Risk analysis gets low marks from NAS
National peer review recommends Alaska take a top-down look at oil and gas failures costliest to state, get industry cooperation
An oil and gas infrastructure risk assessment methodology developed by a contractor for the State of Alaska has received failing marks in a review by the National Academy of Sciences.
After Prudhoe Bay transit line leaks revealed corrosion problems that shut down parts of the Alaska North Slope field for two months in 2006, the state began looking for ways to increase its monitoring of oil and gas facilities in the state. Frank Murkowski, governor at the time of the 2006 spills, envisioned an agency to monitor oil and gas field facilities comparable to the state-federal Joint Pipeline Office which monitors the trans-Alaska oil pipeline. His successor, Gov. Sarah Palin, established a smaller monitoring office in the Department of Natural Resources and also got $5 million funding from the Alaska Legislature for a risk assessment of the state’s oil and gas infrastructure.
In 2008 the state put out a request for proposals for a risk assessment study and awarded a $4.1 million two-year contract to a joint venture of Doyon Emerald Consulting Group LLC and American Bureau of Shipping Consulting.
Part of the risk assessment involved a third-party peer review of the methodology in the risk assessment, and the Alaska Department of Environmental Conservation, which is leading the risk assessment project, commissioned the National Academy of Sciences to conduct that review.
The report of that review, by the Transportation Research Board of The National Academies, was completed in September and was released to the public Oct. 20.
DEC said in an Oct. 20 press release that the risk assessment is one element in the state’s response to oil spills and revelations about the condition of certain North Slope pipelines.
DEC’s Ira Rosen, the state project manager for the risk assessment, said in the release that, “The state oversight team is in the process of weighing the Academy’s input, along with what we are hearing from the public, industry and government, to chart the best way forward.”
Academy: State not well servedThe board’s conclusion was that the state was not well served by the design of the risk assessment, based partly on the state’s request for proposals and partly on the risk assessment methodology developed by Doyon Emerald-ABS. The board reviewed available documentation and meetings in Washington, D.C., and Anchorage, and found the risk assessment methodology to be problematic in three main areas. The board said “the management plan was not feasible given real-world constraints, … the proposed risk methods were too detailed and lacked a sufficient top-down perspective necessary for capturing important risks, and … the proposed results were static and stopped well short of providing the State with a set of tools for evaluating risk mitigation opportunities.”
Focus on reliabilityThe board recommended that the state revise the scope of the project for future risk assessment efforts, “focusing first on risks to infrastructure reliability followed by studies and environmental and safety concerns.” It recommended a combination of top-down and bottom-up approaches, and characterized the initial risk assessment as bottom-up. The board said the state should work with industry from the beginning “so that common goals can be identified and mutual cooperation can be ensured,” and recommended focusing on an end goal of risk management including methods to increase system robustness.
In analyzing the methodology developed by Doyon Emerald-ABS the board said it “appears to be too data intensive given the available resources,” and “assumes significant industry cooperation that is neither promised nor likely to be forthcoming.”
But, the board said, even if industry were willing and able to provide the requested data, “it appears unlikely that the proposed methods are doable and would be useful in identifying and ranking the risk components of the physical and operational infrastructure system.”
The board said it had requested a “worked example” of one small portion of the proposed method but was told that was not available without “significant additional funding.”
And, even if the method proposed was successful in uncovering risk factors, the board said it did not believe that the output of the project in the proposed format would be useful to the state in risk management decision making.
The data issueThe board said that for the proposed assessment to be completed given the time and funding constraints, the contract team “was apparently expecting to draw upon significant industry cooperation and voluntary assistance.”
One document, the board said, pegged a gratis industry contribution at 50,000 hours for a six-month period, some 25 man-years of work or 50 industry personnel working just on providing data for this project for six months.
Industry assistance was needed for all phases of the project, including obtaining detailed facility descriptions and historical failure data.
The board said it was “unclear that the contract team had a clear idea as to what specific data were needed to complete the assessment,” and apparently planned “to cast a very broad net and collect as much information as possible from various industry sources and then to sort through it and select what was needed.”
The approach raised “legitimate concerns” with industry, which did not know how the data would be used and how access to it would be controlled.
System complexThe board characterized two types of events with the potential for harm: large events which cause the state financial harm, such as the 2006 spills which caused BP Exploration (Alaska), the operator at Prudhoe Bay, to do a major shutdown in order to search for additional corrosion that could be the cause of further leaks; and smaller events which would have the potential to harm the environment or people, but would not cause the state financial harm.
“The approach to assessing the risks of major reliability and financial events generally requires a different approach than would be used for the environmental and safety risks,” the board said.
Large-scale events “might include communications failures involving two or more operating organizations, human factors issues …, failures of management systems …, or events outside the oil and gas infrastructure that propagate into the infrastructure. As one might expect, a detailed, bottom-up analysis of all of the infrastructure elements is unlikely to identify such overarching factors that would lead to major events.”
The board recommended starting with “the identification of situations that could cause major financial impacts,” such as an extended shutdown of the trans-Alaska oil pipeline.
The second step would be to identify events which could cause such a shutdown. Once all possible situations that might lead to major reliability events are identified, “it would be possible to conduct specific, detailed analyses in order to develop mitigation strategies to eliminate or at least manage the risks.”