Latest / Safe As Podcast / E72: Fatal Blind Spot: Why Leader Safety Walkarounds Failed Deepwater Horizon

Transcript

E72: Fatal Blind Spot: Why Leader Safety Walkarounds Failed Deepwater Horizon · Safe As Podcast

• 0:00
  7 hours before the Deepwater Horizon rig exploded, 4 senior
• 0:04
  executives flew out for a safety visit.
• 0:06
  They walked the site, asked questions, congratulated the
• 0:09
  crew on seven years without a lost time injury, and then later
• 0:13
  the rig blew up. Andrew Hopkins analysis shows
• 0:16
  something uncomfortable. The visit didn't fail because
• 0:18
  leaders ignored safety. It failed because they focused
• 0:22
  in hindsight on the wrong kinds of threats.
• 0:25
  Today's paper isn't about incompetent people or bad
• 0:28
  apples. It's about how well made safety
• 0:30
  walkarounds can quietly reinforce a risk blindness
• 0:33
  within organisations good everyone welcome to safe as I'm
• 0:37
  doctor Ben Hutchinson. Today's source is Hopkins 2011
• 0:41
  titled management walkarounds lessons from the Gulf of Mexico
• 0:44
  oil well blowout published in safety science.
• 0:47
  Hopkins main claim is that management walkarounds, even
• 0:50
  when well intentioned, often drift towards routine
• 0:54
  Occupational Safety issues like slips, trips, housekeeping, or
• 0:58
  systematically missing major hazard risks and a note upfront.
• 1:02
  This kind of analysis is inevitably shaped by hindsight
• 1:05
  and counterfactuals. Still, Hopkins argument is
• 1:08
  careful and worth seriously considering. 7 hours before the
• 1:12
  blowout, 4 company VIPs helicoptered onto the rig for a
• 1:15
  management visibility tour. Despite their focus on safety,
• 1:18
  their experience as rig managers and drilling engineers, the
• 1:21
  cooperation of rig staff meant that some critical warning signs
• 1:24
  were missed or misinterpreted. Hopkins argues that although the
• 1:28
  visit was framed as a safety, the VIP's paid almost no
• 1:32
  attention to safety critical activities that were occurring
• 1:35
  during their visit. Their attention was drawn
• 1:37
  instead to a recent slip injury elsewhere and on congratulating
• 1:41
  the crew for seven years without a lost time injury.
• 1:43
  The day of the visit, critical well testing was underway.
• 1:46
  Personnel were actively debating how to run the test and how to
• 1:50
  interpret its results. The PS were told this, but
• 1:53
  quoting the paper, asked no more questions about this and moved
• 1:56
  on to a social conversation about the history of the
• 1:58
  company. Later, 1 VIP asked the rig
• 2:00
  manager whether the test had gone well.
• 2:02
  After receiving a thumbs up, he accepted the reassurance.
• 2:05
  Hopkins notes of this question clearly invited the response he
• 2:08
  got and that he did not probe for evidence.
• 2:11
  The VIP said that they would have offered advice if asked,
• 2:14
  but they weren't. Hopkins comments that there was
• 2:16
  no recognition that this was an opportunity to do some ordering,
• 2:20
  to check on competence and to verify that they were complying
• 2:23
  with procedures critical to safety.
• 2:25
  Mudflow monitoring that day was severely degraded, making it
• 2:28
  difficult to reliably match inflow without flow.
• 2:31
  Hopkins argues that probing how monitoring was being done could
• 2:35
  have revealed serious deficiencies, but no one asked.
• 2:38
  The VIPs were also aware that one of the final safeguards
• 2:41
  against blowouts was about to be removed that afternoon.
• 2:44
  Yet their engagement remains social and uninterventional,
• 2:47
  rather than focused on how safety was being managed.
• 2:50
  Hopkins then asked why did this behaviour make sense to them?
• 2:53
  First, the VIP is focused on checking conditions rather than
• 2:56
  behaviours. Inspection records equipment
• 2:59
  status because these are easy to verify than decisions and
• 3:02
  judgments in complex processes. Second, they were reluctant to
• 3:05
  interfere given professional hierarchies and specialised
• 3:08
  roles. They didn't want to interject or
• 3:10
  bypass chain of command authority.
• 3:12
  Third, they saw intervention as justified only for obvious
• 3:16
  immediate danger, not for ambiguous process risks that
• 3:19
  look banal in everyday work. Next, Hopkins draws A
• 3:22
  distinction between personal safety and process safety.
• 3:25
  Personal safety hazards are often their frequent and low
• 3:28
  consequence types. These slips and trips and minor
• 3:31
  injuries. Process safety has it's rare,
• 3:33
  high consequence and often invisible in routine injury
• 3:37
  statistics. One VIP later identified his
• 3:40
  focus was on Occupational Safety.
• 3:42
  So personal safety, not process safety.
• 3:44
  He was concerned about whether pushing a button or turning a
• 3:46
  wrench endangered the person doing it, not whether doing that
• 3:49
  task might trigger an explosion. Hopkins then critiques the stop
• 3:52
  work authority. Workers on the rig had stopped
• 3:55
  work in the past, but only for personal safety risks.
• 3:58
  No one was aware of well operation has been stopped for
• 4:00
  major hazard reasons. One operator felt uncomfortable
• 4:03
  about simultaneous operations that made mudflow monitoring
• 4:06
  nearly impossible. But he didn't treat that
• 4:08
  discomfort as a stop work trigger.
• 4:10
  Hopkins and argues that leadership behaviour reinforced
• 4:13
  this by avoiding disruption and not probing deeply.
• 4:16
  The VIP sent a subliminal message.
• 4:18
  You'd need an exceptionally good reason to stop work.
• 4:22
  Finally, Hopkins proposes how walkarounds could be better.
• 4:25
  Leaders should be briefed on major hazards beforehand.
• 4:27
  They should understand critical controls and and verify that
• 4:30
  those controls work. They should probe our lessons
• 4:32
  from the past accidents have been transferred.
• 4:34
  They should observe safety critical processes directly.
• 4:37
  They should treat some parts of some walk arounds at least as
• 4:40
  partial audits, not just social visits.
• 4:42
  Even if this slows down work, that's it on Safe As.
• 4:45
  Please like, comment and subscribe on YouTube and links
• 4:48
  in show notes if you'd like to support the channel.