Buncefield Incident Revisited

Anyone remember the explosion at the Buncefield fuel depot in December 2005?  At the time I wrote this:

Being a risk engineer in the oil and gas business and all that, here’s my take on it: the root cause is a human error (most probably someone not following procedures) made during a non-routine operation (most probably maintenance). In short, somebody did something he was not supposed to when trying to fix something. Let’s see if I’m right.

Let’s see indeed.  A colleague has pointed out that the final report of the incident has now been published.  So was I right?


It seems the root cause was faulty instrumentation and controls causing the overfilling of a tank during normal, routine operations.  This is pretty unusual. The reason I stated what I did with some confidence is that the vast majority of oil and gas incidents happen during non-routine maintenance, the Piper Alpha disaster having been the most deadly.

So much for my expert analysis.  I’ll stick to throwing rocks over the side of ships.


4 thoughts on “Buncefield Incident Revisited

  1. I think its fair enough Tim though to guess routine analysis. Just looking at the Texas incident which was down to human behaviour error and people becoming lax about the responsibilities.

    But you could also be correct if you question when those instruments were last checked for accuracy, as they should be on the routine schedule for maintenance. I have seen some operators who are religious about checking their instruments and fixing them should they be faulty on an almost monthly basis while other operators i know have routinely ignored what their instruments are saying as they are known to be faulty and have not be replaced yet. Which you could argue is about a failure in human behaviours just as much.

    On a side note, Have you ever seen the BP presentation of their finds from the Texas refinery. They came round 2 years ago to present it as a part of an industry wide door. Very interesting and was amazing to see how it happened but at the same time you could see how they slipped into this way of thinking, a little by little.

  2. Good on you to own up to faulty guesswork.

  3. It was a good guess, though.

    Both the Flixborough and the Piper-A disasters were caused by the sort of thing Tim mentions.

    In the former incident, and temporary pipework diversion that hadn’t been properly designed, and the in the latter a failure of the PTW procedure.

    It’s usually the right thing to look at first.

  4. Actually, you were mostly correct. Key contributing factor (though technically not cause) was human error during maintenance – not properly employing the ultimate high alarm after periodic testing.

Comments are closed.