Managing pipeline risk online 

One example of this approach in action is DNV GL’s Pipeline Evaluation Portal, launched to boost the availability and ease-of-use of its proprietary models, combining advanced structural models and probabilistic assessment with subject matter expertise.The models cover a wide range of threats and hazards, and other operational aspects that must be managed to operate pipelines in a safe manner. 

”Many pipeline assets are ageing and analysis has to rely on historial data,” observed Jeffrey Lachey, group leader for pipeline risk, DNV GL - Oil & Gas.”There are few sensors on them and little data. This emphasizes the need for domain knowledge and data understanding  so we can make more sense of things.”  Re-using and combining results from several models allows quick, case-specific answers without the need for time-consuming customization, he added: ”Combining a model and large data sets lets us predict things that we could not before.” 

For example, DNV GL is developing a service for the Pipeline Evaluation Portal that uses existing data from pipeline operators to be better able to predict corrosion. ”We are doing this by combining online data with information on a pipeline’s surrounding environment, and using machine-learning techniques with large data sets to be able to predict when and how a failure may occur on a specific pipeline segment,” Lachey explained.  

Another application is already in beta version ready for testing through the portal. It demonstrates how a model for pipeline lateral buckling can be provided in a way that allows user interactivity to support decision making.  

“We are using an agile approach to innovation by exploiting digital platforms to work rapidly and collaboratively with industry partners, and get feedback as we develop these tools,” Helle said. “We are inviting the industry to test the lateral buckling application and report their experiences of it back to us. The rationale of the portal is to gather feedback that lets us decide whether to progress with development and if we need to make changes along the way.” 

More pipeline management applications are being developed 

Further applications being developed for the Pipeline Evaluation Portal will enable users to: 

• assess fatigue due to vortex-induced vibrations (VIV) and direct wave loading on free spans of pipeline 
• assess inline inspection data for pipeline corrosion analytics to deliver predictions of remaining operational life
• track potential trawl hazards to pipelines using data from ships’ automatic identification systems combined with vessel-specific information on trawl gear
• estimate potentially serious consequences of anchor load impacts on subsea pipelines. 

The portal also signposts access to a test version of DNV GL’s multi-analytic risk visualization (MARV™) tool. This can be used for pipeline risk assessment with uncertain and missing data by combining data and a pipeline operator's knowledge. It allows easy visualization of the cause-consequence relationships between factors that impact a threat's probability. 

Demonstrating the value of online collaboration 

The way the portal is being developed acknowledges that operators need to see value in gathering data sets that are shared through an independent trusted custodian, such as DNV GL. 

“Significant time and cost are involved in companies accessing and processing data to make it available in one place,” Lachey said. “Sometimes it is spread over different databases within the same company and may be in different formats. It helps if we can demonstrate that it will ultimately deliver benefits such as better corrosion prediction and valve-replacement programmes.”  

Companies once kept all data in-house to avoid surrendering competitive advantage, but some are now more willing to share data, Helle observed. “DNV GL encourages this as best practice where it makes sense; in analysing the root causes of pipeline failure for example. Success is a great motivator. A pilot phase may involve only one collaborator, but others become interested in the next stages of development if they can see potentially significant savings or other benefits from participating.”  

He expressed optimism over industry involvement: “We are already in discussion with a number of major pipeline operators and international oil companies for piloting the use of models. In our experience, some pipeline operators are keen to find ways of working differently, and are showing greater interest in collaboration.”

Disclaimer: 

DNV GL prides itself on providing accurate information but makes no claims or guarantees about the accuracy, completeness or adequacy of contents in this publication, and disclaims liability for any errors or omissions. The authors’ views here do not necessarily reflect DNV GL’s views.