Latest / Safe As Podcast / Ep 10: Are safety myths holding us back?

Transcript

Ep 10: Are safety myths holding us back? · Safe As Podcast

• 0:07
  Are we stuck in a safety trap, constantly pointing fingers at
• 0:10
  human error or adding more layers of ineffective
• 0:14
  protection? What if the very beliefs guiding
• 0:16
  our safety efforts are holding us back?
• 0:20
  Good day everyone. I'm Ben Hutchinson and This is
• 0:22
  Safe as a podcast dedicated to a thrifty analysis of safety, risk
• 0:27
  and performance research. Visit safetyinsights.org for
• 0:31
  more research. Today's paper is from Bernard
• 0:34
  and Holnagel 2014, titled I Want to Believe Some Myths about the
• 0:39
  Management of Industrial Safety in Cognition, Technology and
• 0:44
  Work. The authors argue that many
• 0:46
  industrial safety practises are littered with fragile beliefs,
• 0:50
  rendering safety management flawed and ineffectual.
• 0:54
  They define a myth not merely as an assumption, but as an idea or
• 0:58
  story that many people believe, but which is not true.
• 1:02
  These myths are deeply rooted in our culture, widely shared, and
• 1:06
  profoundly influence decisions and actions.
• 1:09
  The authors contend that acknowledging these myths is a
• 1:12
  crucial first step towards genuinely improving industrial
• 1:16
  safety. Let's explore 6 myths.
• 1:19
  Myth 1. Human error is the largest
• 1:22
  single cause of accidents and incidents.
• 1:24
  This belief is pervasive. For instance, a 2010 report
• 1:29
  declared that human error is involved in over 90% of all
• 1:33
  accidents and injuries in the workplace.
• 1:36
  This idea has a long history. It remains A fundamental part of
• 1:39
  many investigation methods, often marking the deepest point
• 1:43
  of analysis. However, the authors argue that
• 1:47
  this is a simplistic and counterproductive view.
• 1:50
  They highlight that labelling something human error is a
• 1:53
  judgement often made with hindsight bias.
• 1:57
  It kind of implies wrongdoing in seeking A culprit.
• 2:00
  Crucially, it typically focuses only on sharp end operators, the
• 2:05
  people directly involved with the process, ignoring the
• 2:09
  broader context and working conditions imposed by managers
• 2:13
  and the organisation. And if human error is the cause
• 2:16
  of an event going wrong, what about the countless times human
• 2:19
  actions make things go right? They propose A revised
• 2:23
  statement. Human error is an artefact of a
• 2:25
  traditional engineering mindset that treats humans like fallible
• 2:29
  machines, failing to consider the vital role working
• 2:32
  conditions play in shaping performance.
• 2:35
  Myth 2 Systems will be safe if people comply with the
• 2:39
  procedures. We often assume that following a
• 2:42
  procedure not only gets the job done, but gets it done safely,
• 2:46
  and any deviation from procedures automatically creates
• 2:49
  risk. This treats people as if they
• 2:52
  were machines. But this isn't helpful.
• 2:55
  Procedures are inherently incomplete.
• 2:57
  They can't cover every possible situation or fully describe
• 3:01
  their reaction. Humans constantly interpret and
• 3:04
  adapt to procedures based on the situation and their experience.
• 3:08
  The authors draw on a major accident and they highlight that
• 3:12
  rigid blind compliance can actually be detrimental to
• 3:15
  safety and efficiency. Human flexibility is essential
• 3:19
  to compensate for the brittleness of procedures and
• 3:22
  actually contribute to safety. A revised statement is actual
• 3:26
  working situations usually differ from what the procedures
• 3:30
  assume and strict compliance may be detrimental to practise.
• 3:34
  Procedures should be used carefully and intelligently.
• 3:38
  Myth 3. Safety can be improved by
• 3:40
  barriers and protection. More layers of protection
• 3:43
  results in higher safety. This seems intuitive, reflecting
• 3:47
  the defence in depth approach. However, the relationship
• 3:51
  between protection and risk is not straightforward.
• 3:54
  One reason is psychological. People often adjust their
• 3:57
  behaviour based on perceived risk.
• 4:00
  The authors tie this point to risk homeostasis, suggesting
• 4:03
  people maintain a certain level of comfortable risk.
• 4:07
  For example, some limited studies suggested that taxi
• 4:11
  drivers with ABS braking systems drove more aggressively in
• 4:15
  curves and actually had a slightly higher accent rate than
• 4:18
  those without ABS. However, this argument hasn't
• 4:21
  aged too well because the risk homeostasis hypothesis as
• 4:25
  advanced by Gerald Wilde isn't well empirically substantiated.
• 4:30
  But behavioural adaptations is well supported, meaning people
• 4:35
  can and do change their behaviour in response to
• 4:38
  interventions and designs etcetera.
• 4:41
  The second reason is technical. Adding protection inherently
• 4:45
  increases the system's complexity.
• 4:47
  More components, more couplings not only introduce new value
• 4:51
  points, but also exponentially increases the number of
• 4:56
  combinations that can lead to unwanted outcomes.
• 4:59
  A revised statement is technology is not neutral.
• 5:02
  Additional protection changes behaviour so that the intended
• 5:06
  safety improvements might not be obtained.
• 5:09
  Myth 4. Accidents have root causes, and
• 5:11
  root causes can be found. Root cause analysis is a common
• 5:15
  umbrella term for a number of different techniques and
• 5:18
  approaches, assuring that system parts are causally related and
• 5:22
  effects propagate orderly, allowing us to trace problems
• 5:26
  back to their origin. However, this method relies on
• 5:30
  assumptions that don't really hold for complex systems that
• 5:33
  events repeat, outcomes are strictly bimodal, that is
• 5:37
  correct or incorrect, and cause effect relations can be fully
• 5:41
  described. Human performance isn't bimodal,
• 5:44
  it varies. It rarely fails completely, and
• 5:48
  humans can recover from failures.
• 5:50
  When an analysis points to a human error as the root cause,
• 5:53
  it often overlooks how the same human flexibility makes things
• 5:57
  go right most of the time. The paper suggests that the
• 6:00
  preference for simple root cause explanations is due to a
• 6:04
  psychological desire for comfort and satisfaction in tracing
• 6:07
  something unfamiliar back to something familiar.
• 6:11
  Instead, the authors argue that unwanted outcomes in complex
• 6:15
  systems don't necessarily have clear, identifiable causes.
• 6:19
  A revised statement is human performance cannot be described
• 6:23
  as if it was bimodal. In complex systems, things that
• 6:27
  go wrong happen largely in the same ways that things go right.
• 6:32
  Myth 5. Accident Investigation is a
• 6:34
  rational process. While investigations are serious
• 6:37
  undertakings, they are rarely purely rational.
• 6:41
  Practical constraints like deadlines and resource
• 6:44
  limitations often dictate the depth of analysis and the
• 6:47
  methods used, becoming a trade off between efficiency and
• 6:51
  thoroughness. Moreover, every investigation
• 6:55
  method embodies some sorts of assumptions about how accidents
• 6:58
  happen. Perhaps most challenging is the
• 7:01
  need to establish responsibilities.
• 7:03
  This need can heavily bias investigations, making finding a
• 7:07
  culprit more important than understanding the contributing
• 7:11
  or causal factors. As Woods at output it,
• 7:14
  attributing error is fundamentally a social and
• 7:17
  psychological process and not an objective technical one.
• 7:22
  The paper suggests that Accident Investigation is a social
• 7:24
  process where causes are constructed rather than found.
• 7:28
  Indeed, investigators operate on a what you look for is what you
• 7:32
  find principal meeting their chosen method and their existing
• 7:36
  world views directs what they see and don't see.
• 7:39
  To overcome this, it's essential to pursue what they call second
• 7:42
  stories, deeper analysis that go beyond the simplified first
• 7:46
  stories of apparent causes. A revised statement is.
• 7:50
  Accident Investigation is a social process where causes are
• 7:54
  constructed rather than found. Myth 6 Safety first statements
• 8:00
  like safety always has the highest priority and will never
• 8:03
  be compromised, are often made for communication purposes,
• 8:06
  expressing noble values and goals.
• 8:09
  The paper cites a 2004 assessment of the BP Texas City
• 8:13
  refinery accident, which had a low injury rate but experienced
• 8:17
  a major explosion, killing fifteen people before the
• 8:21
  explosion, employees ranked making money, cost and budget
• 8:25
  and production as their top priorities with major incidents
• 8:29
  and the coming in 5th. Safety has undeniable financial
• 8:33
  implications. Safety costs are immediate and
• 8:36
  tangible, while their benefits are often potential and distant.
• 8:40
  This leads to trade offs where safety is often as high as
• 8:43
  affordable. Safety budgets, like any other
• 8:46
  constraint, are limited and decisions involve prioritisation
• 8:51
  and feasibility, often trading safety against economy.
• 8:55
  A revised statement is safety will be as high as affordable
• 8:59
  from a financial, risk, acceptability and ethical
• 9:02
  perspective. So some takeaways is that these
• 9:06
  myths shared across all layers of organisations in society lead
• 9:10
  to flawed safety practises. They are resistant to change
• 9:13
  because they are rarely questioned.
• 9:16
  The authors propose a radical shift.
• 9:18
  Instead of defining safety as a property that a system has,
• 9:22
  safety should be seen as a process, something a company
• 9:25
  does. It's dynamic, constantly
• 9:28
  negotiated, and varies in response to changing conditions.
• 9:33
  More importantly, the goal of safety should shift from
• 9:35
  focusing on what goes wrong to also understanding and enabling
• 9:40
  what goes right. We spend immense efforts
• 9:43
  preventing unsafe sort of functioning, but relatively
• 9:47
  speaking, hardly any efforts are directed towards bringing about
• 9:50
  safe and reliable functioning. Measuring safety solely by a low
• 9:55
  number of negative outcomes is insufficient.
• 9:59
  It should be tied to indicators of an organisation dynamic
• 10:01
  stability, its ability to succeed under varying
• 10:05
  conditions, its capacities to respond, to, monitor, anticipate
• 10:09
  and learn. In a complex world with multiple
• 10:11
  interacting constraints, operating perfectly is
• 10:15
  impossible. The safety myths covered here
• 10:18
  support an unrealistic ideal of safety management.
• 10:21
  To successfully operate increasingly complex systems, we
• 10:26
  need to abandon these myths and adopt more sensible and
• 10:28
  sustainable assumptions about safety.
• 10:31
  Hence, complex systems work because people learn to identify
• 10:36
  and overcome design flaws and functional glitches.
• 10:39
  Quoting Deco. People finish the design in
• 10:41
  practise. People can adjust their
• 10:43
  performance to the current conditions.
• 10:45
  People interpret and apply procedures to match the
• 10:48
  situation. People can detect when something
• 10:51
  is about to go wrong and intervene before the situation
• 10:54
  becomes seriously worsened. This means that systems work
• 10:57
  because people are flexible and adaptive, rather than because
• 11:00
  the systems have been perfectly thought out and designed.
• 11:04
  But of course, these same positive capabilities can also
• 11:09
  be our roads to ruins, because people may adapt in ways that
• 11:13
  are locally optimised but more hazardous at a global,
• 11:17
  organisational level. That's it on Safe As I'm Ben
• 11:21
  Hutchinson. Please help share, rate and
• 11:23
  review, and check out safetyinsightsorg for more
• 11:27
  research. Finally, feel free to support
• 11:30
  Safe As by shouting a coffee link in the show notes.