Part 2: Shaping The Future—Data-Driven Approach To Risk Assessment

Part 1 of this article, published on Oct. 27, can be found here.

Multi-Analytic Risk Visualization (MARV) has been successfully piloted for different pipelines in various locations worldwide. The following four project examples describe the application of MARV™.

PROJECT 1: Internal Corrosion Risk Assessment Of A 36-Inch Pipeline

MARV was employed for an internal corrosion risk assessment of a 36-inch pipeline carrying crude oil. The MARV calculations identified pipeline sections with lower and higher risks. It was deemed that the uncertainty of the input data for the risk assessment may grow in time, and the operator may need to add more information.

The predictions and in-line inspection (ILI) intervals per the initial findings are shown in Figure 2. The graph shows that uncertainties of internal corrosion risk assessment can be reduced and the outcome of MARV™ is useful to optimise the time inspection interval more efficiently. This study also identified the parameters that need to be monitored closely by the operator to reduce the risks even further.

PROJECT 2: External Corrosion Assessment Of A Crude Oil Pipeline

A company required to carry out an external corrosion assessment of a crude oil pipeline where the input data for the conventional risk assessments was incomplete. The assessment was carried out using MARV and subsequently the outcome of the assessment was compared with the ILI data, shown in Figure 3. The most crucial operational risk parameters were identified which benefited from the cause-and-effect relationships, of different parameters, in the form of conditional probabilities implemented in MARV. This project demonstrated that even with uncertainties caused by incomplete data sets, the probability of corrosion damage estimated by MARV was in tolerable agreement with inline inspection data and it could help the operator to make an informed decision before ordering another ILI run.

PROJECT 3: Internal Corrosion And Erosion Risk Assessment Of A Pipeline

In 2010, DNV GL was asked to predict the flaw size in a 36-inch x 52-inch pipeline using only 2004 input data. The pipeline was suffering from internal corrosion and erosion due to a combination of corrosive gases, sand and microbiologically influenced corrosion (MIC). MARV predicted the flaw size distribution along the whole pipe for 2010 compared well with the ILI data (see Figure 4).

It should be noted that DNV GL was not given the 2010 ILI data prior to modeling and therefore this constituted a blind test of the MARV method capabilities.

PROJECT 4: Third-Party Interference Risk Assessment

Third-party interference has been one of the largest threats to pipelines in some locations and has been blamed for lost production. Illegal tapping of pipelines is a global problem that costs operators millions of dollars per year and poses a high risk to safety, the environment and company reputation. Illegal tapping is a complex problem which includes demographically driven factors that needs to be considered in risk assessment of any pipelines.

DNV GL worked with an operator to identify and link the main parameters to different locations of a pipeline. DNV GL experts worked with the operator and established a cause and effect relationship between different parameters. The final deliverable of the assessment was to predict the probability of the next third party damage. This prediction helped the operator to plan necessary mitigation actions and formulate a repair procedure if required. Figure 5 is an example of illegal tapping problem on a major pipeline.

Today’s risk assessment codes are based on number analytical/numerical solutions, engineering judgment and operator specific experiences, where information and knowledge are the foundation of developing new technologies such as MARV. Building on the oil and gas industry’s vast experience, the tool can efficiently carry out the risk assessment, creating a seamless interface of all the input data. MARV has a modular nature and it can be tailored to specific operator’s needs. DNV GL is continuing to enhance the MARV™ capability by working with the industry, benefiting from their experience and data and creating new modules to address industry needs.

There is no limitation to using the tool in terms of risk threats, material, type and size of the pipelines. The concept can also be applied for offshore application. It is designed to improve decision-making process, minimizing the short-term risks, and maximizing long-term maintenance strategy.

MARV has been successfully piloted for 35 pipelines in various locations worldwide. DNV GL has recently collaborated with West Pipeline Co. and CNPC on adopting MARV for their assets. The technology is also supported by a California Energy Commission grant in partnership with California pipeline utilities.

Yonghe Yang from West Pipeline Co. says: “We appreciate the way MARV software displays risk by linking physical threat models with geographical information. All information is available in one interface and future projections do not hide uncertainties. Our decision-making process has been considerably improved.”

Bingchuan Yan from the Pipeline Company of CNPC says: “MARV has helped us to substantially improve our risk management, especially when dealing with a number of uncertainties and complexity in the input data. MARV™ visualization interference makes it easy for our engineers and managers to track the assessment at various levels.”

Ali Mirzaee-Sisan is business development manager – pipelines; Francois Ayello is ‎principal engineer, strategic research and innovation; Shan Guan is principal consultant, strategic research and innovation; and Narasi Sridhar is program director, materials technology and development, all of DNV GL – Oil & Gas.