You are here

Lessons from Macondo

07th November 2013
Lessons from Macondo

Bob Fryar is executive vice president of Safety & Operational Risk at BP.

Posted: 
7th November 2013
A quarter of a century after the tragic Piper Alpha disaster that claimed the lives of 167 people, BP’s head of safety and operational risk Bob Fryar looks back at how offshore safety has progressed in the industry since 1988, drawing on the company’s own experience of and response to the blowout of the Deepwater Horizon rig in 2010 – lessons from the past to prepare for a better future
Lessons from Macondo
Gas from the damaged Deepwater Horizon wellhead is burned by the drillship Discoverer Enterprise, May 16, in a process known as flaring. US Coast Guard Atlantic Area , Gulf of Mexico, 2010. Image by Petty Officer 3rd Class Patrick Kelley

 

When Piper Alpha happened 25 years ago, I was still very new to the industry. I had only worked in it for a couple of years. As for most people, the thing that stood out to me with great sadness was the loss of life and the impact that the accident had on so many families and communities.  It also tragically reinforced the fact that we work in a hazardous industry. Our task is to bring energy to people around the world. That task is full of hazards and risks that must be understood and managed at all times.

 

When we review progress since 1988, it is inevitable that people will focus on the accident we and others experienced three years ago, on 20 April 2010.  That was the day when the Deepwater Horizon rig experienced a blowout as its crew was working on the temporary abandonment of the Macondo well.  126 people were on board.  

 

Hydrocarbons escaped from the well resulting in explosions and a fire that burned for two days until the rig sank. Tragically, 11 men lost their lives and others were injured. Hydrocarbons continued to flow from the well for 87 days. 

 

The response to the oil spill itself was a complex and massive undertaking. Plugging the leak meant working with robot submarines to apply containment equipment to a high pressure well a mile below the ocean. No one had done that before.  Cleaning up the spill meant a huge effort with thousands of people and vessels as well as airplanes and the largest deployment of boom ever – approximately 13.5 million feet. 

 

At the time, I was running BP’s deepwater business in Angola and I was one of many leaders from around BP who flew to the Gulf. I spent four months working on the response. When I arrived, shortly after the accident occurred, my first task was to call other companies to seek their help. I remember making phone calls in the morning, and, by the afternoon, we had experts from Anadarko, Shell, Exxon and others in the office helping us. Additionally, other companies called and asked how they could help.  And as I mentioned earlier, the experience generated a deep desire across the team to learn from the accident and to seek to prevent the repetition of such a tragedy. 

 

Following the accident, Mark Bly, who was my predecessor as head of the safety and operational risk function, led the BP internal investigation of the accident, which was conducted by a team including internal and external expertise.

 

The resulting Bly Report included 26 recommendations addressing important areas of deepwater drilling, including cementing guidelines, equipment certification, assuring the competence of individuals, and testing of blow-out preventers. 

 

The implementation of these recommendations is an on-going major programme of work within BP. Each recommendation has to be applied across multiple locations – and many require new processes or agreements with contractors. 

 

As well as implementing the recommendations of the report, we have also taken a number of broader measures to further support safety and risk management in our upstream organisation. 

 

One of the first things that we did was to look at the organisation itself.  We wanted our upstream organisation to be structured in a way that would encourage the building of capability and the consistent application of standards across the world, wherever they apply. 

 

We did this by moving from an asset model to a functional model. What does that mean? It means that, instead of organising the company in regional teams, we organised our upstream in centralised functions that bring together the people who do the same jobs around the world. All the explorers report to the head of exploration. All the people who build new projects report to the head of the global projects organisation. The people involved in drilling, completions and interventions all report to the head of the global wells organisation. 

 

A benefit of this is that we can build on our expertise within the teams to deliver excellence in each function and also drive standardisation more readily where we wish to do so, with each team using standard procedures. We believe these procedures contribute to consistent implementation and safer execution of work.

 

We have also taken action with regard to the equipment and technology we use in actual deepwater drilling. We have reviewed our requirements for drilling rigs in service on BP-operated wells.  Any proposed departures from those requirements need approval from the appropriate person in our safety & operational risk organisation – what we call S&OR.  

 

But we are mindful that technology, plant and equipment can serve us in many ways.  Strengthening deepwater capability is not all about the kit that actually does the drilling.  We have also set out to use technology to enhance our integrated decision-making on drilling and wells.

 

In Houston, we have created a monitoring centre that enables offshore crews to consult in real time with onshore experts – viewing the same data and linked by video. While the responsibility for well monitoring remains with the rig crew, having a monitoring centre means more people can be available as resources in a given circumstance.

 

Obviously, our top priority is to prevent accidents but part of building deepwater capability is to be prepared for the worst.  Since 2010 there has been a strong industry-wide programme of activity in the area of spill response. 

 

At the international level, the Global Industry Response Group was set up and has launched several work-streams. One is looking at data from incidents and communicating good practice, so the entire industry can learn together. Another relates to developing a well capping toolbox. Another is focused on response in general – capturing the lessons we learned in areas such as relief well drilling and crisis management. BP has built its own capping stack and other containment equipment.  It is stored in Houston but can be mobilised worldwide quickly. 

 

Here in Britain, Oil & Gas UK has taken the lead and worked with the Oil Spill Prevention and Response Advisory Group to develop a response toolkit that includes a capping device.  So we are seeing a number of these devices positioned around the world.

 

We continue to see downward trends in process safety events and losses of primary containment (LOPCs), which are essentially leaks. In 2008, when we first put the LOPC metric in place, we had 658 releases. Last year we had 292. That was a 19 per cent reduction versus 2011.  Process safety events (PSEs) are categorized by tiers depending on their severity, with tier 1 being the most significant.  For BP, we saw a 42 per cent reduction in Tier 1 PSEs in 2012 on 2011. 

 

Tracking this data is only part of BP’s efforts to drive continuous improvement.  But we are beginning to see the benefits of the various on-going activities I have described.  Even one LOPC can have high consequences, and any accident is one too many - and of course there is always more to be done.

 

And while we believe these things are making a difference, we also know there is always more to do at BP and in the industry, and we must remain vigilant.

 

 The Piper Alpha and Deepwater Horizon accidents remind us all of the consequences when things go wrong. They also provide lessons from which we all can learn and improve.