BP Oil Spill

It was on April 20, 2010, that the BP Deepwater Horizon project ran into trouble. While drilling a new oil well in the Gulf of Mexico, the rig became enveloped in flames and collapsed.  Eleven platform workers died and the drill pipe, one mile below the platform, bent and fractured where it entered the ocean floor.  Immediately, oil started gushing through the broken pipe.

The Fix

The company made every effort to close the relief valves where the pipe entered the ocean floor, but each attempt failed and the flow continued.

At first BP minimized the extent of the problem and expressed confidence that it would not take long before it was under control.  But in the days and weeks that followed, it became apparent that conditions were not improving.  At one point, NOAA (National Oceanic and Atmospheric Administration) estimated that as many as 5,000 barrels per day (one barrel contains about 42 gallons of oil) were escaping, although reliable estimates were difficult to obtain at a depth of one mile below the water’s surface.  At the same time, public outcry over damages to wildlife, fishing habitat and the tourist industry were increasing.

Eventually, a containment cap was placed over the well head, a challenging feat because a maneuver like this had never been attempted at such depths and under such great atmospheric pressures.

But there were still concerns. The cap could force a leak elsewhere in the well bore pushing oil and gas through nearby bedrock and mud thereby making the environmental disaster even worse. Moreover, it was a temporary fix.  The permanent one would require the drilling of a relief well parallel to the troubled well.  Once completed, mud and cement would then be pumped to seal the out-of-control well.

In late August 2010, four months after the disaster occurred, the relief well was completed and the damaged well permanently sealed.  But the toll had been substantial.  An estimated 4 million barrels of oil had escaped into Gulf making it the worst environmental disaster in US history.  To put it in perspective, the second worst was the Exon Valdez spill in 1989 when 250 thousand barrels spilled into Alaska’s Prince William Sound.

Project Risk

Project risk assessment focuses on three factors.
•    Likelihood. The likelihood that a risk event will occur.
•    Damages. The magnitude of the damages that is likely to result if the event does occur.
•    Detection. The difficulty in detecting that something is wrong and needs to be addressed.
A serious risk, for example, would be one whose likelihood is high, whose damages could be considerable and one where it would be difficult to detect the early stages of an incident.  The problem is, of course, that if one or more of these issues is underestimated risk management is unlikely to be given the emphasis and attention it deserves.

Consider the case of the BP oil spill.
 
William Reilly, the co-chairman of President Obama’s Oil Spill Commission, said that the oil industry had not kept up with the way it assesses the risk of catastrophic damages from spills. He implied that BP and the industry ignore more effective ways of not only assessing risk but also assessing the extent of the damages should a catastrophic event occur.

Professor Nancy Levenson, a professor at MIT, spoke at a seminar about the challenges in addressing BP project risk. She suggested that there were flaws in the safety culture and that these companies suffer from a culture of denial.  They accept that their industry is risky, that accidents are inevitable, and that safety is improving.  And by accepting these risks they are denying their part in preventing such disasters.

The challenge of detection was raised by Bob Bea, a University of California engineering professor.  In congressional testimony he said that BP ignored the signs that something was wrong through “overconfidence and incompetence.”  He raised the point that ineffective behavior (overconfidence) and inadequate training (incompetence) made it much more difficult to detect problems and properly manage and control the project.

A report issued in November 2010 by a panel of US scientists from the National Academy of Sciences was even more critical of management.  It said that BP and its contractors failed to use the lessons from earlier problems and concluded that better checks and balances might have prevented the disaster.  They mentioned an insufficient consideration of risks, flawed decisions, a lack of management discipline, minimal training standards, lack of expertise on the drilling rig, and the failure of regulators in overseeing oil drilling in the Gulf.

BP Oil Spill and the Columbia Shuttle Disaster
Stanford physics Professor Douglas Osheroff suggested that the BP oil spill disaster shared a management culture not unlike the one that contributed to the Columbia shuttle disaster in 2003.  “None of these systems are fail safe,” he said, “People don’t spend enough time thinking about what could go wrong.”

How were the BP and Columbia (NASA) cultures similar?

The Columbia shuttle had flown many missions before the disaster and on every one of those flights a piece of foam had broken away from the main propellant tank during launch.  Each time the foam did no damage.  But the February 2003 flight was different.  A piece of foam, about the size of a briefcase, fractured the leading edge of the shuttle on takeoff.  Yet it was the kind of risk event that was difficult to detect and as a result went unnoticed by mission control. Upon reentry into the earth’s atmosphere the fracture overheated, the vehicle’s protective shield was breached and the vehicle disintegrated. All seven crew members died.

The Columbia Accident Investigation Board was very critical NASA’s management culture whose mantra was Better, Faster, Cheaper. It was a culture emphasizing schedule and cost at the expense of safety.  It is reasonable to conclude that it was this emphasis that suppressed any “thinking about what could go wrong’.

While the two disasters, BP and Columbia, are quite different in some respects, what they do share is they both failed to create an effective risk management plan.

Lesson Learned

Preparing a risk management plan requires that project managers think objectively about what could go wrong and then make tough tradeoffs between time, money, resources and risk.  But, as we are reminded by the BP and Columbia disasters, taking the time to develop a risk plan is easier said than done. As Osheroff suggested, “People don’t spend enough time thinking about what could go wrong.” And often the reason they fail to do this can be traced to management culture.

What should we do? Is the answer to put even more emphasis on our project management systems?  Levenson doesn’t think so. She says that the emphasis needs to be on changing humans rather than changing the systems in which humans work.