Piper Alpha, 38 Years On

Thirty-eight years ago today, a permit failed to cross a shift handover. By midnight, 167 men were gone.

That was Piper Alpha — 6 July 1988, still the deadliest offshore disaster in history. I've spent the last two weeks writing about it, and about two others: Texas City, 2005. Toledo, 2022.

Different plants. Different decades. Different equipment. A missing safety valve, an overfilled tower, an alarm flood.

But stand back far enough and it's one story, told three times.

At Piper Alpha, the night crew didn't know the valve was missing — the suspended permit never crossed the handover.

At Texas City, the day shift didn't know how full the tower already was — the logbook was incomplete and the face-to-face relief never happened.

At Toledo, the signal was there all along — buried under 3,712 alarms, too noisy to read, let alone hand to the next crew.

184 people. Three inquiries. One finding, repeated for three decades:

The most dangerous place in a process plant isn't inside the equipment. It's the gap between two crews.

Think about what we've engineered since 1988. Interlocks. Trips. Permit systems. Alarm rationalization. Layer upon layer of protection — all of it on the equipment.

And the handover? In most plants it's still a conversation, a notepad, and someone's memory at the end of a twelve-hour night. The one link in the chain we never engineered.

That's what I'll be digging into over the coming posts: what a handover actually has to carry, why plants drown in data but starve for capture, and what it takes for an open item to survive the night without depending on anyone's memory.

For today, one question — the same one those three inquiries kept circling: in your plant, what carries the shift's knowledge across the gap. A system? Or a memory?