Curse of Randomness
Recently, I facilitated a team through root cause analysis. Invariably, one of the team members professes to know the answer before we begin. In this case, Sally said she knew the answer, but she wanted to hold back and see if everyone arrived at the same conclusion.
After an hour of heated analysis, everyone was surprised by the root cause, especially Sally. As we finished up, she said: "I was so sure I knew the root cause, but as it turned out, I was completely off track. How is that possible?
The answer, as I discovered reading Leonard Mlodinow's book on The Drunkard's Walk - How Randomness Rules Our Lives is that people are wired to find patterns in their lives and work, even when there are none.
Fooled By Randomness
In the early 1990s when I worked in the phone company, the VP in charge of our building thought there were too many false fire alarms. Based on limited information and observation, the division's managers all decided, incorrectly, that the problem was microwave popcorn recently introduced into the breakrooms.
Before they pulled popcorn, the VP tasked Debbie, the building manager, and I with doing a little root cause analysis. In a couple of hours, Debbie and I discovered that a phone recently introduced to the market was the root cause. Radio frequency interference (RFI) from cell phones was causing improperly shielded fire detectors to go into alarm mode.
While managers could see and smell popcorn, they couldn't see or smell the RFI from cell phones, leading them to be fooled by randomness.
Mlodinow says: "[In] most of our life experiences, we observe a relatively small sample of outcomes, from which we infer information and make judgements about the qualities that produced those outcomes."
Quoted later in the book, Michael Faraday said: "Human perception is not a direct consequence of reality but rather an act of imagination."
Once a process reaches 3-4 sigma (about 1% error), I have found that our five senses and the mind's ability to analyze data lose their traction. We shift from analysis to imagination. We start to see patterns where none exist.
This is one of the greatest barriers to Lean Six Sigma. Employees and managers alike get promoted and rewarded for detecting and fixing problems within the range of their five senses. But the success of our native problem solving abilities causes us to doubt the need for better methods and tools.
Random is as Random Does
In statistical process control (SPC), we know that every process varies. On a control chart, points should be distributed in a random pattern with more points closer to the center line and fewer farther away. Statistically, however, it's unlikely for certain patterns to exist without a special cause.
Data that hugs the centerline, for example, may suggest that the data is being manipulated to give the desired result, not reported accurately. Mlodinow reports one of the early findings: "the patterns of randomness are so reliable that in certain social data their violation can be taken as evidence of wrongdoing."
Informed by Randomness
Mlodinow concludes: "Our assessment of the world would be quite different if all our judgments could be insulated from expectation and based only on relevant data."
Are you still letting gut-feel, common sense and trial-and-error to guide your improvements or are you using relevant data to guide your efforts?
Are you going to be fooled by randomness or informed by the facts?
Would you pull the popcorn or change the detectors to prevent false fire alarms?
Rights to reprint this article in company periodicals is freely given with the inclusion of the following tag line: "© 2008 Jay Arthur, the KnowWare® Man, (888) 468-1537, firstname.lastname@example.org."