Two hundred sixty-nine regulated pipeline system accidents caused fatalities and/or injuries in the United States between 2010 and 2018, resulting in 106 fatalities and 599 injuries requiring hospitalization. About 84% of these serious accidents occurred on gas distribution systems, which primarily transport natural gas. This study adapts probabilistic risk assessment (PRA) methods which are used predominantly in the space and nuclear industries to gas distribution systems in the U.S. Nationwide system and accident data are used to evaluate natural gas distribution system risks, estimate how many additional resources the public would be willing to dedicate to reduce or eliminate these risks, and determine which improvement areas warrant further evaluation. Recommendations regarding the overall PRA-based framework, as well as the scope, quality, and level of detail of the underlying data, are provided.

While the uncertainties associated with actual pipeline asset condition demand the use of probabilistic methodologies to assess the integrity of pipelines, a realistic and validated probabilistic method to demonstrate post-hydrostatic test (PHT) integrity has eluded the pipeline industry. Traditionally, deterministic methods grow a “just-surviving flaw” (JSF) under worst-case pressure cycling to predict the remaining life of the most severe imperfection which could have survived a high-pressure event, such as hydrostatic test. The deterministic analysis results in a JSF fatigue life but does not identify the likelihood that the flaw exists. Furthermore, identifying the most severe flaw is not intuitive and attempts to probabilistically model material variabilities have failed to match known historical PHT reliability. A pipeline operator has now developed a novel approach to the task of quantifying marginal pipeline reliability after hydrostatic tests. Rather than limiting random values to only material properties, potential defects are assigned sizes and pressure cycling values, randomly sampled from validated distributions of defect size and pressure cycling severity (equivalent to downstream location). The number of generated defects is determined by a validated defect density, and defect size remains limited to what could have physically survived the hydrostatic test. The question posed is no longer “what are possible sizes of JSF close to discharge pressure surviving to a specific time under known load conditions?”, but rather “what proportion of the pipeline segments with similar defect populations would survive to a specific time under known load conditions?”. This represents a fundamental paradigm shift away from considering only a worst-case scenario to the quantification of plausible pipeline health conditions. Monte Carlo simulation time is kept practical by using an equivalent load integral method to project crack growth. This proposed methodology was validated by applying it to a selection of pipeline segments with known historical fatigue failures following hydrostatic tests in order to quantify the predictability of each section’s reliability at the failure time. The initial validation of the method was found to reasonably predict the past incidents. This paper will discuss the methodology, input parameters including their distributions, methods for assigning defect size distributions and densities based on extrapolations of field nondestructive examination (NDE) and in-line inspection (ILI) data, and a minimum defect density floor established based on the PHT fatigue failure of a newly constructed pipeline. While this method originally targets PHT pipeline segments, the development of a similar method for pipelines managed exclusively by ILI data is under development. The largest potential flaw for ILI-managed assets is then dictated by what could have evaded ILI tool detection rather than what could have survived a hydrostatic test. Herein, the progress on this development and future suggested research will be provided.

Digitalization in the oil and gas industry has led to the formation of digital twins. Digital twins bring closer the physical and virtual world as data is transmitted seamlessly between real time sensors, databases and models. The strength of the digital twin concept is the interconnectivity of data and models. Any model can use any combination of inputs ( e.g. operator owned data sets and sensors, third-party databases such as soil composition or weather data, results from other models such as flow assurance, threat modelling or risk modelling). Consequently, the result of one model may become the input of another. This strength is also a weakness, as uncertain (or missing data) will lead to a great source of uncertainty and may lead to wrong results. Worst case scenarios have been used to solve this issue without success. This paper presents a new concept: probabilistic digital twins for pipelines. Probabilistic digital twins do not lose uncertainty as results pass from one model to another, thus providing greater confidence in the final results. This publication reviews the probabilistic digital twin concept and demonstrates how it can be implemented using gas pipeline data from West Pipeline Company, CNPC.

The pipeline industry is increasingly utilizing reliability methods as a basis for decision-making when managing the integrity of pipeline systems. In this paper we present a methodology for developing reliability benchmarks for onshore low vapour pressure (LVP) liquids transmission pipelines to enable an operator to evaluate the performance of their pipelines relative to their industry peers. The goals of this methodology are to use an approach that is calibrated to historical performance, that remains interpretable, that can readily be adjusted over time, and that reflects the increased sensitivity of High Consequence Areas (HCAs). On this basis, the Top Quartile (Q 1 ) and Average historical environmental performance of the top 20 largest liquid pipelines operators were identified as two primary benchmarks of interest. Top Quartile (Q 1 ) environmental performance represents an industry leading benchmark performance while the Average historical environmental performance is representative of typical pipelines. This group of operators was selected on the basis that these major companies operate pipeline networks with comparable complexity, similar integrity challenges and a diverse range of operating environments (urban, rural, etc.). In addition, the length of their respective systems decreases the impact of small sample size issues. The dataset used for calculating the historical environmental performance of these operators was the 2010–2018 PHMSA Hazardous Liquids Incidents and Annual Mileage datasets. The framework presented in this paper, allows specific reliability benchmarks to be calculated for a given pipeline dependent on the environmental consequences associated with failure of the line. Consequences are considered in order to be risk-consistent between pipelines with different outflow potentials and different land use surroundings (HCA vs. Non-HCA). The heightened sensitivity of HCA areas is considered by including an impact ratio in calculating the reliability benchmarks. These benchmarks can be used to identify pipelines for additional scrutiny when the expected environmental performance does not meet benchmark levels and to flag areas for further evaluation. Quantifying the failure potential of pipeline segments gives operators insights into which maintenance actions will make the most effective use of the available resources to improve reliability. These benchmarks are calculated directly from historical data using a simple methodology and are easily recalibrated in the future as the performance of the pipeline industry continuously improves to meet increasing societal expectations.

Enbridge is North America’s premier energy infrastructure company delivering the energy people need and want. Enbridge’s business value is asset intensive. With over 200 onshore liquids pipelines facility assets, operational safety and environmental protection are always top priorities. The embedment of risk management practices in business decisions is an effective way to appropriately optimize asset performance while avoiding catastrophic impacts to people and the environment. This includes understanding and managing these risk events and establishing barriers to prevent the impact. Facility site containment is an independent protection layer that mitigates the consequences of a spill. The United States Environmental Protection Agency and the National Fire Code of Canada provide requirements to contain overland flow of a spill from liquids pipelines facility assets. Although there are specific volumetric requirements for spill containment for facility tanks, there are no specific volumetric requirements for spill containment for pipeline facility assets such as pumps, valves, etc. Industry typically employs an index-based approach to determine the specific design volumes using catastrophic rupture volumes and facility location. This approach has several shortcomings, including design inadequacy, inconsistency, and challenges with scalability. A risk-based approach is appropriate in determining the required site containment volume based on oil spill history, facility assets, and environmental sensitivities. A probabilistic model can be created using historical facility oil spill data based on the Pipeline and Hazardous Materials Safety Administration’s (PHMSA’s) facility incident database to estimate the likelihood of a given size of release occurring. If available, company oil spill history can also be used or integrated with the PHMSA dataset. Combining the likelihood of the size of release occurring with the estimated consequence (by accounting for the volume of a release and the environmental sensitivity at the release location), it is possible to evaluate the risk of a release. This estimation of risk can then be leveraged to support facility site containment design to manage the risk to an acceptable level. By informing facility site containment with volumetric requirements using reliability and consequence models and risk management principles, an organization can prudently balance pipeline safety and capital constraints to comply with federal regulations. This paper demonstrates this approach and describes: • The value of available data and model development • Reliability modeling and consequence assessment • Risk-informed decision-making • Future model enhancements

Effective asset integrity management is supported through the understanding of the condition of the asset, and the quantification of the safety and uncertainty of its properties. Risk based and risk informed decision making can help operators prioritize inspections and repairs on mainline pipe, as well as within operator facilities. Setting operator system specific targets for reliability and risk can help operators better understand the condition of their system, and provide one means of determining whether integrity action or other risk treatment is required on a specific asset, either on the mainline pipe system or on a facility asset system. While mainline pipe condition is better understood through the use of inline inspection technology and non-destructive examination in the field, facility piping and storage condition is more difficult to understand due to the complexity and number of segments of assets within an operator’s facility, as well as the unpiggable nature of the majority of facility piping. To help resolve this issue, a risk quantification can be done for each segmented asset within a facility. A relative ranking of asset risks can help prioritize facility integrity activities and drive the planning and execution optimization. However, simply looking at a relative ranking of asset risks may not be enough to maximize risk reduction and achieve safety and reliability targets. This paper looks to expand on the implementation of Risk Based Inspection (RBI) standard in API 581 and explore more broadly how facility asset risk results can be used in integrity planning and decision making. The paper also examines the application of using finance principals to better quantify risk and carry out a meaningful cost benefit analysis to optimize integrity programs. Interpreting a quantified risk dollar amount is an industry challenge, and shedding light onto the value of applying reliability and risk models beyond the safety of an operator’s system can be extremely beneficial for operators to enhance cost efficiency. The quantification of risk helps support the optimization of spend and resource allocation to bring efficiencies into integrity management systems while maintaining focus on the right risk mitigation across an operator’s system.

Small-diameter, low-pressure pipelines (operated at pressures below 40 percent of specified minimum yield strength [SMYS]) are unlikely to experience a rupture but may be vulnerable to leaks. In general, gas leaks result in less significant consequences to public safety than ruptures; however, under certain circumstances leaks may represent a significant threat. In particular, leaks resulting in belowground gas migration from low-pressure pipelines to a nearby confined space (e.g. the basement of a nearby structure to which the pipeline is not physically connected) may result in significant consequences as accumulation of gas to flammable or explosive levels may be possible. This paper explores the methods by which an Operator may address the threat of gas migration from leaks on low pressure pipelines in its risk assessment and leak management programs. Research was conducted into the specific conditions that contribute to belowground gas migration and included: a literature review of studies into gas migration phenomena, and a review of historical incidents within the United States on similar pipeline systems which resulted in significant property damage, injuries, or fatalities. The research included consideration of the effect of pipeline operating conditions, pipeline attributes (including diameter and depth of cover), leak duration, proximity of nearby structures, attributes of nearby structures (including presence of basements, slabs, and openings in the foundations), leak severity (e.g. pinhole, full separation), ground conditions (including soil permeability and presence of belowground features), and gas odorization. Key conditions impacting the potential for belowground gas migration are identified, and risk assessment methods and leak management practices are recommended based on the likelihood of migration. Based on the research conducted and the environmental conditions identified, a quantitative risk assessment approach, taking into account the likelihood of a leak on a low-pressure pipeline, the likelihood of belowground gas migration, and the potential consequences to public safety, is proposed.

The lack of established acceptance criteria has been one of the key challenges to the application of quantitative risk assessment (QRA) techniques in the Canadian pipeline industry. While a wide range of such criteria have been developed and published, it remains difficult for most operators to commit to specific criteria because such criteria may not be acceptable to all stakeholders. Recognizing this limitation, the Canadian Standards Association formed a Risk Management Task Force (RMTF) under the Technical Committee for the Z662 Standard on the Oil and Gas Pipeline Systems to propose criteria for potential inclusion in its non-mandatory Annex on Risk Assessment. This paper describes the criteria that have been developed by the RMTF and provides the background information needed for users to understand and use them correctly. The discussion includes: a summary of the measures used to quantify the safety risk associated with an ignited product release; a summary of established international and Canadian criteria that have been considered; a description and interpretation of the ALARP (As Low As Reasonably Practicable) principle; and the rationale used by the RMTF to select specific individual risk and societal risk criteria for CSA Z662. The proposed criteria are also compared to the criteria underpinning other risk-based parts of the Z662 Standard, including Annexes C and O. Guidance is provided on the analysis assumptions, methods and parameters required to ensure that the risk calculations produce results that are consistent with the definition and intent of the criteria. Key issues addressed by the guidance include the definition of individual risk (i.e. location risk versus personal risk), the pipeline length over which the frequency versus number of fatalities (F-N) relationship representing societal risk is calculated, and the effect of population density averaging over the pipeline length.

Gas transmission pipeline operators increasingly rely on Electro-Magnetic Acoustic Transducer (EMAT) technology to reliably detect, identify and size stress corrosion cracking (SCC) anomalies in their pipeline system. However, scheduling EMAT in-line inspection (ILI) on every pipeline in the system is not always practicable or achievable in an expeditious manner. A means of conducting a preliminary assessment of the SCC threat on pipelines without EMAT ILI data in an objective and quantifiable manner is useful for understanding the threat level and for prioritizing or deciding on outstanding EMAT inspections. A wealth of system-specific SCC field data from historical integrity excavations across the pipeline system typically exists in a pipeline operator’s dataset and can be readily leveraged for quantitatively estimating the SCC threat reliability in other, similar (“like-in-kind”) parts of the pipeline system. This system-specific data, based on actual SCC findings from integrity excavations, is an improved and more granular alternative to applying industry-wide SCC statistics to estimates of SCC reliability levels on pipelines without EMAT ILI data. This paper presents a robust and direct limit state approach for estimating the SCC reliability level in pipelines that have not yet had an EMAT ILI completed by leveraging system-wide SCC field findings from historical integrity excavations.

The thermal radiation from a jet fire is the dominant hazard resulting from accidental natural gas releases from onshore pipelines or facilities. To assess the consequences to both individuals and equipment, we require models to estimate the incident radiation from the jet fire to the surroundings. Simpler models with shorter implementation and run times are more viable for use in a full probabilistic risk assessment, in which the number of scenarios assessed could number in the millions. However, the level of accuracy within these models must be considered to ensure a reasonably conservative estimate is produced. A review and comparison of semi-empirical models from literature was performed and used to develop a decision tree to recommend the most computationally efficient jet fire modelling approaches based on the release scenario, while maintaining reasonable conservatism. Options for both vertical and non-vertical releases are presented, as well as corrections for lift-off, wind, and buoyancy. Additionally, an efficient algorithm from the area of computer graphics was adapted and applied to a weighted multiple point source jet fire model to account for the reduction in incident radiation to a receptor due to topography or structures partially obstructing the view of the jet fire.

For onshore pipelines transporting hydrocarbon products that remain in a liquid state after release, environmental damage and socioeconomic impact are the dominant concerns. It is reasonable to assume that the magnitude of impact from such a release event is, at least in part, a function of the volume of product released. Analytical models exist for estimating the release volume from pipelines as a function of the physical and operational parameters of the line, and the effective size of the opening at the point of line break. While the estimates of release volumes obtained from such models are an important component of risk assessments and serve to inform emergency response planning activities, it is also important to understand what historical data indicates in terms of actual release volumes and how they are affected by key pipeline attributes. This paper summarizes the findings of an analysis of product release events associated with the onshore United States (US) hazardous liquid pipeline network, as obtained from reportable incident data publicly available from the Pipeline and Hazardous Materials Safety Administration (PHMSA). The findings of the study suggest that, for major releases resulting from large pipeline openings, release volumes are correlated with line diameter; whereas, releases as the result of a leak are largely independent of line diameter. Based on this, simple models have been developed from which both the average release volume and the range of likely release volumes can be estimated as a function of the pipeline diameter and mode of failure. These simple release volume estimation models are useful for benchmarking exercises where the goal is to estimate expected or most likely release volumes for calibrating and/or validating more complex release volume estimation models, and for estimating worst-case release volumes for emergency response planning purposes.

Integrity management of a northern Canadian liquid pipeline required a unique approach due to the presence of methanol-induced, short, axially-oriented, internal stress corrosion cracking (iSCC) adjacent to girth welds. Previous studies focused on investigation of the susceptibility to iSCC, crack detection performance of inline inspection tools, and leak rates due to iSCC. The current feasibility study aimed to identify an approach to estimate the probability of leak due to iSCC that uses information from previous investigations and handles the data gaps. Two probabilistic approaches — a standard reliability-based approach and a Bayesian Network approach — were compared based on a number of criteria related to effective use of existing data, flexibility of the framework, and ability to iteratively improve the results. The results of the comparison indicate that a hybrid of the two approaches is an ideal way to develop a comprehensive integrity management tool.

Oil and gas pipelines are essential infrastructures that sustain the economy of modern society. They are designed for continuous and reliable operations over their service lives. Once installed, however, their reliability is affected by several threats among which external corrosion plays a significant role. Corrosion-based pit depth growth reduces the wall thickness over time that consequently affect the mechanical strength and the hydraulic performance of the pipeline. Pipeline utility managers rely on the corrosion growth rate models to plan their maintenance, rehabilitation and/or replacement. Existing pipeline deterioration models are mostly based on the power law function that relates the pit depth with the exposure time and rarely include the soil factors that can have effect on the corrosion growth rate. Moreover, the way these factors affect the corrosion rate is complex and cannot be captured with simple linear relationship. This paper uses data found in the literature to build a nonlinear pit depth growth model based on Bayesian spectral analysis regression technique. All continuous covariates are allowed to have smooth nonlinear spectral representations of their effect function on the pit depth growth. The discrete (i.e. categorical) factors are modeled using the ordinary least squared algorithm. The final semiparametric model allows to capture all pit depth measurements, even those difficult to be modeled using high degree polynomials. The stochastic nature of the pit depth growth is captured through the Bayesian approach. A time dependent reliability analysis using subset simulation is carried out to evaluate the changes occurring in the probability of failure of the pipe over time and allow for a better planning and management of these important infrastructure. The model is applied on a bare pipe directly exposed to the soil environment over time. The Bayesian pit depth growth model is accurate enough to allow the computation of the time dependent reliability of pipelines considering both the mechanical and hydraulic reliabilities.

Crack or crack-like anomaly is one of the major threats to the safety and structural integrity of oil and gas transmission pipelines. The crack threat is usually managed by hydrostatic test or regular in-line inspection (ILI). For a given crack ILI pipeline tally, operators need to identify critical anomalies, determine appropriate response time, and whether pressure restriction (derate) is required. Traditionally, a deterministic approach is used to determine the mitigation plan based on characteristic values of pipe properties, conservative crack sizing and crack growth rate, and further considering a minimum required safety factor. This study introduces a reliability-based approach to make the mitigation decision with full details. The annual probability of failure ( POF ) is evaluated for each reported crack anomaly by Monte Carlo simulation technique considering all the uncertainties associated with pipe geometry, material properties, crack size measurement, and assessment model error explicitly. Both environmental and cyclic fatigue load driven crack growth are considered following API RP 1176. A reliability-based mitigation threshold is proposed and calibrated against the deterministic minimum required safety factor for maintaining the system to a consistent safety level. Two case studies were conducted to demonstrate the advantages of the reliability-based assessment approach in this paper. Case Study #1 considers an NPS 20 transmission gas pipeline with more than 2000 ILI reported crack anomalies. Compared to the deterministic approach using conservative characteristic values, the reliability-based approach can reduce the number of required mitigation activities significantly without compromising safety. Case Study #2 assumes a severe crack defect on an NPS 34 liquid pipeline to demonstrate the impact of fatigue driven crack growth on POF . The POF of crack anomalies can be compared and combined with other threats, e.g. external corrosion, to evaluate the quantitative risk throughout pipeline systems. The proposed framework in this study could be used by operators to improve the crack assessment programs.

A reliability-based limit states design (LSD) method for assessment of corrosion and crack features has been developed for onshore transmission pipelines as part of a joint industry project. The rule-based LSD approach is a simplified form of the reliability-based approach that reduces the latter to a set of deterministic checks. The LSD corrosion assessment method and a comparison of its performance against one operator’s reliability approach were published in previous IPC papers [1,3]. This paper compares the LSD corrosion and crack assessment methods to another simplified reliability-based approach, namely the Pipeline Integrity Reliability Analysis (PIRA) Level I, that was published in a previous IPC paper [2]. The PIRA model is staged into three levels, where Level I analysis is a simplified form of the reliability-based approach where the probability of failures of predefined features’ sizes are precalculated using conservative assumptions for a fast turnaround screening analysis of the entire pipeline system in order to identify areas requiring more in-depth full probabilistic Level II or III analyses. This paper describes the application of both LSD and Level I methods on two low vapour pressure (LVP) liquid pipelines and provides a comparison of the results. The comparative analysis was based on the number of features not meeting reliability targets in both methods and, thus, are eligible for repair according to each method. Out of the two pipelines considered, one has corrosion and the other has cracks as the dominant threat. The results show that there are noticeable differences between the outputs of the two methods. The root cause of these differences was investigated by conducting a sensitivity analysis on the input parameters, including: reliability target, wall thickness distribution, feature depth and length distributions, pipeline sectioning procedure and the usage of model error (which is used in the LSD method but not in the PIRA Level I method). For cracks, differences between the Modified Ln-Secant model used in the LSD method and the CorLAS model used in PIRA Level I were also considered. It was observed that the discrepancies in the required repairs resulting from the two methods can be mostly attributed to the feature depth distribution, wall thickness distribution and the reliability target. For cracks, the burst pressure model selection also had a significant impact on the results.

The tectonic fault, which is one of the most common geohazards in field, poses great threat to buried pipe segments. Pipes will process to buckling or fracture due to large strain induced by continuously increasing ground displacements during earthquakes. Therefore, it is imperative to conduct safety analysis on pipes which are buried in seismic areas for the sake of ensuring normal operation. However, the highly nonlinearity of pipe response restricts the proceeding of reliability assessment. In this study, a hybrid procedure combining finite element method and artificial neural network is proposed for reliability-based assessment. First of all, the finite element model is developed on ABAQUS platform to simulate pipe response to strike-slip fault displacements. Thus, the strain demand value (the peak strain value obtained by finite element model in each design case) can be collected for database establishment, which is the preparation for neural network training. Thoroughness of the strain demand database can be achieved by a fully comprehensive calculation with consideration of influencing factors involving pipe diameter and wall thickness, operating pressure, magnitude of fault displacement, intersection angle between pipeline and fault plane, and characteristic value of backfill mechanics. Sequentially, Back Propagation Neural Network (BPNN) with double hidden layers is trained based on the developed database, and the surrogate strain demand prediction model can be obtained after accuracy verification. Hence, the strain-based limit state function can be respectively determined for tensile and compressive conditions. The strain capacity term is simply assumed based on published papers, the strain demand term is naturally superseded by the surrogate BPNN model, and Monte Carlo Simulation is employed to compute the probability of failure (POF). At last, the workability of the proposed approach is tested by a case study in which basic variables are referred to the Second West-to-East natural gas transmission pipeline project. It indicates that ANN is a good solver for reliability problems with implicit limit state functions especially for highly nonlinear problems. The proposed method is capable of computing POFs, which is an exploratory application for reliability research on pipes withstanding fault displacement loads.

Risk assessment is the process of risk analysis and evaluation. It is a required component of pipeline integrity management programs (IMP) and is generally the first step in most IMPs. For the risk assessment of natural gas pipelines, the primary concern is the safety of population near the pipeline right of way (ROW). TC Energy’s SWRA uses a quantitative risk assessment model that considers the effect of the thermal radiation due to ignited pipeline rupture and evaluate the consequence on the surrounding population. The overall risk is then evaluated using two specific risk criteria: societal risk and individual risk, with the societal risk measuring the overall level of risk to a community or a group of people and the individual risk measuring the level of risk to specific individuals who are present within the pipeline hazard zone. Natural gas pipeline systems often extend hundreds or even thousands of miles. As such, societal risk criteria for pipelines are typically defined based on a given length of pipeline segment, usually in 1 km or 1 mile (1.6 km). To assess the societal risk of actual pipelines, different approaches are taken on how the risk along the length of a pipeline should be aggregated and compared to the criteria. For example, the PD8010-3 standard in the UK recommends the societal risk of a pipeline through a community to be aggregated and then normalized to the unit length to be compared with criteria; whereas the Dutch regulation requires societal risk at the worst location to be used. In the current SWRA, the societal risk along the length of a pipeline going through development areas or communities is aggregated following the recommendation of the UK PD8010-3, where the risk is aggregated and normalized to the pipeline length. Due to the vast scale of the pipeline system, it is impractical to manually review all development along the pipelines for conducting societal risk assessment on a system wide basis. As such, extent of communities and development areas is determined by a computer program using a simple set of rules. It was found to have led to unsatisfying granularity in the societal risk assessment in certain situations, with some interaction lengths being too long and thus failing to identify the more critical section within the interaction length, and certain development lengths being too short and thus not very meaningful from a societal risk perspective. To overcome issues with the current societal risk assessment method in SWRA, an alternative method largely following the direction of the Dutch approach is introduced in this paper. In this alternative approach, the societal risk is evaluated continuously along a pipeline with a predefined a sliding length, and thus variations in the societal risk levels along the entire length of a pipeline, including the locations with the highest societal risk levels, can be identified. Implantation details and computational efficiency were discussed. The results from the alternative method were compared to that from the current method. The sensitivity of the sliding length method to the predefined sliding length was also investigated. The study showed that this alternative method improves the accuracy and granularity of the societal risk assessment in the SWRA, and, although it is relatively computational commanding, with an efficient implementation, is still practical even for very large gas transmission systems.

As one of the main risks of long-distance oil and gas pipelines, the consequences of pipeline accidents caused by third-party damage (TPD) are usually catastrophic. At present, TPD prevention approaches mainly include manual line patrol, fiber-optical vibration warning, and unmanned aerial vehicle (UAV) line patrol, but there are some limitations such as untimely warning, false alarm, and the missed report. As the location technology of mobile device matures, the user group provides massive data sources for the collection of location information, with which the tracks and features of the third-party activity along the pipeline can be directly obtained. Therefore, this paper proposes a method to identify the TPD behavior based on the location data of mobile devices. Firstly, the characteristics of relevant destruction behaviors were extracted from the historical destruction events. Then, the location information of the third-party activity near the target pipeline is obtained and the data is processed to remove the influence of noise, to reduce the computational burden of the subsequent identification process. Finally, calculate the difference degree of neighborhood trajectory and the similarity with the TPD features based on the data feature grouping (Difference feature and Similarity feature) to classify the type of third-party activity. Taking a 10km pipeline segment as an example, the method of this paper is used to preprocess the collected data and calculate the difference degree and similarity, 232 suspected TPD events are identified. After the on-site verification of the suspected damage by the line patrol, the results show that the method can better identify the third-party activities near the pipeline.

Pipeline operators sometimes face challenges in finding an optimum decision while executing both capital and operational projects. Utility- and risk-based decision-making approaches could create a framework for devising an optimum solution, especially in cases involving multi-parameter decision making. From a utility function perspective, an optimum solution would be the one with the least amount of cost and the highest benefit. In other words, an optimum solution could be achieved by maximizing a utility function. Accordingly, as an optimization problem, constraints related to safety, risk, and reliability for a given scenario could limit or dictate the space of feasible solutions. Addressing the problem of arriving at an optimum solution for a multi-parameter decision is the core topic of this paper. Moreover, optimized utility functions — decision making analyses based on probabilistic risk assessment — are discussed. In addition, the paper introduces a conceptual approach that has been utilized for infrastructure decision making: the life quality index (LQI) approach.

One of the leading threats to the integrity of oil and gas transmission pipeline systems is metal-loss corrosion. This threat is commonly managed by evaluating measurements obtained with in-line inspection tools, which locate and size individual metal-loss defects in order to plan maintenance and repair activities. Both deterministic and probabilistic methods are used in the pipeline industry to evaluate the severity of these defects. Probabilistic evaluations typically utilize structural reliability, which is an approach to designing and assessing structures that focuses on the calculation and prediction of the probability that a structure may fail. In the structural reliability approach, the probability of failure is obtained from a multidimensional integral. The solution to this integral is typically estimated numerically using Direct Monte Carlo (DMC) simulation as DMC is relatively simple and robust. The downside is that DMC requires a significant amount of computational effort to estimate small probabilities. The objective of this paper is to explore the use of a more efficient approach, called Subset Simulation (SS), to estimate the probability of burst failure for a pipeline metal-loss defect. We present comparisons between the probability of failure estimates generated for a sample defect by Direct Monte Carlo simulation and Subset Simulation for differing numbers of simulations. These cases illustrate the decreased computational effort required by Subset Simulation to produce stable probability of failure estimates, particularly for small probabilities. For defects with a burst probability in the range of 10 −4 to 10 −7 , SS is shown to reduce the computational effort (time or cost) by 10 to 1,000 times. By significantly reducing the computational effort required to obtain stable estimates of small failure probabilities, this methodology reduces one of the major barriers to the use of reliability methods for system-wide pipeline reliability assessment.