*As seen in Inspectioneering Journal’s September/October 2021 issue.
Earlier this year, Inspectioneering and Pinnacle co-hosted our 7th bi-annual ‘Meeting of the Minds’ (MOTM) with a select group of mechanical integrity (MI) experts from across industry. This meeting was once again conducted virtually, hopefully for the last time. Nonetheless, our panel of SMEs were excited to gather for another engaging discussion. As with previous meetings, participants come from various sectors of the industry, including refining, petrochemicals, offshore production, and chemical processing.
Previous MOTM Recaps have summarized some of the key takeaways from our discussions over topics like emerging technologies, corrosion under insulation (CUI) programs, integrity operating windows (IOWs), corrosion control documents (CCDs), risk-based inspection (RBI), and most recently, the effects of the Covid-19 pandemic on mechanical integrity programs. This meeting was focused on piping RBI and the challenges operators are facing properly implementing it. Industry surveys attribute some of the biggest and most prevalent MI failures to piping. What can we do to improve this performance? This brief recap of our MOTM discussion on RBI for piping might provide some helpful insight.
Piping Inspection Methodology
This meeting kicked off with a discussion over what inspection methodology(ies) the participants were using for their piping. While most everyone is still using traditional time-based and condition-based inspection strategies to manage piping integrity to a degree, the conversation quickly shifted to who has implemented or is trying to implement RBI for piping at their facilities.
One participant from a major refiner said they were using RBI on piping at 3-4 of their sites, yet his other sites had only partially implemented it. The goal was for all of them to be on an RBI program eventually though. He acknowledged that piping RBI is a little tougher than pressure vessels, but they were seeing benefits to it so they were committed to seeing it through. Some of the benefits include identifying specific damage mechanisms for each piping circuit and being able to really scrutinize the locations where they are looking for them. “We’ve had several finds that weren’t originally detected until we actually went out there to look for them properly,” he said. But of course, this can be done with an effective damage mechanisms review, RBI or not. RBI has the promise of prioritization based on probability of failure combined with criticality or consequence considerations. The structured process also forces us to think more critically and consider things we may have discounted in the past. RBI also forces us to think about confidence in finding the mechanisms of interest by using properly weighted strategies.
Another participant is in the process of overhauling all of their sites’ MI programs, including proper RBI implementation of piping. He admitted that they were just starting their RBI journey for piping and that they’ve traditionally been rule-based. Regardless, they were seeing good initial results. “We’re not necessarily using RBI to identify damage mechanisms, but rather, using it to do exhaustive damage mechanism reviews and optimize our MI program by optimizing condition monitoring locations (CMLs) at the same time.”
A participant with experience in both upstream and downstream O&G said for upstream/offshore, they were primarily rule-based; but for downstream, they were using RBI. They had found that for piping, a lot of the CMLs they’d been doing for a long time weren’t really providing the value that they once thought. However, the inspection plans generated through RBI were paying dividends. He said “some people may think RBI would lead to less time in the field, but for us, it led to more time in the field; which isn’t necessarily a bad thing.” For them, “RBI has resulted in some significant finds because we are now looking in the right way, at the right spot, with the right amount of coverage.”
One of the participants not using RBI for piping said they did an extensive RBI implementation in the late 90s/early 2000’s and it included piping. However, they dropped RBI pretty early on for all of their piping across their plants because they didn’t believe that RBI could correctly model piping like they really needed it to. They had it modeled as one component per circuit, and found it wasn’t really representative of what was going on in the circuit. Now, they invest a lot of time and money on piping inspection and following up with RT and other techniques. They have also developed a statistical model to analyze data outside of traditional thickness monitoring and use it to optimize their system.
Another participant manages six sites with very different backgrounds. Piping strategies at each site are somewhat different, but he hopes over time they will converge to something similar. But in terms of RBI, pretty much all of their sites use some variation of RBI for damage mechanism reviews (DMRs), identifying morphology, expected damage rates, etc. Then they compare that to historical circuit data. But he said most of their inspection scheduling is done per API 570 in 5/10 year default intervals.
Challenges to Making RBI for Piping Work
Everyone appeared to be in agreement that implementing an RBI program for piping is more challenging than it is for pressure vessels. But what makes it so challenging? Here are some of the biggest challenges the participants are facing when it comes to successfully implementing RBI for piping.
One person, who acknowledged they were still in the implementation phase, said the challenge is really two-fold:
- Many of their plants just don’t have their piping properly organized, meaning it hasn’t been properly systemized and circuitized yet.
- The other big challenge is more culture and training — there’s still a lot of misperception that RBI is just a tool to extend intervals. When RBI comes back with a bunch of inspections on the back end, people tend to start questioning its validity, and quite frankly get defensive about the perceived lack of faith in prior inspections.
One of the participants said the biggest limitation for them was that when they implemented RBI for piping, they treated it like a pressure vessel and had every circuit represented by one component (e.g., one corrosion rate, one thickness, etc.). After doing so, they found that model to be too simple for a complex piping system. However, he did admit that nowadays might be different with all the data and computer processing we have available. Another participant was in full agreement, stating that “traditionally, a lot of these RBI programs were standalone, so you had to feed them a corrosion rate and other information from your IDMS. Whereas now, they’re all embedded and we’ve got all of this information at our disposal. There is now a lot more opportunity to get clever about how the logic works in these programs and actually make some logic-based decisions, versus nominating some component that might not actually represent the circuit.”
Another big issue brought up is that with piping, you have a lot of internal and external damage mechanisms to consider. You can apply a consequence to it, but making sure you catch all those mechanisms and get them scheduled appropriately is challenging. One participant said they’ve “historically said let’s set up a special emphasis program, whether it’s dead legs, or touch points, etc., but at the end of the day, you want a holistic program that catches all of those and considers the consequences that go with it.” He said putting that holistic approach together has been their biggest challenge. Piping integrity is definitely one of the most complicated challenges the industry faces, especially in the more complex operating systems.
Value of Consequence Evaluation
One person commented that he believed those that are really using RBI effectively on piping have a significant advantage (“real value”) because of the evaluation of consequence, which doesn’t come very well with time-based inspection and condition-based inspection. He claimed that sites that have effectively implemented RBI can substantially lengthen the majority of their piping intervals, leaving them able to focus much more on the remainder.
Another participant agreed on the importance of evaluating potential consequences of failure (CoF), sharing that he addressed this very issue earlier that week with his team. “I encouraged more of my people to pay closer attention to the CoF side, because that’s where they can really gain some dividends and give some of the low consequence stuff time off for good behavior,” he said.
A participant with a slightly different opinion stated that “a big problem with CoF for piping is that piping systems can physically be very long, and while one section may be near a river, the rest may be in a tank farm. So now it would need to be broken up into different circuits.” They have tried to simplify the consequence side just based on a default hole size, so now they just look at internal pressure and the fluid properties to determine the CoF. Regardless, he admitted it can become extremely challenging and that simplification can over-emphasize consequence for most cases. “Trying to come up with accurate CoF for piping is complicated and extremely difficult,” he said.
In closing, most failures in the O&G and chemical processing industries can be attributed to piping. It’s a complex asset that requires a robust inspection and reliability strategy to manage properly. Effectively implementing RBI for piping continues to be a challenge for operating facilities around the world, but if done properly, it can pay dividends in the form of more effective and efficient inspections, and more reliable operations.
Inspectioneering and Pinnacle would like to thank all of the participants for joining in this discussion of these critical issues. We sincerely appreciate you dedicating your time to share your thoughts and experiences with our community.