Almost exactly 2 years ago there was a massive spill of crude oil from an Enbridge pipeline in Michigan. Ultimately about 3 ML of oil (a good couple of Olympic pools) was released and entered the Kalamazoo River, with short-term health effects on about 300 local residents plus extensive environmental damage that is still continuing. The oil did not ignite but in every other sense this was a very serious pipeline failure.
The US National Transportation Safety Board yesterday released its conclusions and recommendations on this failure (press release and report synopsis). The Board is scathing about Enbridge actions in both the lead-up to the spill and the response once it occurred. I find it particularly interesting that the opening words in the NTSB press release are “Pervasive organisational failures …”. This yet again reinforces that in technological systems the ultimate causes of failures are human and organisational and not purely technical.
In this case the immediate cause of the failure was stress corrosion cracking. SCC is a known problem in many pipelines, with known consequences and known methods of monitoring and management. However the NTSB says Enbridge failed to monitor SCC adequately and misinterpreted the data that they obtained – organisational failures at several levels. Then when the pipeline ruptured it took the Enbridge control room 17 hours, including two attempted restarts and multiple breaches of their own procedures, to realise that the oil they were continuing to push into the line was not being delivered to its destination but was being pumped into the environment – more organisational failures. There were also significant regulatory failures.
The message of the Andrew Hopkins and Energy Pipelines CRC Research Program 4 on the sociology of safety needs to be spread far and wide, if only those who need to know could be persuaded to listen.