PPTS guides drop in liquids release frequency, scale

May 1, 2010
Both the frequency and scale of releases from hazardous liquids pipelines in the US have declined since the 1999 introduction of the Pipeline Performance Tracking System.

Both the frequency and scale of releases from hazardous liquids pipelines in the US have declined since the 1999 introduction of the Pipeline Performance Tracking System.

A data mining team (DMT) uses PPTS data to develop operator advisories, focusing industry's efforts and priorities on areas in which it can improve safety and reduce risk. The DMT evaluates data from both a consequence and frequency perspective to assess risk. Based on these risk analyses it shares the data and findings with industry to facilitate ongoing dialogue on additional areas of focus for improvement and refinement.

The DMT strives to improve industry performance by reviewing industry trends to create cause-specific operator advisories. These advisories discuss industry lessons and make recommendations for operators. These lessons and recommendations may allow operators to see potential leading indicators they may not have experienced yet or allow them to focus on ones they have.

In addition to industry advisories, each participating operator receives a confidential annual report card allowing it to compare its performance against its own previous performance, that of its peers, and the industry as a whole.

Background

PPTS has become a major tool of a joint environmental and safety initiative of the US American Petroleum Institute and the Association of Oil Pipe Lines, applying a multidisciplinary approach to understanding and improving industry performance by collecting information about releases, analyzing it, and finding ways to reduce or prevent releases.

When it was developed, the opportunity to identify leading or lagging indicators and learn from them was limited to the incidents reportable to the US Department of Transportation's Office of Pipeline Safety (now part of the Pipeline and Hazardous Materials Safety Administration, or PHMSA). With a volume threshold for hazardous liquids releases of 50 bbl, the accident data reported represented just the tip of the pyramid of releases. PPTS has allowed the industry to sharpen its focus on safety, develop new ways to keep its pipelines safe and secure, and strengthen its commitment to protect the environment and strive for zero incidents.

PPTS participants operate 85% of the liquids pipeline mileage subject to PHMSA jurisdiction. PPTS participants voluntarily report spills of 5 gal or more, plus any smaller spills to water. The lower PPTS reporting threshold resulted in many more spills reported and provided more opportunities to learn from the data. In 2002, PHMSA decided to lower the threshold for its accident reports to match the PPTS threshold.

While specific incident information recorded in PPTS remains confidential, aggregate data from all of the reported spills are compiled by the DMT, which publishes results of the analyses in advisories the pipeline industry can use to better manage pipeline risk.

The DMT as of early 2010 includes representatives from Shell Pipeline Co. LP, Sunoco Logistics Partners LP, Kinder Morgan Energy Partners, LP, Colonial Pipeline Co., BP Pipelines (North America) Inc., ExxonMobil Pipeline Co., Marathon Pipe Line LLC, Chevron Pipe Line Co., ConocoPhillips Pipe Line Co., American Petroleum Institute, Association of Oil Pipe Lines, and Allegro Energy Consulting.

Data use

The framers of PPTS created a survey asking enough detailed questions for the industry to understand the causes of and circumstances surrounding accidents and releases. The original PPTS structure prescribed a simple report for small, low-consequence releases (≥5 gal, <5 bbl or 210 gal) but a more complex and detailed report for larger, higher-consequence releases (>5 bbl). In addition to the volume threshold, incidents involving deaths, injuries, fire, or explosion also used the more detailed form.

As the typically larger releases along the right-of-way became better understood and more attention was focused on releases (typically smaller) from station piping (piping and equipment inside pump or meter stations or tank farms), it was determined the smaller amount of data collected by the simple short form was inadequate to identify the factors driving these incidents, leading to the short form's discontinuance in 2007.

The industry also expanded PPTS in 2007 to collect more information about excavation-related accidents by adding a module matching the Common Ground Alliance's (CGA) Damage Information Reporting Tool (or DIRT). PPTS DIRT data is uploaded to CGA annually.

For incidents occurring in 2009 and later, PPTS revised its survey form to provide greater granularity on release causes and contributing factors.

Each reporting year, PPTS participants also complete an infrastructure survey, a supplemental survey of information needed to understand the industry's footprint, and provide data to normalize accident information (e.g., system miles, number of tanks, pump and meter stations, and volumes moved). The infrastructure survey was also revised for 2009 reporting to capture information on broader categories of pipelines, equipment types, and transported commodities (e.g., biofuels).

10-year findings

The 10 years spent analyzing PPTS data have resulted in refinement of what information is collected and how the industry uses it. The industry uses the data to identify areas of focus for additional investigation or improvement. Since the liquids pipeline industry considers protection of the general public as the number one reason for maintaining the integrity of its pipeline systems, consequences such as a fatality, injury, fire, explosion, or a public evacuation also remain an area of focus. PPTS uses this lens to assess data and industry practices to determine the most frequently occurring events, those with the highest consequences, and those occurring over the rest of the frequency-consequence risk continuum.

Table 1 shows releases from station piping account for more than 60% of the incidents but only 10% of the volume spilled in 2006-08. Onshore pipe, in contrast, accounts for less than 30% of all spills and 70% of the volume spilled. Station piping accounts for most of the spills <5 bbl, and onshore pipe accounts for most spills ≥50 bbl or more. These contrasts underscore the correctness of giving a higher priority to addressing releases on the pipeline ROW. Facilities shouldn't be ignored, but more emphasis should be placed on addressing risks on the ROW first.

Fatalities and injuries resulting from liquids pipeline releases are uncommon, but PPTS data helped show the industry casualties were disproportionately occurring along the ROW (Table 2). About 80% of both incidents involving a fatality and incidents involving an injury occurred on the ROW. Another factor giving the ROW high priority in prevention efforts is the effect a pipeline release involving a fire or explosion, or requiring an evacuation has on the surrounding community and its resources, such as emergency personnel.

Right-of-way

The industry's record shows dramatic improvement along the ROW (Fig. 1). Releases have fallen from 2.0/1,000 miles in the earliest 3-year period (1999-2001) to 0.7/1,000 miles in the latest 3-year period (2006-08), a decline of 63%. The volume released has fallen to about 330 bbl/1,000 miles in the most recent period from just over 600 bbl in the earliest period, a decline of 48%.

Fig. 2 shows the industry has reduced accidents along the ROW across the diverse causes. Corrosion-related releases, the most frequent, fell by 74% between the most recent 3-year average, 2006-08, and the earliest 3-year average, 1999-2001. Releases involving excavation damage, the second most frequent cause, fell by 63%. Both causes have been the subject of PPTS Operator Advisories, discussed more fully below.

Corrosion

Early analysis of PPTS onshore pipe incidents showed corrosion accounted for the largest share of all onshore pipeline incidents. The vast majority of corrosion-related incidents occurred on crude oil systems. While corrosion releases tend to be small and seldom result in casualties (no fatalities, one injury over the 1999-2008 period), their frequency brought corrosion issues to the forefront.

The Data Mining Team prepared a series of advisories to the industry focused on corrosion incidents, their causes, and assessment options. These advisories provided operators with factors and strategies to consider in their risk assessments and ongoing integrity-management programs.

The industry's own initiatives and the integrity-management programs mandated by PHMSA have caused the number of spills caused by corrosion to drop by 74%, the largest decline among the major cause categories. The industry invested billions of dollars in integrity-management programs since 2000 to improve pipeline safety and comply with PHMSA's risk-based approach.

Inspection and repair have reduced corrosion-related spills, identifying vulnerable spots and allowing repair before the pipe failed. Corrosion, however, remains the leading cause of onshore pipeline spills and is the largest cause of spills along the ROW at 45%. Most of these releases, however, are low consequence, in contrast to excavation damage incidents.

Third-party excavation

Excavation or digging around or above a pipeline can seriously damage the pipeline and injure or kill individuals living and working on the ROW. PPTS considers a "third party" to be a person or persons not involved with operating or maintaining the pipeline. Third-party damage incidents account for <7% of overall incidents but tend to be larger releases (60% are 50 bbl or more on the ROW, and third-party damage accounts for 36% of these).

About 90% of third-party incidents occur on the ROW, causing 50% of the incidents involving a fatality (and the most deaths), 38% of the incidents involving an injury (and the greatest number of injuries), and 38% of the incidents involving an explosion. The majority of individuals killed or hurt in these incidents were members of the public not involved with the pipeline.

Analysis suggested potential for improvements to internal, peer, and external damage prevention programs. Analysis of who the damaging party was provided insight into outreach and prevention strategies. It also demonstrated prevention of these incidents as a high priority from both safety and integrity perspectives. Between the first 3 years of PPTS data, 1999-2001, and the most recent 3 years, 2006-08, releases caused by third-party excavation fell 63%.

The high consequences associated with excavation damage prompted the DMT to write a variety of advisories on the subject. Most recently, it published advisories addressing the different types of parties responsible for excavation.

One set of excavators includes operators of underground facilities such as other pipelines in a shared ROW, sewers, telecommunications, and electric utilities. These entities were of particular interest because they fund the "one-call systems," through which an individual or company planning to dig, excavate, trench or drill can register their intent with "one-call" and each underground facility operator with an asset in the area will mark its location.

One might have assumed these "one-call partners" would be most attuned to the potential dangers of digging and the need for using safe practices. Data showed, however, almost 90% of incidents involving these stakeholders were caused by a failure to adhere to one of the Common Ground Alliance's best practices. These hits were also taking place deeper than other excavation incidents, largely as a result of directional drilling (see box on this page).

PPTS data also showed operators (first parties) and their contractors (second parties) as occasionally damaging the operator's own pipeline or other assets while working or excavating around it. Releases caused by the PPTS operator or its contractor are classified as operator error in PPTS.

Station piping

While the industry began to see declines in pipeline releases, the number of releases from station piping remained relatively stable and became an area of greater focus. Facility releases account for most incidents but tend to be small, have limited public safety effects, and are contained on site.

Understanding facilities releases requires finer cuts of the data. Releases from valves, for instance, do not have the same causes or involve the same circumstances or operation-maintenance tasks as releases from piping itself.

One of the first advisories on the subject of releases in facilities therefore arrayed the data by the type of equipment. Pumps stood out as being involved in a disproportionate share of incidents, especially events involving a fire. These releases generally involve seals or pump packing. While the pump-related incidents are few, and often small, they require a shutdown of the pump, constraining the movement of products through the pipeline.

DMT is examining the issue of pump failures further but in the meantime has decided to also pursue more straight-forward issues. Pipe, for example, was the piece of equipment most frequently involved in facilities incidents, and the most frequent cause of failure was internal corrosion in crude oil systems. Industry experience suggested these may involve dead legs or other low-flow lines, but data to quantify the issue or clearly define the mechanisms at work in these incidents were missing.

An industry team sought to understand the subset of facilities releases occurring on tubing and small piping, dead legs, drain lines, and relief lines. The team created a series of special surveys to obtain additional detail on failures involving these different types of piping. Of 180 incidents reported on these special surveys, about half involved tubing and small piping. More importantly, they often involved fittings, rather than the tubing or pipe body itself and were frequently tied to poor installation and vibration, suggesting the issues are not asset-related, but operations-related.

Corrosion also remained an issue, especially where flow is intermittent and water might collect. Microbial-induced corrosion (MIC) also bore examination. Because expansion of water in freezing conditions can also cause issues, possible water traps need to be removed in existing systems or avoided in the design of new systems. The team examined the role of valve vault design, but the special survey participants had encountered too few incidents to suggest conclusions.

The team also surveyed facilities' inspection practices–frequency, scope, methodology–and found a wide variance. The team's subject matter experts assembled possible take-aways for operators to consider in designing and implementing inspection protocols. Station piping continues to be an area of focus.

Operator error

Operator-error incidents are important both because they are in the operator's direct control and because they have caused deaths and injuries to pipeline workers (employees and contractors). Operator error accounted for about 14% of incidents overall, but 20% of the incidents involving a fatality and 24% of the incidents involving an injury.

An operator-error incident involves tasks completed by an individual who can be either an operator's employee or its contractor. About 70% of operator-error incidents involve station piping. While particularly infrequent along the ROW, accounting for only about 6% of all ROW incidents, operator-error incidents occurring there involved excavation damage by an operator or its contractor in about 40% of the incidents; these incidents also caused an injury.

PHMSA is also stepping up reviews of operator-error incidents and their causes, whether occurring inside facilities or along the ROW. Analysis recommended additional focus on operator training programs, operating procedures, management of change, analysis of abnormal operating events, and root-cause analysis. Modifications to PPTS occurred to improve the reporting path and options for these incidents.

The importance of operator error for pipeline operators has led to a series of PPTS Operator Advisories on the subject, including one covering incidents occurring because operators damage their own pipeline during excavation. The DMT also surveyed operators on their use of root-cause analysis in accident investigations. A key question centered on whether increased use of root-cause analysis in recent years had led to reclassification of some incidents to operator error, and for 60% of the respondents, it had. This conclusion encouraged the DMT to provide clearer reporting paths for incidents involving operator error.

Editor's note: If you would like to learn more about PPTS or read its advisories, go to
http://www.api.org/ppts.

The authors

Adrianne Cook ([email protected]) is an industry analyst at the Association of Oil Pipe Lines in Washington, DC. She has also served as a regulatory and telecommunications attorney with RJGLaw LLC, in Silver Spring, Md., and as a regulatory analyst at the Edison Electric Institute in Washington, DC. She holds a JD from The Catholic University, Columbus School of Law, and a BS in business management from Clemson University. She is currently pursuing an MBA at George Mason University. She is a member of the Energy Bar Association and the Junior League of Northern Virginia.
Peter Lidiak ([email protected]) is director, pipeline, at the American Petroleum Institute. Lidiak has been with API since April 2000 working on air quality, fuels and refining, and pipeline issues. He directs API's pipeline segment, overseeing federal and state activities related to hazardous liquid pipeline safety, security, operations, and environmental performance. Prior to joining API, Lidiak worked in the mobile source program in the Office of Air and Radiation at the US Environmental Protection Agency for 15 years. He is a graduate of Colgate University.

Tom J. Kelly is environmental compliance manager for Colonial Pipeline Co., Alpharetta, Ga., which he joined in 1989. Previous positions have included environmental team leader, regulatory engineer, quality-assurance coordinator, southeast district project manager, and Atlanta area engineer. Kelly holds a BS (1988) in chemical engineering from the University of Tennessee, Knoxville, and is a member of AIChE and NAEM.

Cheryl Trench ([email protected]) is president of Allegro Energy Consulting, New York City, a firm she established in 1997. She has also served in a variety of positions during 22 years at Petroleum Industry Research Foundation. She holds a BA (1978) from the City University of New York. She is a member of the International Association for Energy Economics and its US affiliate. She serves on the board of The Energy Forum Inc., a New York-based not-for-profit organization that sponsors seminars on topics of interest to energy professionals.

One operator's drilling experience

Our experience suggests drilling risks could be quite high when compared with trenching, and we've instituted practices such as requiring that drilling contractors "bell hole" over our pipe to show they crossed it safely. We have also had repeated experience with drilling operations not using One-Call, or giving wrong or incomplete information on the location. As an example, a cable company's contractor gave the location where they were going to start, which led us to believe we were not involved. A subsequent in-line inspection revealed significant pipe damage from that same job after the contractor moved down the street for a new bore. He thought he hit a rock, withdrew the tool, and made another pass, which further gouged the pipe as it slid underneath. Pay attention when the excavator is going to be using directional drilling.

More Oil & Gas Journal Current Issue Articles
More Oil & Gas Journal Archives Issue Articles
View Oil and Gas Articles on PennEnergy.com