From Sustained Success to Sudden Tragedy

⌕ Opening

Repeated success can sometimes conceal the conditions of an approaching failure.

When capable people and high-performing organizations continually achieve their objectives, they naturally develop confidence in their practices and decisions. That same confidence can also produce blind spots. For example, when small departures from established standards result in no apparent consequences, those departures can gradually become accepted practice. Even when the departures are noticed, their significance may go unrecognized.

On January 28, 1986, the Space Shuttle Challenger broke apart just 73 seconds after launch, killing all seven crew members aboard. The tragedy was sudden, but the conditions that contributed to it had developed over years of seemingly successful operations.

The Challenger disaster reveals a disturbing paradox: Repeated success can often hide gradual erosion of performance standards.

▣ Brief Background

By January 1986, NASA's Space Shuttle program had successfully completed 24 missions. But behind those successes, a critical performance standard was gradually eroding.

The solid rocket boosters relied on rubber seals called O-rings to prevent extremely hot gases from escaping through joints in the boosters. The original design did not anticipate any O-ring erosion during flight. The seals were expected to perform their intended function every time without damage.

However, following the second shuttle mission (STS-2) in November 1981, inspectors discovered erosion of a primary O-ring. Although this represented a departure from the original design standards, the mission itself had been successful.

Additional evidence of O-ring damage emerged during subsequent missions. By March 1984, NASA engineers were formally discussing acceptable erosion, using calculations and testing to establish how much damage could be tolerated. What had originally been an unanticipated condition was increasingly treated as a manageable one.

This gradual redefinition of acceptable performance illustrates what sociologist Diane Vaughan later described as normalization of deviance: a process in which departures from established performance standards become regarded as normal through repeated experience without catastrophic consequences.

⌕ Performance Investigation

The critical change in NASA's performance standards did not begin with a single decision to lower safety expectations. It developed gradually through changes in how acceptable performance was understood.

The original design and performance expectation was zero O-ring erosion. Yet the organization repeatedly experienced successful missions despite evidence of erosion. Each successful mission provided further reassurance that the original standard might have been unnecessarily high. Engineering analyses increasingly supported accepting a degree of erosion. Gradually, what had initially been considered a defect became an increasingly tolerable condition.

By January 1986, the question had shifted from whether O-ring erosion was acceptable to how much erosion could be tolerated.

The evening before Challenger's launch, engineers warned that unusually cold temperatures could compromise the O-rings' ability to seal. Their initial recommendation against launching was ultimately reversed, and the shuttle launched the following morning.

The absence of previous catastrophic failures had helped create confidence that the underlying problem did not justify stopping operations.