On 16 Jan 03, a piece of insulating foam struck the left wing of Space Shuttle Columbia during launch. On 01 Feb 03, Columbia broke apart during re-entry, killing all seven crew members.

The foam strike was not simply an unforeseen technical failure. Foam shedding had occurred on previous missions, and because those missions had returned safely, the abnormal gradually became accepted. Previous success began to function as evidence that the risk was tolerable.

In Episode 10 of Safety Jon: The Real Risk Podcast, Jon Hillis examines the Columbia disaster through the lens of normalisation of deviance and organisational risk management.

The episode looks at how repeated deviations can become accepted practice, how confirmation bias affects decision-making, why uncertainty can be interpreted in favour of continuing operations, how technical concerns can fail to reach decision-makers with sufficient force, and what happens when the burden shifts to engineers to prove something is unsafe rather than requiring decision-makers to establish that it is safe.

Columbia is an extreme example, but the organisational mechanisms are familiar. A damaged guard that has “always been like that”, a vehicle defect repeatedly deferred, a procedure routinely bypassed, or a control that fails without consequence can all become normal because nothing happened last time.

That is the trap. An absence of consequences is not evidence that a control is adequate.

Safety Jon: The Real Risk Podcast looks past paperwork, slogans and polished dashboards to examine what actually controls risk, and what happens when organisations convince themselves that abnormal conditions are normal.

For more Safety Jon content, articles, resources and practical WHS tools, visit Safety Jon.