Risk assessments become risky

Robb Knock, SimuTech Group Inc., Calgary

Have we as an industry – oil and gas drilling, production, refining and pipelines – learned anything from recent disasters and world events? The predictable answer is, "Yes, of course we did" followed by a litany of changes in operational management procedures, regulatory proposals, and enhancements to the risk analysis parts of our engineering practices.

However, a more candid answer might be, "Not enough." The risk analysis game is changing right under our noses, and not for the better. The petroleum industry may be grossly underestimating its potential liabilities. Over the years, potential liabilities from accidents have jumped from calculable, with a decent level of certainty, to incalculable, with a high degree of uncertainty.

Diminishing access to hydrocarbon resources and geopolitical uncertainties in the world's developing countries are compounding these challenges. In North America, exploration and production must deal with tighter regulations and greater scrutiny, And, in the US, with a litigious legal culture. Risk analysis and risk management are struggling to keep up.

Numerically speaking, risk can be simplified to a formula: Anticipated cost multiplied by likelihood or, if deemed likely, multiplied by frequency. However, common sense tells us that these estimates and predictions will never be easy. Actual costs from roughly similar operating failures and accidents can vary from a few million dollars to more than a billion dollars, an enormous range to fit into any calculation. Because they require predicting the future, likelihood and frequency can be even harder to quantify.

Among many other imponderables in risk analysis are the consequences of on-the-spot operational decisions that have enormously different consequences. An example comes from two drilling rigs owned by Transocean Ltd., one of the world's largest offshore drilling contractors, and decisions made just a few days apart in April 2010. The decisions were to replace drilling fluids (mud) with seawater, a common practice in offshore operations. On both rigs, blowouts occurred.

The blowout on one rig, Sedco 711 in the North Sea, was contained by the rig's blowout preventer (BOP). There were no serious injuries and damage was less than US$10 million. The other blowout was the Deepwater Horizon in the Gulf of Mexico. As it caught fire, exploded and sank, 11 workers died. For BP Plc (British Petroleum), to which Transocean had contracted Deepwater Horizon, that disaster has led to pretax charges against earnings of US$43.5 billion as of May 2011.

Virtually all the industry's players are rethinking how they design, engineer, and build for the Gulf of Mexico. The same is true for Canada's sprawling new heavy (unconventional) oil sands in northern Alberta and the pipeline projects that get this oil to market.

Technological advances need to go hand in hand with changes in corporate culture. Designers need to speak out about risks they perceive and explain why they are skeptical. Risk analysts and risk managers need to be more open to the skeptics and be more skeptical themselves about safety margins, engineering rules of thumb, and industry standards that have been in use for many years.

Upper management needs to allocate time and money in globally competitive marketplaces for deep dives into real-world risks and liabilities and ascertaining the ramifications of responsibilities and decisions.

In oil and gas, engineers are tasked with designing equipment that grows bigger and more complex every year. As every engineer is taught, uncertainty can grow geometrically and perhaps even exponentially, as size and complexity increase. Layers of uncertainty pile on, starting with who really, deep down, fully understands how these enormous new oil and gas structures work.

Economy of scale leads to mammoth projects to extract hydrocarbons profitably from increasingly difficult places. These places include depths below 20,000 feet (where pressures per square inch approach incalculable) and stormy, frigid arctic waters. It is little wonder that the costs of producing energy rise incessantly.

As the oil and gas industry drills six miles down with rigs floating in water a mile deep, it's easy to see why the industry's customary risk assessments fall woefully short of today's geopolitical and environmental realities.

Consider new unconventional reserves such as shale oil and gas. Recent findings in Pennsylvania show that the hydraulic fracturing, or "fracking," process may affect local underground water sources. Could this have been foreseen? Perhaps. There are lots of hard science and disk drives filled with statistical data. But with unconventional reserves, that may not count for much. And data (or lack of it) does not automatically point to negligence. Nevertheless, there is a potentially big environmental and litigation problem.

Developing unconventional sources of hydrocarbons and tapping deep reserves demand innovation and (forgive us) cautious daring rather than incremental steps. This is a time for leaders rather than managers.

Industry standards, safety factors, and rules of thumb have a built-in bias toward incremental steps and away from innovation. And in part due to litigation, a hostile media and unfriendly regulators, risk taking is being discouraged just when it's most needed. Unfortunately these challenges aren't going away anytime soon. From a business perspective, however, understanding risks properly can be an opportunity to differentiate one's firm from the competition.

Risk-analysis experts are fully aware that many time-honored guidelines no longer stand up to scrutiny. Common sense tells us these guidelines bear hidden risks of their own. Over-reliance on them makes iffy situations worse. This adds up to what my company, SimuTech Group, regards as Exhibit A for the transformation in the ways we look at risk.

Engineering-intensive organizations such as ours are being contracted more frequently to apply the latest numerical techniques and solutions, especially for analytics, the science of understanding inherent behaviors of systems. Expert insights, based on decades of experience, are sought by many oil and gas companies to beef up their internal risk-analysis processes.

Over the past 20-plus years, we have learned that nothing is ever foolproof. Nor are there any straightforward, easily grasped answers, much as we wish there were. Woe to the engineer who is unable to explain to upper management why. We doubt if easy answers ever existed, and our savviest customers concur.

The oil and gas industry has entered a new era of risk analysis and risk management. In this new era, anything may be dangerous if its engineering relies on outdated company safety margins and safety factors, obsolete engineering rules of thumb, or industry standards that no longer apply.

Risk analysis and risk management today should draw on numerous oil and gas experts to become an engineering team's "simulation arm." These experts can help build the requisite body of knowledge, put it through peer review, and then transfer that knowledge to the users' simulation tools, and finally, add it to their best practices. This third-party validation lets companies become more self-sufficient in dealing with risk.

There are four fundamental factors in up-to-date risk analyses:

• Accidents and spills, even in remote areas, are televised on the evening news and often "go viral" on YouTube, Facebook, and other social media.
• There is no real security for corporate and professional reputations in conforming to guidelines, yesterday's or today's. Many once-reliable margins, rules, and standards have been made irrelevant. They are industry roadkill, kicked to the curb by technology, economics, and the steady accumulation of experience and understanding of the systems and components we build.
• Caps on damage liabilities, statutory or otherwise, have been made meaningless by litigation, especially in the US. Litigation is not just a risk factor. For engineers, it is becoming the risk factor because litigation can dwarf all the other costs of an accident combined.
• Liabilities have become open-ended, and can quickly add two or three zeroes to a company's exposure by two or three orders of magnitude.

This means conventional, tried-and-true risk metrics in oil and gas cannot reliably assess the scale, scope and magnitude of foreseeable impacts-especially if the metrics are simply based on the costs of reimbursing customers for failed components, assemblies or systems. This is an outstanding example of the Fallacy of Misplaced Precision.

Constraining the liability metrics to a bill of materials (BOM) is completely inadequate and unrealistic. This blinkered approach can lead to grossly underestimating potential impacts. This is why traditional risk analysis is losing ground to the more predictive approach of Failure Mode and Effects Analysis (FMEA). (See "Specialized tools for advanced risk analysis," page 32.)

Other factors are at work in the persistent underestimation of risk. At the leading edge in any industry, accidents happen more frequently than across the entire industry; this is obscured by industry averages commonly used to calculate risks. Historical industry data underestimate future costs of spills, blowouts, fires, and rig explosions. Some indirect costs may have been left out.

Conventional risk analyses usually represent too conservative a view of conditions in the field, and often embody outdated views of components and systems. This gets to an engineering and risk-analysis paradox: What seems like a prudent, and even cautious, approach itself has hidden risks, in some of them unacceptably high.

Even harder to understand is that the knottier uncertainties are often zeroed out. They are simply wished away, in other words. The cost of a failure cannot be fully defined until its liability is understood as the sum of all the risks in the system. Conventional risk analyses also ignore the damage done to reputations by failure. (See "Traditional approaches to risk analysis," page 33.)

From a risk-analysis standpoint, this is culture shock. The oil and gas business grew to span the globe by taking risks, wise or otherwise, not by avoiding them. Liability always equals the product of two estimates: the cost of an accident or failure times the odds of its happening. To the extent they are ruled by physics and governed by design, both are directly calculable.

Beyond that, not so easy. This is more about the numbers than the calculations those numbers are pushed through.

Risk lies at the heart of engineering. Dealing successfully with risk is the heart of engineering, and thus at the heart of innovation as well.

What to do? Start with words, numbers, and expertise. "Failures in Waiting" is a much more descriptive, and more accurate, way to characterize risks than "potential liabilities." Viewing risk as the sum of potential liabilities can never suffice. It's too easy to ignore or rationalize away and won't grab the attention of hard-pressed upper management executives.

"Potential" merely implies a probabilistic eventuality, a statistical function, a bell-shaped curve. Statistics, dry or otherwise, rarely change the course of events except maybe after the fact, when it's too late.

Insist on questioning all the numbers in every risk evaluation. That won't be easy, of course, but avoiding euphemisms in favor of more compelling terminology should help garner the necessary support in time, data, and budget. And finally, outside expertise can be of enormous value.

About the author

Calgary-based Robb Knock is regional director of SimuTech Group Inc., a high-end engineering software and services company with 14 US and Canadian offices specializing in engineering simulation solutions from ANSYS Inc., Autodesk (Moldflow), FE Design, and Safe Technology.

More Oil & Gas Financial Journal Current Issue Articles
More Oil & Gas Financial Journal Archives Issue Articles
View Oil and Gas Articles on PennEnergy.com