized. This study demonstrates how integrated lift-off sensors can be used to quantify the lift-off as the MFL sensors pass over a dent. This technology integration has allowed the confident application of sizing specifications for many dents with metal loss, thereby permitting robust ECAs. Several case studies are examined in this paper, demonstrating how the integrated MFL and lift-off technology can serve to reduce excavations while still ensuring safe pipeline operations.  


The United Kingdom Onshore Pipeline Operators' Association (UKOPA) and ATCO Pipelines and Liquids, Canada have been investigating the quality of girth welds in vintage (pre-1972) gas and liquid pipelines in both the United Kingdom and Canada, welded to API 1104.  Workmanship acceptance criteria and QA/QC requirements for pipeline girth welds have evolved over time. The overall quality of vintage girth welds might not meet modern workmanship acceptance levels, but it can be demonstrated that the welds are not an inherent risk to pipeline integrity. 

A programme of testing on girth welds from pipeline assets (with varying diameter and wall thickness combinations) constructed prior to 1972 within the UK and Canada has been undertaken. Samples were analysed using non-destructive examination, chemical analysis and detailed microscopy. The results indicated that several of the welds tested contained a wide variety of welding defects that would not be acceptable according to modern pipeline welding standards (e.g. CSA Z662, API 1104 or BS 4515-1). Subsequent mechanical testing (tensile and Charpy V-notch) showed that the welds were over-matched and had an adequate toughness for the intended service.  Analysis of the results indicated that the girth welds were fit-for-service. Additionally, fatigue tests were performed on a sample of these welds, the results of which showed that the fatigue performance of the welds was conservatively predicted by the class E curve in BS 7608. 

The conclusion drawn from this work is that vintage girth welds are not an inherent risk to pipeline integrity. 


The anticipated “Gas Mega Rule” finally started the implementation process with the release of the first part in September 2019 for publication in the Federal Register.  The first part of the updated Gas Rule will address Congressional mandates and NTSB recommendations from the 2010 San Bruno gas pipeline rupture, and is expected to nearly double the size of 49 CFR Part 192, including provisions relative to MAOP confirmation/reconfirmation and expanded assessment criteria to newly defined areas known as Moderate Consequence Areas (MCAs).  At the same time, the Liquid “Mega” rule cleared the OMB process and is expected to be published soon after, it is intended to address Congressional mandates from the 2011 reauthorization of the Pipeline Safety Act, recommendations from the NTSB report on the 2010 Marshall Michigan oil pipeline rupture and GAO recommendations from 2012 regarding the collection of data from unregulated onshore hazardous liquid pipelines.

To address these new requirements, operators need to effectively apply proven tools to manage the added criteria to existing processes, or the additional mileage volume to current assessments.  This paper will illustrate advancement in engineering and technology processes and tools to aid in the safe and efficient management of records, confirmation of MAOP and class, and HCA / MCA assessments, in order to minimize the impact to current resources and budgets while ensuring compliance with the new regulations.