Failures Do Not Happen Overnight

Jim Rohn said, “Failure is not a single, cataclysmic event. You don’t fail overnight. Instead, failure is a few errors in judgment, repeated every day.” That quote struck a chord with the recent failure at Oroville dam, as indeed failures do not happen overnight.

In Oroville Dam’s “Sudden” Crisis, We Can Cite Various Historical Deviances

Here is a tentative list of historical deviances:

  • Oroville Dam annual inspections carried out by California Division of Safety of Dams found water “seepage” on the face of the dam and warned about potential structural steel failures since 2014.
  • The emergency spillway had insufficient drainage apron maintenance and protection.
  • The Hyatt power plant had one turbine down for “routine” maintenance for four years.
  • The Thermalito powerhouse, with 114MW capacity, has been unrepaired for five years since it was damaged in a fire.
  • The dam/river bypass valve was decommissioned and not repaired for eight years.
  • There were apparently concerns regarding cracks in the outlet works, nearing end of design life.

This list looks a lot like an example drawn from Jim’s quote or, as we would call it in risk adviser glossary: normalization of deviance.

Failures do not Happen Overnight

When we perform a risk assessment, or the review of one prepared by others, we usually carry out inquiries with key personnel. Key personnel covers key figures, not just management. This simple technique is highly efficient in revealing “untold truths.” That is especially true when compared to the usual workshops where big personalities tend to dominate.

The inquiries are also good for understanding the system under consideration. Defining the system is paramount. That is especially since normalization of deviance can creep up and transform a perfectly functional system into a deficient one, unbeknownst to people close to the system itself.

Examples of catastrophes due to normalization of deviance are:

  • Upper Big Branch Mine disaster: This disaster was the worst in the United States since 1970.
  • The Space Shuttle Challenger disaster: This explosion caused the death of seven crew members in 1986.
  • The cruise ship Costa Concordia crash: Negligence of the captain and crew were the cause of the ship sinking in 2012.

So, How Do We Solve This?

Below are a few points to consider when designing a solution.

  • Do you keep track of near misses in a database that can be crosschecked by type of hazard, threats-from and threats-to? Those data are important to detect normalization of deviance.
    To do this, proper definitions of “business as usual” and “force majeure” are necessary.
  • Defining success criteria and failure criteria is paramount. Without these clear definitions, any attempt to evaluate performances, operational and tactical risks will be in vain.
  • If your risk register has automatic triggers that update the values of likelihood and consequences after each event or at predefined time interval, you are in control!
  • Your risk register should fully empower you to act where you need to maximize efficiency and give you a clear roadmap.