The clock is ticking. The Pipeline Safety, Regulatory Certainty, and Job Creation Act of 2011 (H.R.2845) signed by President Obama in January further confirms the direction federal regulators are taking with respect to pipeline security and safety. The Act, along with advisories from the Pipeline Hazardous Materials and Safety Administration (PHMSA) largely address the post-failure findings of the San Bruno, CA pipeline failure in 2010.
The signed Act with a subsequent PHMSA Advisory reminds operators of covered pipelines to “ensure that the records accurately reflect the physical and operational characteristics of the pipelines…and confirm the established maximum allowable operating pressure…” PHMSA directives stress that operators “…assure that all MAOP and MOP are supported by records that are traceable, verifiable and complete”.
The Challenge For Pipeline Operators
How This Came About
On Sept. 9, 2010, a high-pressure natural gas pipeline ruptured in a residential neighborhood of San Bruno, CA. The subsequent ignition resulted in eight fatalities and numerous injuries, destroying 38 homes while damaging 70.
In part, the NTSB concluded that the operator’s integrity management program was deficient and ineffective because it:
1. Was based on incomplete and inaccurate pipeline information, and;
2. Did not consider the design and materials contribution to the risk of a pipeline failure.
(Source: NTSB Accident Report NTSB/PAR-11/01, PB2011-916501)
PHMSA made clear that audits starting in 2013 will focus on measures taken by pipeline operators to verify MAOP calculations. The clock is now ticking for operators to verify records by March 2013, or otherwise rely on another method to verify MAOP (as allowed for in 49 CFR 192.619 or 49 CFR 195.406). In late July, PHMSA confirmed that a “single, quality document” does indeed suffice provided that the document is traceable and complete. Notwithstanding PHMSA’s clarification, the subject matter of this article pertains to issues related to the use of multiple pipeline records for verifying MAOP.
A Phase-Out For Grandfathered Pipelines?
Post-San Bruno, the National Transportation Safety Board (NTSB) recommended PHMSA consider lifting the so-called “grandfather” clause [49 CFR 192.619 (a)(3)] and “require that all gas transmission pipelines constructed before 1970 be subjected to a hydrostatic pressure test…”. PHMSA will be addressing this recommendation in a future rulemaking. For some operators with significant portions of their systems constructed prior to 1970, and without a pressure test at some point during its history, lifting of the clause will mark a “step change” as it relates to their integrity management programs and ease of compliance.
A New Risk Outlook
The prospect of losing the grandfather exemption for older assets coupled with the costs of verification by pressure testing, inline inspection (ILI), or other means, can prove to be quite costly, time-consuming and operationally constraining. And pipeline systems targeted for acquisition will now potentially hold a much greater exposure to liability and potential non-compliance due to documentation (or lack thereof) of the targeted pipeline.
Lessons Learned In Records Verification – DNV’s 5-Step Process
DNV has been instrumental in the records verification process for several U.S. pipeline operators and we’ll share some of our experiences with you. A more detailed explanation of some of our findings can be found in the white paper, “A Practical Approach to Pipeline System Materials Verification” presented to the International Pipeline Conference (IPC) 2012 by DNV’s Andy Lutz and Dr. Tom Bubenik.
Though pertinent data capture alone can be a tedious task, we contend that the verification process is woefully incomplete until a digital, auditable records structure is devised. A structured framework is essential in supporting continual improvement while efficiently satisfying future audits. For sure, such a structure helps to establish the traceable component of PHMSA’s directives.
We think most will find the empirical approach of records verification will be preferred in verifying continued safe MAOP (vs. other testing methods to be proposed and/or developed). In some cases, a graduated approach will be necessary by using available documentation that is reliable but supplemented by other measures to close gaps. Though functionally unique, the five steps of our process will at times be performed concurrently. As the documentation review and capture process can be quite laborious and time consuming, we have concentrated our efforts towards maximizing efficiencies.
And finally, we’ve found that a “fresh set of eyes” by a trusted third party in the review process can better ensure that gaps and inconsistencies are more readily identified, with the added value of a pre-audit from the same critical perspective to be encountered in subsequent formal audits.
Step 1: Discriminate Pertinent From Non-Pertinent Records (Document Search, Collection and Tabulation)
The goal is to verify MAOP as early in the records review process as possible, realizing that the MAOP of the entire pipeline is based on its “weakest link.” DNV’s process will identify records gaps which may prevent ready verification. Pipeline attributes critical in establishing MAOP include pipe diameter, wall thickness, grade, seam type, component ratings and historic pressure test data.
The first step in the process will identify records that are pertinent to the pipeline’s material properties and testing used to establish MAOP. Records targeted will relate to design, fabrication (i.e. purchase orders, mill certification records), construction (i.e. as-built drawings, alignment maps), maintenance (i.e. replacements, repairs), and testing (i.e. pressure test records). Previous studies and analyses of class location and MAOP calculations performed during the line’s operational history will also be reviewed.
Both paper and digital records are reviewed by a team led by an engineer familiar with pipeline construction and maintenance documentation and who holds experience with data mining for the auditing process.
It is important that the client designate a company representative to act as ombudsman to answer the inevitable questions that will arise during the search. This client overseer will have sufficient knowledge of the pipeline system and the organization allowing him/her to readily respond to queries or to recognize the subject matter expert (SME) who is able to formulate a response. In addition to the use of SMEs, interviews with operations and maintenance employees (current and past) may aid in the data collection process.
During this phase, discrepancies, inconsistencies and omissions will be identified as gaps. Not surprisingly, we have found that older pipeline assets can generally be more prone to having gaps in design, construction and testing records. And changes made to a line during its operational life may not be properly documented, as well. Such problematic gaps can then be further compounded by multiple (non-continuous) owners and the lines’ susceptibility of incomplete records transfer during owner transition.
Step 2: Establish A Reliability Index For Captured Data (Document Precedence)
Even pertinent records will come in varying degrees of accuracy and trustworthiness. Step 2 in the process is therefore one of the most important: document precedence, or, establishing a reliability index. The process involves placing a ‘weighting’ on the various records in the context of accuracy, consistency, completeness and reliability. For example, pipe grade documentation from purchasing records will be given a higher precedence (weighting) than pipe grade tallies taken from ILI questionnaires – where the information source as well as the credibility or authority of the person completing the form may be uncertain.
To further ensure reliable indexing, key operator employees are consulted to identify those records historically given more emphasis due to their contents’ inherent reliability.
Creating a document reliability index, or document hierarchy, is essential for large-scale verification projects involving substantial volumes of data. Records gaps can’t be identified if data cannot be readily trusted; and, verification attempts in the absence of such an index can greatly add to inefficiencies and wasted effort.
Step 3: Make Permanent And Structured Records (Digital Capture)
As document imaging is beyond DNV’s normal remit, we’ve partnered with 3SG, a respected imaging company strategically located adjacent to our Dublin, OH headquarters. Proximity is key when considering the close interaction required for capturing critical, sometimes one-of-a-kind data.
As conclusions drawn from the records verification process must be traceable and subject to future audits, documents are captured digitally – typically in portable document format (PDF) – as they’re identified as pertinent and a structure is put in place for referencing. A systematic and consistent file naming procedure is implemented for the database filing system to be used in Step 5, Auditable Reference.
As the process can entail capturing huge volumes of data, we will usually perform the digital capture using mobile-scanning hardware and a dedicated team brought to the operator’s site for the task-specific job, thus avoiding overwhelming the clients’ scanning equipment. On larger, more massive projects, it can make logistical sense to perform imaging at 3SG’s Dublin facility, where specialized, high-capacity imaging equipment can be utilized.
Step 4: Identify/Reconcile Gaps (Verify or Re-establish MAOP)
The U.S. is home to some of the oldest pipeline systems in the world, as this is where the industry is rooted. Records for vintage pipeline systems will typically be more prone to deficiencies than with newer systems. As for assets with a history of multiple owners, the seller’s pipeline records can be incomplete or inaccurate and the legacy brain trust of the seller – the intimate asset knowledge held by those who may have laid and/or operated the asset for years – is not necessarily passed on.
Not infrequently, gaps will be a result of inadequate recording of material properties and testing for historical modifications, replacements or re-routes. The concern, for example, is that inadequately recorded pipe replacement opens the possibility of having a lower yield-strength pipe, perhaps due to a different wall thickness, grade or seam weld type, than the legacy pipe. Strategies must be formulated to make the verification process complete or to re-establish MAOP.
Various strategies open to consideration by regulators are offered in varying degrees of conservatism to address such gaps: from wholesale replacement (most conservative), to pressure testing, to engineering critical assessment/fitness for service assessment (less conservative). Such engineering assessments can be used to draw inferences from known results in order to answer the unknown.
Step 5: Build From A Traceable, Auditable And GIS-Compatible Framework (Risk-informed, Sustainable, Auditable Reference)
Verifying or re-establishing MAOP is, of course, the primary point of the verification process. A secondary goal is to establish an easily accessible and navigable data structure to facilitate (on-going) records management and for reference in audits. Certainly, a single system of record for all parameters affecting MAOP calculations and related data will aid in achieving that goal.
DNV has found that a linear format associated with pipeline footage or stationing is a practical form of spatial alignment, easing integration, navigation and user visualization of the entire pipeline as it traverses its route. With a linear spatial alignment, installations such as pipe type, valves and other applicable features can be made to appear in the data as they appear (spatially) on the pipeline.
It is important to couple the construction of this reference listing with sufficient revisions to company procedures to ensure that pipeline changes (i.e. re-routes, pipe replacements) are dutifully recorded in this central listing. Such measures help to ensure that the data remains traceable, verifiable and complete through the pipeline’s continuous operation.
Reconciling Records Gaps
Although PHMSA has yet to rule on remedies to reconcile records gaps, we would contend that there may be a role for supplemental methods such as using ILI and/or materials testing. Our ILI SMEs are exploring new applications of existing technology in efforts to reconcile some records gaps. For example, if replacement spools were used in a maintenance program with less than adequate records, ILI results can act as a supplement to as-built records by providing a “fingerprint” of the pipeline’s make-up. ILI can identify exactly where such spools were installed, as ILI will typically identify wall thickness changes and general seam type.
Inferences from the known can help reconcile the unknown.
If pipe grade is in question and the replacement spools can be grouped with a degree of certainty as a single pipe type (from the ILI results), then a single excavation with perhaps a single cut-out for metallurgical testing can offer a rational inference that all of the like spool pieces hold the same properties. Though such reconciliation may be far from ideal in terms of effort, it will almost certainly transcend alternative actions such as wholesale replacement or pressure testing.
The concept of extracting more information out of ILI devices and their data has recently been the focus of the Interstate Natural Gas Association of America (INGAA) and by Pipeline Research Council International (PRCI). Among other things, these industry groups are interested in the pipeline “fingerprint” concept for showing pipe changes. And research is being considered for possibly expanding ILI capability to better identify material properties from pigging data.
Few will argue that there won’t be continued, if not increased emphasis, on records to verify the calculated MAOP of U.S. pipelines. Where gaps heretofore were largely (and simply) defined by virtue of ‘known” vs. “unknown,” reliability of that information will now play a central role.
For grandfathered pipelines, it remains to be seen how future rulings will further affect the risk landscape.
Regulators have yet to rule on methods to be used for closing records gaps. It’s quite possible that at least some field investigations (i.e. in the ditch verification and testing) will be required to augment deficient information. In such cases, elements of this five-step verification process will help identify and prioritize locations to excavate by means of a risk-based approach.
At any rate, use of a rational, systematic approach to identify and mitigate records gaps will help operators produce verifiable materials results and should therefore aid in meeting regulatory requirements while improving pipeline safety. This approach will help to ensure that MAOP calculations are supported by traceable, verifiable and complete records, and pipeline safety will ultimately be improved.
Traceable – Records which can be clearly linked to original information about a pipeline segment or facility.
Verifiable – Records in which information is confirmed by other complementary, but separate, documentation.
Complete – Records in which the record is finalized as evidenced by a signature, date or other appropriate marking.