International Journal of Pressure Vessels and Piping 204 (2023) 104973

Contents lists available at ScienceDirect

International Journal of Pressure Vessels and Piping

journal homepage: www.elsevier.com/locate/ijpvp

Review

Strain-based design and assessment for pipeline integrity management: A

review of applications and gaps
Colin A. Schell a ,∗, Ernest Lever b , Katrina M. Groth a
a Systems Risk and Reliability Analysis Lab (SyRRA), Center for Risk and Reliability, University of Maryland, College Park, MD 20742, USA
b
GTI Energy, 1700 S Mt Prospect Rd, Des Plaines, IL 60018, USA

ARTICLE INFO ABSTRACT

Keywords: Pipelines have become a popular method of energy transportation for natural gas and oil due to their
Strain-based high reliability and efficiency. However, geohazards pose a significant threat to buried pipelines, as ground
Risk assessment movement can excessively deform pipelines, leading to a leak or rupture. Given the unpredictable nature of
Pipeline integrity
these geohazards, it is crucial to ensure the design of durable pipelines and to actively maintain their integrity.
Geohazards
In the past several decades, the pipeline industry has increasingly adopted strain-based design and assessment
(SBDA) methods in order to manage pipeline integrity challenges stemming from geohazards. Though the
use of these methods has steadily increased and some standards have adopted SBDA methods into their
pipeline design codes, there are still several challenges in integrating SBDA methods into a complete risk
assessment framework. This work presents a brief introduction to SBDA methods as well as a review of the
current research efforts. Gaps and opportunities for future work are identified and discussed with a focus on
SBDA’s role in pipeline risk assessment. The identified gaps suggest a need for system level modeling so that
causal relationships between geohazards and pipeline risk may be more clearly understood. Finally, connections
between SBDA and probabilistic risk assessment are presented to provide a road map for moving towards an
integrated pipeline integrity management framework.

1. Introduction cross geohazard prone terrain, ground movement poses a significant

risk to pipeline integrity.
Pipelines have become the primary transportation method for natu- Expressing the effects of external loads on a structure is typically
ral gas and oil resulting in over 2.6 million miles of pipelines stretching done using stress and/or strain. Stress is a measure of the structure’s
across the United States [1]. The majority of these pipelines are buried internal forces while strain is a measure of the structure’s deformation.
underground in order to conceal and protect the pipelines. However, Historically, the pipeline industry has used stress-based design methods
buried infrastructure is prone to ground movement geohazards such as for pipeline integrity management by limiting hoop stress and by
earthquakes and landslides. If left unchecked, or when combined with designing the pipeline not to yield [5]. However, stress-based design
other pipeline defects such as corrosion, cracks, and dents, excessive methods struggle to accurately design for geohazards. The first reason
ground movement can damage pipelines and result in oil or natural for this is that loads generated by geohazards are not readily available;
gas releases. The average failure rate of onshore pipelines due to the distance that the ground moves is easier to measure than the
geohazards is low but it is important to note that a large percentage of resulting forces [6]. Ground movement is primarily characterized as
the pipeline network crosses terrain where geohazards are unlikely to a displacement-controlled load (loads that are subject to change based
occur [2]. The pipeline incident 20 year trend data [3] collected by the on the structure’s deformation) as opposed to load-controlled loading
U.S. Department of Transportation Pipeline and Hazardous Materials where the load is independent of the structure’s deformation, such as
Safety Administration (PHMSA) recorded that earth movement hazards weight force. In reality, pipelines typically experience a combination
were responsible for 1% of pipeline incidents and $391M US dollars in of displacement and load-controlled loading [7]. However, geohazards
damages from 2002 to 2021. However, Sweeney et al. estimated that resulting in ground movement are primarily displacement-controlled
most geohazard-induced pipeline incidents occur in mountainous ter- and as such, measures of displacement are more readily available
rain containing just 5% of the pipeline network [4]. For pipelines that than measures of forces. Secondly, many stress-based design methods

∗ Correspondence to: 0144.B Glenn L. Martin Hall, 4298 Campus Drive, College Park, MD 20742, USA.
E-mail addresses: [email protected] (C.A. Schell), [email protected] (E. Lever), [email protected] (K.M. Groth).

https://doi.org/10.1016/j.ijpvp.2023.104973
Received 6 February 2023; Received in revised form 4 May 2023; Accepted 8 May 2023
Available online 11 May 2023
0308-0161/© 2023 Elsevier Ltd. All rights reserved.
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

define failure by an equivalent stress exceeding the yield strength 2.1. Strain-based design and assessment
of the material but due to the ductile nature of steel and the thin-
walled design of pipelines, some loading conditions allow for pipelines This section aims to provide a brief overview of the SBDA methodol-
to deform past their elastic limit (a transition point marked by the ogy. This discussion is not exhaustive but there exist detailed descrip-
material’s yield strength) and remain operational [7]. Not only does tions of these methods elsewhere [6,8,10–13]. Onshore and offshore
this make the failure criteria of many stress-based design methods pipelines are exposed to displacement-controlled loads during instal-
inappropriate for ground movement scenarios, but stress also becomes lation and operation. Onshore pipelines may experience landslides,
ground subsidence, and frost heave, while offshore pipelines are reeled
an ineffective failure indicator in the plastic region as large changes in
during installation, may experience temperature induced deformation,
strain may correspond to small changes in stress. Strain-based design
and may experience earthquakes and general soil movement [14].
is not a replacement to stress based design but in order to design
Onshore, buried pipelines will be the main focus of this paper but
pipelines to withstand geohazards, the industry has begun adopting the general methods discussed in this section are also applicable for
strain-based design methods as a solution to these noted shortcomings offshore pipelines.
of stress-based design methods. Strain-based design methods are performed before installation with
The basic concept of strain-based design is that a pipeline has a specific design allowances taken from a standard with strain-based
strain capacity, the maximum allowable strain that a pipeline can design equations as discussed in Section 2.1.3. These standards describe
withstand before failing, and a strain demand, the strain imposed on the strain-based design using limit state functions which are classified into
pipeline by its environment [8]. When the strain demand exceeds the groups (serviceability, ultimate, fatigue, and accidental) depending on
strain capacity of a pipeline, there will be a decrease in its load carrying the consequence for if strain demand exceeds strain capacity. SBDA
capacity, potentially leading to tensile rupture and/or compressive limit state functions are defined as strain demand subtracted from strain
(localized) buckling. A pipeline’s strain capacity is estimated based on capacity and if the result is less than or equal to 0, failure is expected
the pipe material, geometry, weld profiles, and nearby defects while to occur. Pipeline designers must also inspect the pipeline to ensure
that its welds are of sufficient quality and ensure that the pipeline is
strain demand is estimated based on the pipeline’s environment and
carefully installed to limit any accumulated strain at this phase [13].
its operating conditions. More recently, strain-based design has been
Strain-based assessment is performed after pipeline installation and
expanded to strain-based design and assessment (SBDA). Strain-based
uses site-specific estimates of strain demand. Experts will often perform
design takes place at the design phase of a pipeline and accounts for
quick strain-based assessments to screen for hazards and progress to
future pipeline hazards while strain-based assessment is performed on more detailed and thorough assessments in order to reduce uncertain-
buried and operating pipelines to understand the nature of current ties. The general process for performing SBDA for a buried pipeline can
pipeline threats [6]. In the past few decades, researchers from varied be seen in Fig. 1 and will be discussed in the following subsections.
disciplines have expanded the capabilities of strain capacity and strain
demand models. However, based on a review of these works, SBDA 2.1.1. Strain capacity
research efforts are not fully standardized and have not been properly Strain capacity is defined as the maximum strain that a pipeline
integrated into a system-level methodology. SBDA efforts could benefit of a specified set of parameters can withstand before failing. The type
from a road map focused on bringing its sub-disciplines together into of failure depends on which strain capacity is being calculated, tensile
a system-level pipeline integrity management framework. A logical strain capacity (TSC) or compressive strain capacity (CSC). These strain
source for this framework is probabilistic risk assessment (PRA), given capacity models are mostly developed using a combination of experi-
its strength in systematically modeling complex engineering systems mental testing, finite element method (FEM), and analytical methods.
and their hazards [9]. As such, this paper has been written with the Some standards present equations that calculate strain capacity but
many more sophisticated models cannot be translated into surrogate
following three objectives in mind:
models and require expert, site-specific analysis. TSC and CSC are the
two primary components of SBDA but some standards also include
• Presenting the current state-of-the-art methods in strain-based
ovalization limits for use in strain-based design.
design and assessment (SBDA) with a focus on its role in pipeline
Tensile strain capacity: TSC has been the primary subject of
risk assessment and integrity management.
SBDA research due to the prevalence of tensile strain failures. Ground
• Reviewing and categorizing the current SBDA applications and movement or thermal expansion can result in either bending strains or
research directions. uniform axial strains. Bending strains consist of a compressive strain at
• Establishing connections between SBDA and probabilistic risk as- the inner bend-radius of the pipeline and a tensile strain around the
sessment (PRA) for an integrated pipeline integrity management outer bend-radius. Excessive tensile strain can promote crack initiation
framework. and growth at existing flaws and result in a leak or rupture. The most
common location of tensile leak/rupture is at pipeline girth welds
The organization of the paper is as follows. Section 2 presents as welding often results in material defects and negatively affects
general background information for strain-based design and assessment the microstructure of the surrounding pipe steel. Additionally, weld
as well as probabilistic risk assessment. Section 3 describes the methods material is often stiffer than the pipe steel which results in excessive
used to conduct the literature review and Section 4 discusses the stresses at the girth weld when displaced. The importance of weld
reviewed literature. Section 5 is where the gaps and opportunities for quality in determining TSC makes it particularly difficult to calculate
TSC for vintage pipelines that were held to different weld standards or
future work are discussed. Section 6 describes how strain-based design
have incomplete or missing construction records [11]. These missing
and assessment can be integrated with probabilistic risk assessment and
or incomplete construction records also make it difficult to estimate
finally, Section 7 presents the paper’s conclusions.
important material properties such as yield strength [15]. Tensile strain
models are also used to inform new weld standards as weld quality
can be a limiting factor for TSC [16]. However, this is not to say that
2. Background
pipelines only fail at girth welds; failures can also occur in the pipe
body when flaws are present in the wall of the pipeline and girth welds
This section will briefly discuss the basics of strain-based design and are of a sufficient quality. The most commonly considered factors in
assessment (SBDA) and probabilistic risk assessment (PRA). predicting TSC are listed below [17].

2
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

Fig. 1. A general flowchart illustrating how different areas of knowledge are connected to each other in strain-based design and assessment.

∙ Pipe diameter a pipeline’s cross-section and is typically due to external compres-

∙ Wall thickness sive forces. Excessive ovality means the pipeline is deviating from
∙ Pipe steel and weld steel material strength its circular form and as a result, may have a reduced load carrying
∙ Pipe steel and weld steel strain hardening capacity or damaged protective coatings. Ovality mostly depends on
∙ Pipe steel chemical composition the diameter/wall thickness ratio and the amount of bending strain
∙ Weld steel toughness present [7].
∙ Weld profile
∙ Weld misalignment 2.1.2. Strain demand estimation
∙ Heat affected zone (HAZ) softening Strain demand is defined as the estimated applied strain for a
∙ Defect properties given pipeline segment. Strain demand can be measured using iner-
∙ Internal pressure tial measurement units (IMUs) or direct measurement methods if a
∙ Loading rate pipeline is already suspected to have deformed. Alternatively, pipe-
soil interaction models can be used to predict a pipeline segment’s
Compressive strain capacity: CSC has not been studied to the strain demand given knowledge of the surrounding ground movement
extent that TSC has partially because CSC related failures occur less and geotechnical parameters. Each of these methods have their own
frequently. Excessive compressive strain due to bending can result in strengths and weaknesses so choosing the correct method depends on
buckling and wrinkling of the pipeline along the inner radius of the the scenario itself.
bend. This buckling and wrinkling rarely leads to a gas/oil release on its
own, but will leave the pipeline vulnerable to other failure mechanisms. • Inertial measurement unit: Inertial measurement units (IMU)
Stress concentrations created by the wrinkles may accelerate fatigue are sent through the pipeline of interest during a pigging opera-
damage due to excessive loading. Wrinkling and buckling can also tion and record the centerline profile of the pipeline. Successive
damage protective coatings leaving the pipe vulnerable to corrosion runs on the same pipeline highlight changes to the curvature
damage [6]. Additionally, if the buckling is severe enough, the re- of the pipeline allowing operators to locate areas of high bend-
duced pipe cross-section could even impede pigging runs though this ing strain [20]. Analyzing a single IMU run without prior data
is uncommon. offers limited information to the operator but there are some
Buckling has been commonly observed at pipeline bends, around new techniques for approximating pipeline bending strain using
girth welds [18], or at other inconsistencies in pipeline stiffness. Ad- a single run [21]. Additionally, recent advancements in pipeline
ditionally, any defects in the pipe body may also encourage a leak or sensing technology allows for the measurement of axial strains
rupture once the buckling or wrinkling has already occurred. The most using linepipe material magnetic properties [22].
commonly considered factors in CSC models are listed below [19]. • Direct measurement: There are two main methods for directly
measuring pipeline strain, strain gauges and fiber optic cable.
∙ Diameter/wall thickness ratio
∙ Yield strength/tensile strength ratio Strain gauges can be attached to pipelines at key locations and
produce highly accurate strain measurements. However, these
∙ Internal pressure
values are extremely localized as it is impractical to mount strain
∙ Defect properties
gauges across many pipeline segments. Fiber-optic cables can be
∙ Pipe uniform strain
installed over greater lengths of pipelines but the application of
∙ Pipe Lüder’s strain
fiber-optic cables to pipeline strain measurement is less mature
∙ Net-section stress
than the use of strain gauges [11]. A benefit of applying strain
Ovalization: Some pipeline design standards also set limits on gauges or fiber-optic cables is that it allows for real-time strain
the ovalization of a pipe. Ovalization refers to the deformation of monitoring unlike other methods. As a result, these methods

3
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

Fig. 2. A general probabilistic risk assessment framework from Moradi and Groth [29].

are best used to monitor at-risk pipeline segments where future 2.2. Probabilistic risk assessment
ground movement is expected.
• Pipe-soil interaction models: Pipe-soil interaction models calcu- Probabilistic risk assessment (PRA) is a formalized process for un-
late the expected pipeline displacement based on the surrounding derstanding how complex engineering systems operate and for evaluat-
ground movement and the local geotechnical data. There are ing system/component-level risks, commonly expressed as the product
a number of different pipe-soil interaction model types, each of the event probability and the consequence of the event [9]. PRA
requiring a different amount of time and expertise. Analytical also highlights what factors contribute most to system-level risks so
pipe-soil interaction models solve the problem by calculating the that decision makers may enact more effective risk management mea-
estimated pipe displacement profile given a ground movement sures. These factors may be human, software, or hardware elements of
profile and soil data. Other methods use finite element method the system. PRA elements are defined probabilistically which enables
(FEM) to model the pipe and the surrounding soil. These FEM uncertainty analyses in order to understand where knowledge may
models represent the soil either as a series of simple springs be limited within the system. A generalized PRA process taken from
connected to the pipeline, as a continuous medium surrounding Moradi and Groth [29] can be seen in Fig. 2.
the pipeline, or as a collection of discrete particles [23]. FEM PRA may also be known as quantitative risk assessment (QRA)
models require more time to solve, must be done on a case-by- depending on the industry performing the risk assessment (the nuclear
case basis, and typically require a FEM expert to produce trusted industry uses PRA while QRA is more common in the oil and gas
results. However, they are generally accepted to produce more industry). Despite the different names, a QRA analysis can also be
accurate strain demand estimations than the analytical models. done probabilistically. PRA has been incorporated into the management
Pipe-soil interaction models have been the subject of a significant of several other complex engineering systems such as nuclear power
amount of research and have only been briefly discussed here. A plants [43] and space systems [44]. Currently, the U.S. Department of
more in-depth discussion can be found in [23–25]. Transportation Pipeline and Hazardous Materials Safety Administration
(PHMSA) states that probabilistic risk models are the only models to be
Pipe-soil interaction models can be very powerful in estimating categorized as the ‘‘best practice’’ in all decision scenarios [45]. These
strain demand but do require ground movement data as an input. probabilistic methods require more time and data than other risk assess-
Monitoring and measuring ground movement can be done using a wide ment methods but the quality of their results is recognized throughout
variety of technologies, each with its own strengths and weaknesses. multiple industries. Risk assessment procedures are provided in several
The details of these methods lie outside of the scope of this paper standards such as ASME B31.8S as well as CSA Z662. The QRA process
but a detailed discussion can be found in [10]. There are a number provided in CSA Z662 can be seen Fig. 3 which has been adopted from
of geohazards that have the potential to move a pipe from its original CSA Z662 Annex B [30]. For more information on pipeline QRA, a
location including landslides, fault movement, ground subsidence, ero- review of pipeline risk models was performed in the PHMSA funded
sion, frost heave, avalanches, and many more [11]. Landslides, ground project, ‘‘Critical Review of Candidate Pipeline Risk Models’’ [46].
subsidence, and fault movement in particular have been extensively
studied for their potential to damage pipelines [11,12,26–28]. 3. Literature screening methodology

2.1.3. Current state of the art The literature surrounding SBDA covers many disciplines from frac-
Strain-based design methods are currently addressed in a number ture mechanics to geospatial analysis. Therefore, the structured review
of notable standards. Currently, CSA-Z662 is the only standard that method shown in Fig. 5 was used to search, filter, and organize the
includes TSC and CSC equations for onshore pipelines. These equations recent work of SBDA. The goal of the literature review was to un-
are meant to be widely applicable and thus provide conservative esti- derstand the current usage of strain-based design principles in the
mates. DNV-OS-F101 and API RP 1111 include TSC and CSC equations reliability modeling of onshore pipelines. The bolded words formed
but are targeted specifically for offshore pipelines [19]. Additionally, the basis of the keyword list (strain-based, reliability, pipeline, strain
some standards allow for the application of strain-based design but capacity, and failure). First, these keywords were used to search a set of
do not include specific guidelines [8]. These standards include ASME journals chosen based on their relevance to SBDA. These journals can
B31.8, API 1104, ABS 2001, and NEN 3650. be seen in Fig. 5. These same keywords were then used to search Google
In addition to these standards, there are several TSC models devel- Scholar and Web of Science so as not to neglect any excluded journals.
oped by the industry in recent years. Some notable TSC models include Candidate articles were identified by their title, keywords, recency
CRES/PRCI [17,31], SINTEF [14,32], ExxonMobil [33], JFE [34], U (published after the year 2000) and abstracts at this stage. Additionally,
Ghent [35], and TWI [36]. A more detailed discussion of TSC models project partners from an ongoing PHMSA funded project conducted
is provided by Wang et al. [16]. The same is true of CSC models. their own literature reviews and recommended articles based on their
Some notable CSC models include CRES/PRCI [19], JFE [37,38], C- search results. The total number of articles collected at this stage was
FER [39], University of Alberta [40], and Atkins Boreas [41]. A detailed 85.
discussion of CSC models including those from standards can be found The articles identified in the initial search were then read in detail in
in Andrews et al. [42]. order to confirm their relevance to the literature review. Articles were

4
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

Fig. 3. A risk management process from CSA Z662 [30].

filtered out if they did not include at least one of the following items: many pipeline utility and consulting companies would not reveal the
an investigation of how key input variables affect strain capacity, a full details of their integrity management research for security reasons.
novel methodology for calculating strain capacity, a novel methodology
for calculating strain demand, or the application of SBDA methods for
assessing pipeline integrity. After filtering out unrelated articles, 54 4. Current research efforts
papers were saved and sorted based on several criteria: whether or not
the model is probabilistic, whether or not the model used operational The identified articles covered a wide variety of research topics but
data, the output of the model, the methodology, and what variables the can be broadly grouped by what variable(s) are related to pipeline
researchers related to strain limits. The distribution of articles within strain. There are a litany of variables that must be considered in a
these criteria can be seen in Fig. 5. It should be noted that categories complete strain-based design/assessment so many researchers focus on
within ‘‘Methodology(s)’’ and ‘‘Variable(s) related to strain’’ criteria better understanding the effects of a small set of variables or inves-
are not mutually exclusive. For example, some researchers chose to tigating the effects of previously neglected variables in SBDA. These
investigate the effects of pipeline parameters and defect characteristics variables were categorized using five groups which will be used to
on strain in the same study. Additionally, the number of identified organize the discussion of the surveyed literature. The papers and their
articles published each year is displayed in Fig. 4 and shows the recent categorical assignments can be seen in Table 1.
increase in SBDA related research. Lastly, almost all of the papers
identified a specific gap in SBDA which were recorded and grouped into 4.1. Loading conditions
identified gap categories. These will be discussed in Section 5. It should
be noted that this literature review only contains work that is publicly The group loading conditions encompasses the study of how pipeline
available which has likely caused a sampling bias. It is assumed that loads affect a pipeline’s strain capacity. This includes the effects of

5
 C.A. Schell et al.
Table 1
Reviewed literature categorized and tabulated.
Literature Probabilistic? Model output Operational Variable(s) related to strain limits Methodology(s)
data?
Loading Pipeline Defect Geohazard Geotechnical FEM Analytical Experimental
conditions parameter parameter parameter parameter testing
Abdelmoety et al. [47] Yes Failure probability No – X X – – X – –
Abdulhameed et al. [48] No Strain capacity No X – X – – X – X
Agbo et al. [49] No Strain capacity No X – X – – – – X
Agbo et al. [50] No Strain capacity No X – X – – X – X
Agbo et al. [51] No Strain demand No – – – X X X – –
Anderson et al. [52] Yes Failure probability Yes – X X – – X X –
Bastola et al. [53] Yes Strain capacity No X – X – – X – X
Berg et al. [54] No Other No X – X – – X X –
Cao et al. [55] No Strain capacity No X X – – – X X X
Chen et al. [56] No Strain capacity No – – X – – X – X
Cheng et al. [28] No Other No X X – X X X – –
Cheng et al. [27] No Strain demand No – X – X X X – –
Dake et al. [57] No Strain capacity No X – X – – X – –
Fraser and Koduru [58] Yes Strain demand No – X – X X X X –
Gu et al. [59] Yes Failure probability No – X – X X X – –
Han et al. [60] No Strain demand No – X – X X X – –
Hasan et al. [61] Yes Failure probability No – X X – – – X X
Hertele et al. [62] No Strain capacity No – X – – – – X X
Jang et al. [63] No Strain capacity No – X X – – X – X
Jang et al. [64] No Strain capacity No X X X – – X X –
Jayadevan et al. [65] No Strain capacity No X X X – – X – –
Ji et al. [66] No Strain demand No X X – X X – X –
Kim et al. [67] No Strain capacity Yes X – X – – X X X
Koduru et al. [68] No Strain capacity No – X – – – X X –
6

Koduru [69] Yes Failure probability Yes – X – X X – X –

Law and Bowie [70] No Strain capacity No – X – – – – – X
Lee et al. [71] No Other No – X X – – X X –
Li et al. [72] No Strain capacity No – – X – – X X X
Liu et al. [73] Yes Failure probability No X X X – – X X –
Liu et al. [74] No Strain capacity No X X – X X X – –
Liu et al. [75] No Strain demand No X X – X X X – –
Ma et al. [76] No Strain capacity No – – – X – X – X

International Journal of Pressure Vessels and Piping 204 (2023) 104973

Melissianos et al. [77] Yes Other No – X – X X X – –
Noronha2010 [78] No Strain demand Yes – – X – – X – –
Nourpanah and Taheri [79] No Other No X X X – – X X –
Oestby et al. [32] No Strain capacity No X X X – – X – –
Sakonder et al. [80] No Strain capacity No – X – – – X X –
Sandvik et al. [81] Yes Failure probability No – – X – – X X –
Shen et al. [82] No Other No – X X – – X – –
Shuai et al. [83] Yes Strain capacity No X X X – – X – X
Vazouras et al. [84] No Strain capacity No – X – X X X – –
Wang et al. [85] Yes Failure probability Yes – – X X X X – –
Wang et al. [86] No Strain capacity No X X X – – X – X
Wang et al. [87] Yes Failure probability No X – – X X X X –
Wang et al. [88] No Strain demand No – X – X X X X –
Wu etal. [89] Yes Failure probability No – X – X – X – –
Wu et al. [90] Yes Failure probability No – X – – – X – –
Yang et al. [91] Yes Strain capacity No – X – – – X – X
Yi et al. [92] No Strain capacity No X – X – – X – –
Yoon et al. [92] Yes Failure probability No – – – – X X – –
Zeng et al. [26] No Other No – – – X – X X X
Zheng et al. [93] No Strain demand No – X – – X – X –
Zhou [94] Yes Failure probability No – X – X X – X –
Zhou et al. [95] No Strain capacity No X X X – – X – X
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

effects on strain capacity as most SBDA models only consider girth

welds. Lee et al. [71] found that existing models performed well in
weld strength over-matching scenarios but performed poorly in under-
matching scenarios and developed a new reference-strain method that
accurately modeled weld strength mismatch. Aside from weld param-
eters, material anisotropy has been the focus of numerous studies for
its effects on both TSC and CSC [62,68,80]. Lastly, buried pipelines
are typically lowered into trenches dug in the ground during the
installation phase. Cheng et al. [27] and Gu et al. [59] investigated how
different trench design parameters affect the pipeline strain demand in
ground movement scenarios.

4.3. Defect characteristics

Pipe defects are known to reduce strain capacity but the precise
relationships can be difficult to model. Girth welds commonly pro-
duce small cracks and inconsistencies in the material and as such,
a considerable amount of work has been done to better understand
Fig. 4. Bar chart displaying how many of the referenced articles were published in
how different cracks in or around girth welds might reduce strain
each year.
capacity. Sandvik et al. [81] studied cracked offshore pipelines and
did so probabilistically using the tangency criterion. Several input pa-
rameters, including crack depth, were assigned probability distributions
internal pressure or external loads on a pipeline such as bending loads. and the resulting strain capacity and strain demand distributions were
Jayadevan et al. [65] showed that fracture toughness (using CTOD used to calculate the probability of failure. Shen et al. [82] used a
as the measure of toughness) is sensitive to tensile loads at high FEM model to better understand the effects of internal surface cracks
plastic strains. Østby et al. [32] found that the same holds true for located in the weld seam or HAZ region on crack driving forces. Strain-
bending loads but that tensile loading encourages crack propagation based failure assessment diagrams were generated for different internal
more than bending loads do. Both papers conclude that when a pipeline surface cracks as well as for different levels of HAZ softening and weld
experiences plastic deformation, a linear relationship exists between strength mismatch. Shen et al. identified a number of combined effects
CTOD and strain, regardless of the type of loading on the pipeline. Dake as discussed in Section 5.2.
et al. [57] studied the effects of both internal pressure and uniform ten- The applicability of X80 steel for use in offshore pipelines was in-
sile loads on the fracture capacity of offshore pipelines. Multiple load vestigated by Bastola et al. [53] using FEM and full-scale pipe tests with
paths were simulated and pressure was found to increase CTOD, but varied crack dimensions and locations. Cracks in the HAZ region had a
the magnitude of this effect depended on the load path. Similarly, Cao large effect on pipeline integrity meaning HAZ softening could limit the
et al. [55] investigated the effects of strain paths on strain capacities. acceptable crack size for offshore applications of X80 steel. Similarly,
Strain capacity models were developed for the pipeline base metal, the Dake et al. [57] developed new failure-assessment diagrams and strain
HAZ material, and the weld metal and it was found that the location of capacity estimates for offshore pipelines cracked at the girth weld. Chen
the minimum strain capacity changed depending on the strain path. et al. [56] developed a model for predicting the TSC for X70 pipelines
For offshore pipelines, reeling is a necessary procedure for properly with a crack located next to the girth weld seam and tested a variety of
laying pipe but also deforms the pipeline significantly. Nourpanah and crack sizes. Berg et al. [54] tested the impact of circumferential crack
Taheri. [79] were able to accurately predict the fracture toughness growth and different constraint theories on the CTOD of surface cracked
(using the J-integral) of reeled, cracked pipelines allowing for more pipelines and found that modeling circumferential crack growth had
accurate strain capacity predictions for offshore pipelines. a large impact on CTOD, but only for certain crack sizes. Constraint
theories were also investigated by Li et al. [72] and it was found that
4.2. Pipeline parameters different constraint theories changed the fracture toughness predictions
dramatically. Jang et al. [64] used both CTOD and the J-integral in a
The group, pipeline parameters, contains the largest number of re- TSC model for pipelines with internal cracks located at the girth weld
viewed papers and contains work relating to pipeline material, ge- and parameterized many of their input variables, including crack sizes.
ometry, its girth welds, or its installation. Many papers fall in this Lastly, Jang et al. [63] tested the impact of using different fracture
category as it is common to include parametric studies of pipeline toughness measures as well as different limit state definitions (not
geometry. However, there are many other pipeline parameters that discussed in this paper but a description of these different limit state
are not fully understood. For example, the study of vintage pipelines definitions can be found in [16]) on TSC estimates.
has been a major challenge for SBDA due to the variance of vintage In addition to cracks, other defects such as metal loss and dents
weld qualities and the limited data surrounding their manufacture [96]. can reduce the strain capacity of a pipeline. Corrosion defects have
This is further shown in the work by Law and Bowie [70] where been extensively studied for their ability to reduce the burst pressure
the performance of several burst pressure and strain capacity models of a pipeline but corrosion defects can also have detrimental effects
were compared against each other using a variety of pipelines. The on a pipeline’s strain capacity in ground movement scenarios. Shuai
models especially struggled with predicting accurate strain capacities et al. [83] analyzed the affects of different corrosion features such as
for the vintage pipelines. Agbo et al. [50] developed a TSC model for spherical pitting corrosion and cylindrical grooved corrosion on the
cracked X42 vintage pipelines and Abdulhameed et al. [48] developed CSC of a pipeline. Zhou et al. [95], Liu et al. [73], and Kim et al. [67]
a TSC model for cracked X52 vintage pipelines. On the other hand, also modeled different corrosion features and studied their effects
Wu et al. [90] showed that building new pipelines with high strength on strain capacity. Hasan et al. [61] probabilistically characterized
steels encourages weld strength under-matching and investigated the transgranular stress corrosion cracking (TGSCC) using a strain-based
effects of weld misalignment and strength mismatching on failure approach. In-line inspection (ILI) tools can also be used to detect
probability using an SBDA approach. Nontraditional girth welds such as existing flaws in operational pipelines. Anderson et al. [52] used ILI
elbows and bends were also studied by Liu et al. [73] for their unique data to calculate the conditional probability of failure for flaws detected

7
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

Fig. 5. Literature review methodology and distribution of articles across different categories.

in an operational pipeline in order for operators to prioritize repairs to FEM model to predict the TSC of a pipeline crossing a strike-slip fault.
the most critical flaws. Wang et al. [85] used ILI data as an input into a Lastly, Ma et al. [76] used a small scale pipe test and a FEM model to
corrosion growth model and used both a stress-based and strain-based obtain new TSC and CSC values in the event of longitudinal landslides.
approach to calculate probability of pipeline failure due to earthquakes. Conversely, other researchers have used existing strain capacity
Dents have also been studied for their effects on pipeline strain capaci- models in tandem with strain demand estimation techniques to bet-
ties. Abdelmoety et al. [47] used a response surface method to calculate ter understand the relationship between these natural hazards and a
the probability of failure for a dented pipeline while considering the re- pipeline’s failure behavior. Cheng et al. [28] calculated the critical fault
rounding of dents. Liu et al. [73] developed a CSC model considering displacement for pipelines crossing oblique-reverse faults considering
plain dents and Wang et al. [86] developed a CSC model considering the effects of rock strata. Zhou [94] calculated the probability of
kinked dents. Lastly, Noronho et al. [78] developed new procedures to
tensile rupture and compressive buckling due to longitudinal ground
accurately estimate dent dimensions and bending strain using imperfect
movement using an analytical model. Wang et al. [87] calculated
ILI data.
pipeline failure probability for spatiotemporal earthquakes considering
a number of different sources of uncertainty. Additionally, Koduru [69]
4.4. Geohazard parameters
developed a Bayesian network to predict the probability of tensile
rupture and compressive buckling given slope movement activity.
SBDA’s main focus is in mitigating pipeline geohazards and the
relationship between different geohazards and pipeline strain demand
is complex. This has motivated many researchers to develop new strain 4.5. Geotechnical parameters
demand models for select geohazards. Wang et al. [88] developed a
strain demand model for frost heave displacement and Ji et al. [66]
Lastly, a small portion of the papers identified focused on geotech-
developed a strain demand model for pipelines subjected to thaw
nical parameters. This mostly includes parametric studies of soil pa-
slumping loads. Similarly, a strain demand model for pipelines at
rameters as different soils can result in very different pipe movements
a fault crossing was developed by Liu et al. [75]. Han et al. [60]
given the same ground movement. For example, Wang et al. [85]
used a FEM model to predict the strain distribution along a pipeline
accounted for geotechnical discontinuities (a change in the soil along a
experiencing a landslide. In an effort to model multiple geohazards,
pipeline segment) in their pipe-soil interaction model and found that
Agbo et al. [51] studied the effects that different permanent ground
deformation patterns have on strain demand. Zheng et al. [93] also these discontinuities increased the dynamic response of the pipeline
developed a strain demand model for a range of geohazards using the when exposed to earthquakes. Similarly, Zhou [94] and Fraser and Ko-
finite difference method. duru [58] studied how spatial variability in soil parameters along pipe
Using existing strain demand models, a number of studies have segments affect pipe movement when exposed to ground movement.
simulated pipelines exposed to natural hazards until failure in order to Yoon et al. [97] developed seismic fragility curves using a strain-
calculate theoretical strain capacities. Vazouras et al. [84] simulated a based approach with and without including uncertainty in their soil
pipeline crossing a strike-slip fault and calculated the resulting TSC, parameters. Variables such as internal friction angle, unit weight, and
CSC, and ovalization limit in order to identify the dominant failure cohesion were studied and the uncertainties were found to increase
mode. This model was meant as a method for operators to focus their the pipeline failure probability. The effects of many geotechnical pa-
assessment efforts towards the most likely failure mode in the event of rameters on pipeline strain were studied in the PHMSA funded project,
an active strike-slip fault. Melissianos et al. [77] also created a failure ‘‘Definition of Geotechnical and Operational Load Effects on Pipeline
mode identification model but for reverse faults. Liu et al. [74] used a Anomalies’’ [25].

8
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

Fig. 6. Bar chart depicting the number of times each gap category is mentioned in the reviewed work.

5. Strain-based design and assessment gaps and widely-applicable SBDA models that can be integrated into a
comprehensive pipeline risk assessment. Strain capacity models are
There are many opportunities for future work to improve the ac- often translated into equations making it possible for reliability mod-
curacy and applicability of SBDA methods. Almost all of the papers els [27,28,66] to leverage strain capacity estimates without having to
included in this review identified a gap in SBDA at the time of their run computationally demanding FEM models. Although strain capacity
publication. These gaps covered a wide array of topics but can be equations are not as accurate as FEM models, they are capable of
roughly categorized into nine gap categories as seen in Fig. 6. For some efficiently calculating strain capacity which could be applied to mul-
of these gap categories, there exists a large amount of literature ad- tiple pipeline segments given adequate data. As such, the application
dressing the gap; for others, there has been little to no associated work of analytical strain capacity equations in reliability models should
partially due to the gap only being identified recently and partially due be further explored. For strain demand, there are existing analytical
to the difficulty of addressing said gap. Table 2 contains the papers pipe-soil interaction models but these are typically only applied in
that have identified each gap, and the papers that are considered to specific scenarios due to their limited model scope. As stated in Zheng
address each gap. Papers were considered to address a gap if the gap et al. [98], there is a need for straightforward strain demand estimation
they identified was part of the motivation for their research or if the methods that are useful in a variety of scenarios. Ji et al. [66] developed
paper’s conclusions addressed a gap identified elsewhere. Though none an analytical method for estimating strain demand due to thawing
of these gaps should be considered addressed, gaps such as ‘‘model- landslides. Wang et al. [88] also developed a lightweight surrogate
ing the effects of defect parameters on strain capacity’’ has been the model to calculate frost heave induced strain demand based on a
focus of numerous research efforts. By contrast, the gaps ‘‘developing FEM analysis. Furthermore, Zheng et al. [93] built a simple strain
simple and widely applicable models for broader applications’’ and demand model using the finite difference method that can be applied
‘‘including input variable interactions in strain capacity and strain in a range of geohazards. To the best of the authors knowledge, the
demand models’’ have been mostly neglected and have become crucial potential for using these widely-applicable strain demand and strain
for improving the field of SBDA. From reviewing the literature, the capacity models in a comprehensive pipeline risk assessment has not
authors have identified another gap that builds off of these gaps which yet been explored. Additionally, SBDA researchers should continue to
is ‘‘research has not been properly integrated into system-level models’’. refine their lightweight strain capacity and strain demand models to
The difficulty in integrating this research into system-level models, as expand their scope and improve their accuracy to support future risk
well as the gaps deemed as being crucial for improving the field of assessment modeling efforts.
SBDA will be discussed in this section.
5.2. Including input variable interactions in models
5.1. Developing models for broader applications
Most of the work reviewed in this paper investigates the effects of a
The SBDA methods discussed in this paper model the effects of few select variables on strain capacity or demand. The work will often
ground movement on pipeline integrity. Ground movement hazards include a parametric study of these select variables while any other
are just one of many threats to pipeline integrity. Additionally, the variables considered in the study are kept constant. The selection of
current application of SBDA methods are limited in scope to a pipeline which variables to parameterize depends on the focus of the study and
segment due to the site-specific data that strain capacity calculations the parameter’s range of possible values. In some of these parametric
require. Extending the application of SBDA methods beyond a single studies, researchers have identified interacting variables or combined
pipeline segment could help decision makers identify at-risk pipeline effects (the effect of two variables on the output is greater than the
segments that require further analysis. Thus, there is a need for simple sum of their individual effects). Hertelé et al. [62] noticed a combined

9
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

Table 2
Papers that have identified and addressed each gap category.
Gap category Papers that identify the gap Papers that address the gap
Developing simple and widely applicable models [26,48,93] [66,69,75,88,93]
for broader applications
Including input variable interactions in strain [55,62,72] [55,62,72,82,85]
capacity and strain demand models
Identifying limitations due to data [58,87,97] [69,78,85]
quality/availability
Modeling the effects of pipeline environment [27,59,85,94] [27,59,85,94]
parameters on strain demand
Modeling pipeline strain demand due to specific [28,51,60,66,74–77,84,88] [28,51,60,66,74–77,84,88]
geohazards
Modeling the effects of weld parameters on strain [32,62,71,73,81,90,91] [32,62,71,73,81,82,90,91]
capacity
Modeling the effects of different materials and [49,50,55,68,70,80,89] [49,50,55,68,70,80,89]
material parameters on strain capacity
Including uncertainty in strain capacity and strain [69,72,81] [47,52,53,58,59,61,69,73,77,81,83,85,87,89–91,94,97]
demand models
Modeling the effects of defect parameters on strain [47,52–54,56,57,61,63,67,73,78,83,86,92,95] [47,48,50,52–54,56,57,61,63,64,67,72,78–83,85,86,92,95]
capacity

effect of pipe steel heterogeneity and internal pressure on TSC. Cao Variability in wall thickness, coatings, and soil variables were found to
et al. [55] also identified how strain paths affect strain capacities but affect the ultrasonic ILI tool accuracy. Pandey and Lu [102] considered
that these effects depend on the stress state of the pipe. Li et al. [72] the noise and sizing error inherent in ILI data but developed simple
stated that in-plane and out-of-plane constraints can have combined models for forecasting degradation growth rates for corrosion defects.
effects and pointed out the potential for other unforeseen interacting Inaccurate dimensions, probability of detection, as well as probability
input variables. Similarly, Shen at al. [82] found that weld strength of false alarms are all potential sources of uncertainty in ILI data.
mismatch and HAZ softening had variable effects on pipeline integrity These studies highlight the various data quality problems that must be
depending on the location and dimensions of an internal surface crack. considered for ILI data but from the reviewed literature, only Noronha
For deep cracks, weld strength mismatch and HAZ softening had almost et al. [78] connected the associated uncertainties to pipeline strain. It
no affect on the failure assessment curves. However, weld strength is possible that part of the reason why there is not much peer-reviewed
mismatch and HAZ softening still influenced the failure assessment work using ILI data or operational data is because companies do not
curves for pipelines with shallow cracks. One must be careful when want to make this information publicly available for security reasons.
integrating these works into more comprehensive system-level models. That being said, only five articles [52,67,69,78,85] used or provided a
Parameterizing a variable that was kept constant in the original study method for including operational pipeline data in SBDA models.
could lead to modeling dependent variables as independent from one Bayesian networks are probabilistic data fusion models and have
another. Additional work in identifying and modeling these combined become the preferred method for managing the integrity of corroded
effects is not only crucial for motivating future research on these pipelines [103]. The application of Bayesian networks to SBDA could
combined effects, but also for scaling these works into system-level help leverage operational data and demonstrate how data quality issues
models. propagate into uncertainty in strain capacity estimates. Koduru [69]
used a Bayesian network allowing for data fusion but the model only in-
5.3. Identifying limitations due to data quality/availability cluded a limited number of variables; a comprehensive set of Bayesian
network nodes must be developed to enable insight into the wide
An accurate SBDA model requires an immense amount of data given range of risk influencing factors that exist in a ground movement
the number of different variables that affect a pipeline’s reliability in scenario. Although data limitations add considerable uncertainty to
the event of ground movement. Many authors noted data quality as a current SBDA models, the use of data fusion models may help overcome
limitation to their work, most commonly pointing at a lack of geotech- these limitations.
nical data. Moisture content and soil parameters (which can vary along
a pipeline segment) change the mechanics of the soil which affects the 5.4. Developing system-level models
relative impact of ground movement on pipe movement [11]. Geotech-
nical expertise is often required for site-specific inspections to obtain The majority of papers reviewed in this paper chose to advance
both accurate soil data as well as ground movement measurements. the state of SBDA by building a deeper understanding of a select set
Increasing the availability of location specific geotechnical data could of variables such as soil parameters, defect dimensions, etc. However,
bolster the speed at which SBDA can be performed and limit the given the number of variables and disciplines in SBDA, there is a need
uncertainty in strain demand estimates. for additional system modeling efforts. The study of strain capacity is
In-line inspection (ILI) data is another source of data with data mostly isolated from studies of strain demand or geohazards and data
quality challenges that must be addressed in a complete SBDA model. from multiple sources is rarely fused together. The causal relationships
ILI data is gathered during pigging operations and conducts non- between geohazards and pipeline risk are difficult to communicate to
destructive examinations of a pipeline. ILI tools are commonly used decision makers through independent studies but by merging these
to detect defects such as corrosion, cracks, and dents but produce works together into system-level, causal models, a clear picture can be
imperfect data that must be processed to remove data errors. Tech- painted for the industry.
niques such as signal processing are commonly used to process ILI This is likely the most difficult gap to address as it requires other
data when assessing the integrity of corroded pipelines [99]. Dann discussed gaps to be addressed first; improving data quality, fusing data
and Huyse [100] showed that the largest reported corrosion defects together from multiple sources, identifying and understanding input
are often oversized when collected with magnetic flux leakage ILI tools variable interactions, and developing models for broader applications
and developed deterministic and probabilistic tools to correct for this are all necessary steps for building a useful system-level model. As
over-sizing bias. Wang et al. [101] developed a two-stage model for discussed in Section 5.1, the scope of current SBDA methods is lim-
calibrating ultrasonic ILI data as well as predicting the true metal loss. ited to a single pipeline segment making it difficult to apply these

10
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

Fig. 7. A general PRA framework overlaid with an idealized SBDA process and its associated gaps.

methods to pipeline risk models. Another problem with developing For each initiating event (geohazards), the analyst should identify
these system-level models is the lack of SBDA guidelines for onshore, any safety barriers that may protect the pipeline and the factors that
buried pipelines. CSA-Z662 is the only standard with compressive and affect the functionality of said safety barriers. Event trees are commonly
tensile strain capacity equations for onshore, buried pipelines and these used to represent these different failure scenarios. Logic Modeling is
equations are highly conservative and do not consider many of the where the analyst should compute the strain capacities and compare
variables listed in Section 2.1.1. Moving to system-level models is no them against the estimated strain demand. This should be done for each
simple task and requires direction in order to properly combine existing chosen limit state and the reserve strain capacity should be recorded.
modeling efforts and to identify the most important future work. A The next step is performing a Consequence Analysis for each failure
potential source of such a road map is PRA as discussed in Section 2.2. scenario. This may involve estimating the size of a potential hazardous
Establishing connections between SBDA and PRA could move the indus- release and the different losses that may result from said release in
try towards an integrated pipeline integrity management framework for order to calculate risk. The Results Assessment is where the analyst es-
improving pipeline safety. tablishes the conclusions of the assessment. The analyst should evaluate
the different risks, identify the critical pipeline risks, identify the im-
6. A path towards system-level pipeline risk models portant risk-influencing factors, and establish future steps. The analyst
must decide if the pipeline requires risk management measures or if
Integrating SBDA research into system-level pipeline risk models further assessment is required. It is possible the assessment stops here
is a long-term goal because the gaps discussed in Section 5 make it if the pipeline is not at risk. If the pipeline is at risk, risk management
should be the next step per the guidelines in CSA Z662. However, if
difficult to create widely applicable and accurate SBDA models for
the pipeline could be at risk but the uncertainties are too great yet
use in system-level modeling. Developing a road map for integrating
reduceable, then the analyst should return to the Objective and Scope
SBDA methods with PRA could quicken the pace that the industry
step to begin a more in-depth assessment.
develops these system-level models. A comprehensive pipeline integrity
The depicted gaps complicate the PRA step they lie within and
management framework takes time to develop but current pipeline
thus, the quality of the PRA as a whole. Identifying limitations due to
standards offer pipeline specific risk management guidelines and offer
data quality/availability makes the Information Assembly and Initiating
a clear path for combining pipeline failure models. Researchers must
Events Identification steps more difficult because the risk analyst will
take time to expand and add detail to a framework through preliminary
have to deal with limited data and the resulting uncertainty. Next,
modeling efforts.
developing simple and widely applicable models for broader applications
In an effort to move towards a comprehensive pipeline integrity
will allow for a more comprehensive Scenario Development step as
management framework, the general PRA framework presented in
models become more flexible and accurate. Including input variable
Section 2.2 was used as a template in which an idealized SBDA process
interactions in strain capacity and strain demand models will help improve
could be overlaid and is shown in Fig. 7. General steps in SBDA model performance in the Logic Modeling step. Additionally, a proper
were placed in the PRA step that they belong in along with the gaps Results Assessment that considers all pipeline risks is still challenging
mentioned in Section 5. The first step in performing a PRA using SBDA seeing as how research has not been properly integrated into system-
methods should be defining the Objectives and Scope. This is where the level models. By addressing the mentioned gaps and expanding upon
analyst should decide on the pipeline segment(s) of interest, the depth this preliminary framework, SBDA can be more clearly connected to
of assessment, the risk acceptability, etc. The Information Assembly PRA forming a systematic process for accurately modeling pipeline
step is where the data needed to perform the assessment is gathered systems and their hazards in order to develop a comprehensive pipeline
such as geotechnical data, details of the pipeline’s manufacture, ILI integrity management framework.
data, or satellite data for ground movement estimations. The data
gathered here will depend on the scope and chosen depth of assessment 7. Conclusions
from the previous step. Next is the Initiating Events Identification step.
The analyst should identify which geohazards pose a threat to the In this paper, the current state-of-the-art methods in strain-based
pipeline depending on the pipeline environment and take note of any design and assessment (SBDA) were presented with a focus on their
girth welds and large defects as potential failure locations. Scenario role in pipeline risk assessment and integrity management. The current
Development involves deriving a set of scenarios where the pipeline research efforts and applications of SBDA were surveyed, organized,
fails. The identified failure scenarios should be used to select which and discussed. From this review, a set of gaps within SBDA were iden-
limit states are needed. tified from the reviewed literature. The gaps most critical to pipeline

11
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

risk assessment and integrity management were discussed. These gaps References
can be summarized as follows:
[1] PHMSA, General pipeline FAQs, 2018, URL https://www.phmsa.dot.gov/faqs/
• Data that is required to perform accurate strain-based assessments general-pipeline-faqs.
[2] M. Porter, G. Ferris, M. Leir, M. Leach, M. Haderspock, Updated Estimates
is limited and disparate. Researchers should develop SBDA data of Frequencies of Pipeline Failures Caused by Geohazards, in: International
fusion models, such as Bayesian networks, in order to better Pipeline Conference, Volume 2: Pipeline Safety Management Systems; Project
leverage operational data and integrate different data sources. Management, Design, Construction and Environmental Issues; Strain Based
• Input variable interactions within SBDA models are not well Design; Risk and Reliability; Northern Offshore and Production Pipelines, 2016,
http://dx.doi.org/10.1115/IPC2016-64085, V002T07A003.
enough understood. Developing system-level models requires ac- [3] PHMSA, Pipeline incident 20 year trend, 2022, URL https://portal.phmsa.dot.
counting for the interactions between variables that span studies, gov/analytics/saw.dll?Go.
e.g., the relationship between weld strength mismatch and crack [4] M. Sweeney, A.H. Gasca, M.G. Lopez, A.C. Palmer, Pipelines and landslides in
location. rugged terrain: A database, historic risks and pipeline vulnerability, in: Inter-
national Conference on: Terrain and Geohazard Challenges Facing Onshore Oil
• There is a need for simple and widely-applicable SBDA models and Gas Pipelines, 2015, pp. 647–660, http://dx.doi.org/10.1680/tagcfooagp.
for broader applications. Researchers should explore the potential 32781.0045.
value of these models for integration into system-level modeling [5] Y.-Y. Wang, M. Liu, F. Zhang, D. Horsley, S. Nanney, Multi-Tier Tensile Strain
to support more exhaustive pipeline risk assessments. Models for Strain-Based Design: Part 1 — Fundamental Basis, in: International
Pipeline Conference, Volume 4: Pipelining in Northern and Offshore Environ-
• SBDA research has not been coalesced into system-level models. ments; Strain-Based Design; Risk and Reliability; Standards and Regulations,
Research efforts from different SBDA disciplines remain separated 2012, pp. 447–458, http://dx.doi.org/10.1115/IPC2012-90690.
limiting the method’s ability to illustrate exactly how geohazards [6] Y.-Y. Wang, Strain-based design and assessment - Concepts and gaps, in:
affect pipeline integrity. Pipeline Integrity Management under Geohazard Conditions, PIMG, ASME Press,
2020, http://dx.doi.org/10.1115/1.861998_ch3.
[7] B. Liu, X. Liu, H. Zhang, Strain-based design criteria of pipelines, J. Loss Prev.
Although the move towards system-level models is a long-term Process Ind. 22 (6) (2009) 884–888, http://dx.doi.org/10.1016/j.jlp.2009.07.
goal, preliminary modeling efforts may highlight specific areas within 010, Papers Presented at the 2007 and 2008 International Symposium of the
SBDA that require additional research. As a first step, SBDA steps were Mary Kay O’Connor Process Safety Center and Papers Presented at the WCOGI
connected to a general PRA framework to highlight how SBDA can 2007.
[8] M. Lower, Strain-Based Design Methodology of Large Diameter Grade X80
be applied to a pipeline integrity management framework. SBDA gaps Linepipe, Tech. Rep. ORNL/TM-2014/106, Oak Ridge National Laboratory,
were also mapped onto the PRA framework showing what work is 2014.
most critical and why. By establishing connections between SBDA and [9] M. Modarres, M.P. Kaminskiy, V. Krivstov, Reliability Engineering and Risk
PRA, SBDA methods can be properly utilized in an integrated pipeline Analysis, CRC Press, 2016.
[10] R.L. Baum, D.L. Galloway, E.L. Harp, Guidelines for Constructing Natural
integrity management framework in order to improve pipeline safety. Gas and Liquid Hydrocarbon Pipelines Through Areas Prone to Landslide
and Subsidence Hazards, Tech. Rep. dot:34640, C-CORE and D. G. Honegger
Consulting and S. S. D. Inc., 2009.
CRediT authorship contribution statement
[11] Y.-Y. Wang, Management of Ground Movement Hazards for Pipelines, Tech.
Rep. CRES-2012-M03-01, Center for Reliable Energy Systems (CRES), 2017.
Colin A. Schell: Methodology, Formal analysis, Data Curation, [12] Geosyntec Consultants Inc., Golder Associates Inc., Center for Reliable Energy
Systems (CRES), Guidelines for Management of Landslide Hazards for Pipelines,
Writing – original draft, Visualization. Ernest Lever: Conceptualiza-
Tech. Rep., Interstate Natural Gas Association of America, 2020.
tion, Writing – review & editing, Project administration, Funding ac- [13] W. Mohr, Strain-Based Design of Pipelines, Tech. Rep. 45892GTH, EWI, 2003.
quisition. Katrina M. Groth: Conceptualization, Methodology, Writ- [14] E. Østby, C. Thaulow, B. Nyhus, A new approach to ductile tearing assessment
ing – review & editing, Supervision, Project administration, Funding of pipelines under large-scale yielding, Int. J. Press. Vessels Pip. 84 (6) (2007)
337–348, http://dx.doi.org/10.1016/j.ijpvp.2007.02.002.
acquisition.
[15] J. Chen, D. Ersoy, Y. Liu, Probabilistic bulk property estimation using multi-
modality surface non-destructive measurements for vintage pipes, Struct. Saf.
Declaration of competing interest 87 (2020) 101995, http://dx.doi.org/10.1016/j.strusafe.2020.101995.
[16] Y.-Y. Wang, B. Liu, B. Wang, Tensile strain models and their applications, in:
Pipeline Integrity Management under Geohazard Conditions, PIMG, ASME Press,
The authors declare the following financial interests/personal rela- 2020, http://dx.doi.org/10.1115/1.861998_ch22.
tionships which may be considered as potential competing interests: [17] Y.-Y. Wang, M. Liu, X. Long, M. Stephens, R. Peterson, R. Gordon, Second
Generation Models for Strain-Based Design, Tech. Rep. DTPH56-06-T-000014,
Colin A. Schell, Katrina M. Growth, Ernest Lever reports financial
Pipeline Research Council International, Pipeline Research Concil International,
support was provided by U.S. Department of Transportation Pipeline 2011.
and Hazardous Materials Safety Administration. [18] D.W. Murray, Local buckling, strain localization, wrinkling and postbuckling
response of line pipe, Eng. Struct. 19 (5) (1997) 360–371, http://dx.doi.org/
10.1016/S0141-0296(96)00096-X, Stability of structures.
Data availability [19] M. Liu, Y.-Y. Wang, F. Zhang, K. Kotian, Realistic Strain Capacity Models
for Pipeline Construction and Maintenance, Tech. Rep. DTPH56-14-H-00003,
Center for Reliable Energy Systems, 2013.
No data was used for the research described in the article.
[20] B. Purvis, T. Huewener, Pipeline mapping and strain assessment using ILI tools,
2009.
Funding and acknowledgments [21] A. Young, K. Kol, R. Nijland, D. Dotson, Maximising value from single run IMU
inspection data, Pipelines Int. (2019) 18–21.
[22] I. Yablonskikh, M. ElSeify, J. Dawson, Strain demand and capacity assessment
This information, data or work presented herein was funded in based on in line inspection of axial and bending strains, in: Baker Hughes,
part by the U.S. Department of Transportation Pipeline and Hazardous Process & Pipeline Services, 2021.
[23] D. Yu, Y.-Y. Wang, B. Liu, X. Chen, A Review of Pipe-Soil Interaction Models
Materials Safety Administration’s Pipeline Safety Research and Devel-
for Strain Demand Estimation, in: International Pipeline Conference, Volume
opment Program, under Award Number 693JK32110002POTA. The 2: Pipeline Safety Management Systems; Project Management, Design, Con-
views and conclusions contained in this document are those of the struction, and Environmental Issues; Strain Based Design; Risk and Reliability;
authors and should not be interpreted as representing the official Northern, Offshore, and Production Pipelines, 2020, http://dx.doi.org/10.1115/
IPC2020-9678, V002T06A016.
policies, either expressed or implied, of the Pipeline and Hazardous
[24] M.J. O’Rourke, Response of Buried Pipelines Subject to Earthquake Effects
Materials Safety Administration, the Department of Transportation, or (Monograph Series / Multidisciplinary Center for Earthquake Engineering
the U.S. Government. Research), 1999.

12
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

[25] A. Fredj, A. Dinovitzer, Definition of Geotechnical and Operational Load Effects [51] S. Agbo, K. Roy, S. Adeeb, Y. Li, Effects of permanent ground deformation
on Pipeline Anomalies, Tech. Rep., BMT Fleet Technology, 2017. patterns on strain demand of pressurized buried continuous steel pipelines,
[26] X. Zeng, F. fei Dong, X.-D. Xie, G.-F. Du, A new analytical method of strain J. Pipeline Syst. Eng. Practice 13 (4) (2022) 04022042, http://dx.doi.org/10.
and deformation of pipeline under fault movement, Int. J. Press. Vessels Pip. 1061/(ASCE)PS.1949-1204.0000680.
172 (2019) 199–211, http://dx.doi.org/10.1016/j.ijpvp.2019.03.005. [52] T. Anderson, J. Andrew, J. Moritz, A probabilistic method for prioritizing
[27] X. Cheng, R. Huang, L. Xu, C. Ma, X. Zhu, Parametric study on the trench repairs following an ILI crack tool run, J. Pipeline Eng. 17 (4) (2018) 281–289.
designing for X80 buried steel pipeline crossing oblique-reverse fault, Soil [53] A. Bastola, J. Wang, H. Shitamoto, A. Mirzaee-Sisan, M. Hamada, N. Hisamune,
Dyn. Earthq. Eng. 150 (2021) 106824, http://dx.doi.org/10.1016/j.soildyn. Investigation on the strain capacity of girth welds of X80 seamless pipes with
2021.106824. defects, Eng. Fract. Mech. 180 (2017) 348–365, http://dx.doi.org/10.1016/j.
[28] X. Cheng, C. Ma, R. Huang, S. Huang, W. Yang, Failure mode analysis of X80 engfracmech.2017.06.010.
buried steel pipeline under oblique-reverse fault, Soil Dyn. Earthq. Eng. 125 [54] E. Berg, B. Skallerud, C. Thaulow, Two-parameter fracture mechanics and
(2019) 105723, http://dx.doi.org/10.1016/j.soildyn.2019.105723. circumferential crack growth in surface cracked pipelines using line-spring
[29] R. Moradi, K.M. Groth, Modernizing risk assessment: A systematic integration elements, Eng. Fract. Mech. 75 (1) (2008) 17–30, http://dx.doi.org/10.1016/j.
of PRA and PHM techniques, Reliab. Eng. Syst. Saf. 204 (2020) 107194, engfracmech.2007.03.023.
http://dx.doi.org/10.1016/j.ress.2020.107194. [55] J. Cao, W. Ma, W. Hui, K. Wang, K. Cai, X. Li, S. Cong, J. Ren, H. Nie, W.
[30] CSA, CSA Z662: Oil and Gas Pipeline Systems, Tech. Rep., CSA Group (CSA), Dang, Effects of strain path and girth weld heterogeneity on strain limits of
2020. X70 pipeline, Int. J. Press. Vessels Pip. 198 (2022) 104682, http://dx.doi.org/
[31] Y.-Y. Wang, M. Liu, X. Long, M. Stephens, R. Peterson, R. Gordon, Validation 10.1016/j.ijpvp.2022.104682.
and Documentation of Tensile Strain Limit Design Models for Pipelines, Tech. [56] H. Chen, J. Niu, Q. Chi, Y. Bi, Y. Wang, F. Yang, J. Ren, Strain capacity
Rep. DTPH56-06-T-000014, Pipeline Research Concil International, 2011. of girth weld joint cracked at ‘‘near-seam zone’’, Int. J. Press. Vessels Pip.
[32] E. Østby, K. Jayadevan, C. Thaulow, Fracture response of pipelines subject to 139–140 (2016) 77–85, http://dx.doi.org/10.1016/j.ijpvp.2016.03.006, Special
large plastic deformation under bending, Int. J. Press. Vessels Pip. 82 (3) (2005) Issue: ICPVT-14 International Conference on Pressure Vessel Technology.
201–215, http://dx.doi.org/10.1016/j.ijpvp.2004.08.012. [57] Y. Dake, I. Sridhar, X. Zhongmin, S.B. Kumar, Fracture capacity of girth welded
[33] X. Wang, S. Kibey, H. Tang, W. Cheng, K. Minnaar, M. Macia, W. Kan, S. pipelines with 3D surface cracks subjected to biaxial loading conditions, Int. J.
Ford, B. Newbury, Strain-based design-advances in prediction methods of tensile Press. Vessels Pip. 92 (2012) 115–126, http://dx.doi.org/10.1016/j.ijpvp.2011.
strain capacity, Int. J. Offshore Polar Eng. 21 (2011). 10.019.
[34] S. Igi, N. Suzuki, Tensile Strain Limits of X80 High-strain Pipelines, in: [58] A. Fraser, S.D. Koduru, Effect of soil variability on strain demand associated
International Ocean and Polar Engineering Conference, vol. All Days, 2007, with moving slopes, in: 11th International Pipeline Conference. Volume 2:
ISOPE-I-07-504. Pipeline Safety Management Systems; Project Management, Design, Construc-
[35] M. Verstraete, W. De Waele, R. Denys, S. Hertelé, Pressure correction factor for
tion and Environmental Issues; Strain Based Design; Risk and Reliability;
strain capacity predictions based on curved wide plate testing, in: International
Northern Offshore and Production Pipelines, 2016, http://dx.doi.org/10.1115/
Pipeline Conference, Volume 4: Pipelining in Northern and Offshore Environ-
IPC2016-64432.
ments; Strain-Based Design; Risk and Reliability; Standards and Regulations,
[59] X.-T. Gu, X.-R. Zang, Z.-Y. Zhang, P. Yang, W.-Z. Miao, P. Cao, B. Zhao,
2012, pp. 359–365, http://dx.doi.org/10.1115/IPC2012-90592.
Reliability analysis of large-diameter high-grade-steel natural gas pipelines
[36] S. Smith, Development of the BS 7910 failure assessment diagram for strain
under fault action, Pet. Sci. (2022) http://dx.doi.org/10.1016/j.petsci.2022.06.
based design with application to pipelines, in: International Conference on
017.
Offshore Mechanics and Arctic Engineering, Volume 3: Pipeline and Riser
[60] B. Han, Z. Wang, H. Zhao, H. Jing, Z. Wu, Strain-based design for buried
Technology, 2012, pp. 431–436, http://dx.doi.org/10.1115/OMAE2012-83527.
pipelines subjected to landslides, Pet. Sci. 9 (2) (2012) 236–241, http://dx.
[37] N. Suzuki, S. Igi, Compressive strain limits of X80 high-strain line pipes, in:
doi.org/10.1007/s12182-012-0204-y.
International Ocean and Polar Engineering Conference, vol. All Days, 2007,
[61] S.M. Hasan, F. Khan, S. Kenny, Probabilistic transgranular stress corrosion
ISOPE-I-07-506.
cracking analysis for oil and gas pipelines, J. Pressure Vessel Technol. 134
[38] N. Suzuki, J. Zhou, M. Toyoda, Compressive strain limits of high-strength
(5) (2012) http://dx.doi.org/10.1115/1.4006125, 051701.
linepipes, in: 2008 7th International Pipeline Conference, Vol. 3, in: Inter-
[62] S. Hertelé, K. Van Minnebruggen, M. Verstraete, R. Denys, W. De Waele,
national Pipeline Conference, 2008, pp. 737–744, http://dx.doi.org/10.1115/
Influence of pipe steel heterogeneity of the upper bound tensile strain capacity
IPC2008-64526.
of pipeline girth welds: A validation study, Eng. Fract. Mech. 162 (2016)
[39] C.M. Timms, D.D. DeGeer, M.R. Chebaro, Y. Tsuru, Compressive strain limits of
121–135, http://dx.doi.org/10.1016/j.engfracmech.2016.03.042.
large diameter X80 UOE linepipe, in: International Ocean and Polar Engineering
[63] Y.-Y. Jang, J.-Y. Kang, N.-S. Huh, I.-J. Kim, Y.-P. Kim, Strain-based CTOD and
Conference, vol. All Days, 2009, ISOPE-I-09-332.
J-integral estimations for pipelines with a surface crack under large plastic
[40] A.B. Dorey, D.W. Murray, J.J.R. Cheng, Critical buckling strain equations for
strain and internal pressure, J. Pressure Vessel Technol. 142 (5) (2020) http:
energy pipelines—A parametric study, J. Offshore Mech. Arct. Eng. 128 (3)
//dx.doi.org/10.1115/1.4046979, 051503.
(2005) 248–255, http://dx.doi.org/10.1115/1.2199561.
[41] M. Carr, I. MacRae, D. Bruton, Local buckling of pressurised seamless linepipe: [64] Y.-Y. Jang, N.-S. Huh, I.-J. Kim, Y.-P. Kim, J and CTOD-based tensile strain
Results of the SAFEBUCK JIP, 2009. capacity prediction for pipelines with a surface crack in girth weld, J. Pressure
[42] D.R. Andrews, M. Stephens, M. Carr, J. Brückner, A review of strain capacity Vessel Technol. 144 (1) (2021) http://dx.doi.org/10.1115/1.4051629, 011305.
assessment methods for transmission pipelines, J. Pipeline Eng. (2018). [65] K. Jayadevan, E. Østby, C. Thaulow, Fracture response of pipelines subjected to
[43] R. Bartel, WASH-1400 (The Reactor Safety Study), Tech. Rep. NUREG/KM-0010, large plastic deformation under tension, Int. J. Press. Vessels Pip. 81 (9) (2004)
United States Nuclear Regulator Commission, 2016. 771–783, http://dx.doi.org/10.1016/j.ijpvp.2004.04.005.
[44] M. Stamatelatos, H. Dezfuli, G. Apostolakis, C. Everline, S. Guarro, D. Mathias, [66] B. Ji, X. Liu, D. Bolati, Y. Yang, J. Jiang, Y. Liu, H. Zhang, Improved analytical
A. Mosleh, T. Paulos, D. Riha, C. Smith, W. Vesely, R. Youngblood, Probabilis- method for longitudinal strain analysis of buried pipelines subjected to thaw
tic Risk Assessment Procedures Guide For Nasa Managers And Practitioners slumping load, Front. Energy Res. 9 (2021) http://dx.doi.org/10.3389/fenrg.
(Second Edition), Tech. Rep. 20120001369, NASA, 2011. 2021.742348.
[45] PHMSA, Pipeline Risk Modeling: Overview of Methods and Tools for Improved [67] I.-J. Kim, Y.-C. Jang, Y.-Y. Jang, J.-H. Moon, N.-S. Huh, Estimation of tensile
Implementation, Tech. Rep., U.S. Deparetment of Transportation, 2020. strain capacity for thin-walled AP X70 pipeline with corrosion defects using
[46] S. Koduru, R. Adianto, J. Skow, Critical Review of Candidate Pipeline Risk the fracture strain criteria, J. Mech. Sci. Technol. 34 (7) (2020) 2801–2812,
Models, Tech. Rep., C-FER Technologies, 2016. http://dx.doi.org/10.1007/s12206-020-0613-6.
[47] A.K. Abdelmoety, M. Kainat, N. Yoosef-Ghodsi, Y. Li, S. Adeeb, Strain-based [68] S.D. Koduru, A.M. Fraser, N. Yoosef-Ghodsi, Q. Chen, Influence of material
reliability analysis of dented pipelines using a response surface method, J. anisotropy on the compressive strain capacity, in: International Pipeline Con-
Pipeline Sci. Eng. 2 (1) (2022) 29–38, http://dx.doi.org/10.1016/j.jpse.2021. ference, Volume 2: Pipeline Safety Management Systems; Project Management,
11.002. Design, Construction and Environmental Issues; Strain Based Design; Risk and
[48] D. Abdulhameed, C. Cakiroglu, M. Lin, R. Cheng, J. Nychka, M. Sen, S. Reliability; Northern Offshore and Production Pipelines, 2016, http://dx.doi.
Adeeb, The effect of internal pressure on the tensile strain capacity of X52 org/10.1115/IPC2016-64293, V002T06A006.
pipelines with circumferential flaws, J. Pressure Vessel Technol. 138 (6) (2016) [69] S.D. Koduru, A Bayesian network for slope geohazard management of buried
http://dx.doi.org/10.1115/1.4033436, 061701. energy pipelines, in: 13th International Conference on Applications of Statistics
[49] S. Agbo, M. Lin, I. Ameli, A. Imanpour, D.-M. Duan, J.R. Cheng, S. Adeeb, and Probability in Civil Engineering(ICASP13), 2019, http://dx.doi.org/10.
Experimental evaluation of the effect of the internal pressure and flaw size on 22725/ICASP13.444.
the tensile strain capacity of welded X42 vintage pipelines, Int. J. Press. Vessels [70] M. Law, G. Bowie, Prediction of failure strain and burst pressure in high
Pip. 173 (2019) 55–67, http://dx.doi.org/10.1016/j.ijpvp.2019.04.010. yield-to-tensile strength ratio linepipe, Int. J. Press. Vessels Pip. 84 (8) (2007)
[50] S. Agbo, A. Imanpour, Y. Li, M. Kainat, N. Yoosef-Ghodsi, J.J.R. Cheng, S. 487–492, http://dx.doi.org/10.1016/j.ijpvp.2007.04.002.
Adeeb, Development of a tensile strain capacity predictive model for American [71] J.-S. Lee, M.-H. Kim, Strain-based failure assessment based on a reference strain
petroleum institute 5L X42 welded vintage pipelines, J. Pressure Vessel Technol. method for welded pipelines, J. Offshore Mech. Arct. Eng. 142 (4) (2020)
142 (6) (2020) http://dx.doi.org/10.1115/1.4047561, 061506. http://dx.doi.org/10.1115/1.4045917.

13
C.A. Schell et al. International Journal of Pressure Vessels and Piping 204 (2023) 104973

[72] Z. Li, B. Gong, G. Lacidogna, C. Deng, D. Wang, Strain-based fracture response [90] K. Wu, H. Zhang, Y. Yang, X. Liu, Strength matching factor of pipeline girth
of X80 steel pipe welded girth based on constraint-modified J-R curves: From weld designed by reliability method, J. Pipeline Sci. Eng. 1 (3) (2021) 298–
SENT specimen to full-scale pipe, Eng. Fract. Mech. 258 (2021) 108114, http: 307, http://dx.doi.org/10.1016/j.jpse.2021.09.002, Special Issue on Risk and
//dx.doi.org/10.1016/j.engfracmech.2021.108114. Reliability Assessment of Pipelines.
[73] M. Liu, H. Zhou, B. Wang, Y.-Y. Wang, J. Bergman, B. Ayton, M. Stephens, [91] Y. Yang, H. Zhang, K. Wu, P. Chen, Y. Sui, D. Yang, X. Liu, Strain capacity
T. Weeks, J. Gianetto, Strain-Based Design and Assessment in Critical Areas analysis of the mismatched welding joint with misalignments of D 1,422 mm
of Pipeline Systems with Realistic Anomalies, Tech. Rep. DTPH56-14-H-00003, X80 steel pipelines: An experimental and numerical investigation, J. Pipeline
CRES Americas, C-FER Technologies , NIST, and Natural Resources Canada,
Sci. Eng. 1 (2) (2021) 212–224, http://dx.doi.org/10.1016/j.jpse.2021.05.002,
2017.
Special Issue on Strain-Based Assessment and Design of Pipelines.
[74] W. Liu, Q. Guo, C. Qiao, W. Hou, Strain design method of buried pipeline
[92] D. Yi, Z.M. Xiao, S. Idapalapati, S.B. Kumar, Fracture analysis of girth welded
crossing fault, Eng. Fail. Anal. 105 (2019) 659–671, http://dx.doi.org/10.1016/
pipelines with 3D embedded cracks subjected to biaxial loading conditions, Eng.
j.engfailanal.2019.07.036.
[75] X. Liu, Q. Zheng, K. Wu, Y. Yang, Z. Zhao, H. Zhang, Development of a novel Fract. Mech. 96 (2012) 570–587, http://dx.doi.org/10.1016/j.engfracmech.
approach for strain demand prediction of pipes at fault crossings on the basis 2012.09.005.
of multi-layer neural network driven by strain data, Eng. Struct. 214 (2020) [93] Q. Zheng, Y. Li, N. Yoosef-Ghodsi, M. Fowler, M. Kainat, S. Adeeb, A finite
110685, http://dx.doi.org/10.1016/j.engstruct.2020.110685. difference-based approach for strain demand prediction of inelastic pipes
[76] H. Ma, B. He, X. Luo, W. Cai, D. Liu, C. Hou, J. Han, Investigation on strain subjected to permanent ground displacements, Eng. Struct. 273 (2022) 115072,
characteristic of buried natural gas pipeline under longitudinal landslide debris http://dx.doi.org/10.1016/j.engstruct.2022.115072.
flow, J. Nat. Gas Sci. Eng. 86 (2021) 103708, http://dx.doi.org/10.1016/j.jngse. [94] W. Zhou, Reliability of pressurised pipelines subjected to longitudinal ground
2020.103708. movement, Struct. Infrastruct. Eng. 8 (12) (2012) 1123–1135, http://dx.doi.
[77] V.E. Melissianos, D. Vamvatsikos, C.J. Gantes, Methodology for failure mode org/10.1080/15732479.2010.505244.
prediction of onshore buried steel pipelines subjected to reverse fault rupture, [95] H. Zhou, Y.-Y. Wang, M. Stephens, J. Bergman, S. Nanney, Tensile and com-
Soil Dyn. Earthq. Eng. 135 (2020) 106116, http://dx.doi.org/10.1016/j.soildyn. pressive strain capacity in the presence of corrosion anomalies, in: International
2020.106116. Pipeline Conference, Volume 2: Pipeline Safety Management Systems; Project
[78] D.B. Noronha, R.R. Martins, B.P. Jacob, E. de Souza, Procedures for the strain Management, Design, Construction, and Environmental Issues; Strain Based
based assessment of pipeline dents, Int. J. Press. Vessels Pip. 87 (5) (2010)
Design; Risk and Reliability; Northern Offshore and Production Pipelines, 2018,
254–265, http://dx.doi.org/10.1016/j.ijpvp.2010.03.001.
http://dx.doi.org/10.1115/IPC2018-78802, V002T06A007.
[79] N. Nourpanah, F. Taheri, Development of a reference strain approach for
[96] J.J. Chen, D. Williams, K. Leewis, D. Aguiar, Analysis of pressure test failure
assessment of fracture response of reeled pipelines, Eng. Fract. Mech. 77 (12)
performance for vintage pipe, in: International Pipeline Conference, Volume
(2010) 2337–2353, http://dx.doi.org/10.1016/j.engfracmech.2010.04.030.
1: Pipelines and Facilities Integrity, 2016, http://dx.doi.org/10.1115/IPC2016-
[80] C. Sakonder, L. Xue, M. Paredes, R. Savioli, D. Sarzosa, Directional dependence
of critical axial strain in x65 pipeline steel subject to combined internal 64518, V001T03A031.
pressure and bending loading, Int. J. Press. Vessels Pip. 196 (2022) 104610, [97] S. Yoon, D.H. Lee, H.-J. Jung, Seismic fragility analysis of a buried pipeline
http://dx.doi.org/10.1016/j.ijpvp.2022.104610. structure considering uncertainty of soil parameters, Int. J. Press. Vessels Pip.
[81] A. Sandvik, E. Østby, C. Thaulow, Probabilistic fracture assessment of surface 175 (2019) 103932, http://dx.doi.org/10.1016/j.ijpvp.2019.103932.
cracked pipes using strain-based approach, Eng. Fract. Mech. 73 (11) (2006) [98] Q. Zheng, W. Qiu, N. Ergezinger, Y. Li, N. Yoosef-Ghodsi, M. Fowler, S.
1491–1509, http://dx.doi.org/10.1016/j.engfracmech.2006.01.026. Adeeb, Development of an online calculation tool for safety evaluation of pipes
[82] S. Shen, B. Yan, T. Wang, G. Wu, S. Zhang, M. Zhu, Strain-based fracture subjected to ground movements, in: 14th International Pipeline Conference.
analysis for internal surface cracks of X80 pipe girth welds, Int. J. Press. Vessels Volume 1: Pipeline Safety Management Systems; Project Management, Design,
Pip. 203 (2023) 104944, http://dx.doi.org/10.1016/j.ijpvp.2023.104944. Construction, and Environmental Issues; Strain-Based Design and Assessment;
[83] Y. Shuai, X.-H. Wang, C. Feng, Y. Zhu, C.-L. Wang, T. Sun, J. Han, Y.F. Cheng, A Risk and Reliability; Emerging Fuels and Greenhouse Gas Emissions, 2022,
novel strain-based assessment method of compressive buckling of X80 corroded http://dx.doi.org/10.1115/IPC2022-86485.
pipelines subjected to bending moment load, Thin-Walled Struct. 167 (2021) [99] A.A. Soomro, A.A. Mokhtar, J.C. Kurnia, N. Lashari, H. Lu, C. Sambo, Integrity
108172, http://dx.doi.org/10.1016/j.tws.2021.108172.
assessment of corroded oil and gas pipelines using machine learning: A
[84] P. Vazouras, S.A. Karamanos, P. Dakoulas, Mechanical behavior of buried
systematic review, Eng. Fail. Anal. 131 (2022) 105810, http://dx.doi.org/10.
steel pipes crossing active strike-slip faults, Soil Dyn. Earthq. Eng. 41 (2012)
1016/j.engfailanal.2021.105810.
164–180, http://dx.doi.org/10.1016/j.soildyn.2012.05.012.
[100] M.R. Dann, L. Huyse, The effect of inspection sizing uncertainty on the
[85] Y. Wang, P. Zhang, G. Qin, Reliability evaluation of local corrosion of X80
maximum corrosion growth in pipelines, Struct. Saf. 70 (2018) 71–81, http:
pipeline subjected to accidental earthquake considering geotechnical discon-
tinuities, Int. J. Press. Vessels Pip. 189 (2021) 104254, http://dx.doi.org/10. //dx.doi.org/10.1016/j.strusafe.2017.10.005.
1016/j.ijpvp.2020.104254. [101] H. Wang, A. Yajima, R.Y. Liang, H. Castaneda, A Bayesian model framework
[86] J. Wang, Y. Shuai, R. He, X. Dou, P. Zhang, C. Feng, Ultimate strain capacity for calibrating ultrasonic in-line inspection data and estimating actual external
assessment of local buckling of pipelines with kinked dents subjected to bending corrosion depth in buried pipeline utilizing a clustering technique, Struct. Saf.
loads, Thin-Walled Struct. 169 (2021) 108369, http://dx.doi.org/10.1016/j.tws. 54 (2015) 19–31, http://dx.doi.org/10.1016/j.strusafe.2015.01.003.
2021.108369. [102] M. Pandey, D. Lu, Estimation of parameters of degradation growth rate
[87] Y. Wang, A. Xia, G. Qin, Probabilistic modeling for reliability analysis of distribution from noisy measurement data, Struct. Saf. 43 (2013) 60–69, http:
buried pipelines subjected to spatiotemporal earthquakes, Probab. Eng. Mech. //dx.doi.org/10.1016/j.strusafe.2013.02.002.
69 (2022) 103315, http://dx.doi.org/10.1016/j.probengmech.2022.103315. [103] A.A. Soomro, A.A. Mokhtar, J.C. Kurnia, N. Lashari, U. Sarwar, S.M. Jameel,
[88] Q. Wang, X. Li, Q. Wu, H. Jin, New methods for predicting strain demand of M. Inayat, T.L. Oladosu, A review on Bayesian modeling approach to quantify
arctic gas pipelines across permafrost under frost heave displacement, Geofluids failure risk assessment of oil and gas pipelines due to corrosion, Int. J.
2022 (2022) 9094890, http://dx.doi.org/10.1155/2022/9094890. Press. Vessels Pip. 200 (2022) 104841, http://dx.doi.org/10.1016/j.ijpvp.2022.
[89] Y. Wu, B. Meng, L. Wang, G. Qin, Seismic vulnerability analysis of buried 104841.
polyethylene pipeline based on finite element method, Int. J. Press. Vessels
Pip. 187 (2020) 104167, http://dx.doi.org/10.1016/j.ijpvp.2020.104167.

14