Birthday Ramblings on Errors and Work Design
Kevin Meyer, over at Evolving Excellence, recently wrote a good piece on “endorsing” a U.S. presidential candidate. Today, I'm technically Constitution-eligible to be President of the United States, as I'm now 35 years young. However, I'm not nominating myself to be a write-in candidate. You'd have to be crazy to want that job.
Maybe it's age, but when I returned home last night (it's good to be home on your birthday), I lost my car at the airport. OK, it wasn't age, it was a process problem.
I went to follow my usual DFW parking routine. To track which terminal and level I parked in, I always jot a note in my BlackBerry calendar when I depart, such as “A21B first row” so I can find my car when I get back. Last night, I looked back to my departure date — NO NOTE! Oops. What happened? My BlackBerry has been glitchy lately and it deletes old emails (not 30 days old, but like yesterday's messages). Maybe the same thing happened with my calendar.
So I jogged my memory (being less than half Sen. McCain's age) and rememered we left from Terminal A and I parked on the second row… but which of the three parking structures? I had to walk around for 20 minutes, hitting the alarm key on my car remote (thank God for that feature). At one point, I had a false alarm… it was some other guy using the same feature to find his car.
My lesson learned — I relied on technology… and the technology failed. Maybe I need a better, more robust system (like a good ole' scrap of paper).
This reminded me of something I read on the way home. USA Today (via a WSJ Blog) reported that the recent airliner crash in Spain was caused by the pilots forgetting to set the flaps properly for takeoff. This was human error. But…. the technology failed them. The warning system (a “poka yoke” of sorts) failed to sound an alarm before takeoff, leading to the crash. The USA Today reports that this same error (forgetting to set the flaps) has occurred 50 times in the US recently… but the warning system worked.
So, sadly, it's just a matter of time before the same crash is repeated.
This type of crash hits home to me because I remember vividly, as a kid, the Northwest Flight 255 disaster, where the plane crashed on takeoff from Detroit's Metro Airport, killing nearly everyone on board. The pilots forgot to set the flaps. Something about them being distracted in the cockpit caused this, perhaps. And it led to the technology solutions… that work except when the process is still bad. Maybe it's the impact that this error had on my young mind that led me to be so interested in systems, work design, and error proofing?
Rather than relying on technology, how can we improve the system (or the standardized work) to prevent:
- Distractions in the cockpit (or from tower radio communications), cited in many recent incidents
- Fatigued pilots
- Other systemic causes that might lead pilots to not check the flaps, such as being rushed to hit a takeoff slot or poor management oversight
That's why we need process. Technology can't do it all.
Look at hospitals. Many are addressing the problem of medication errors with technology — bar code systems. These systems work great unless clever nurses work around them, by printing out extra bar codes so they can “save time” by batching up the scans instead of doing each one at the patient's bedside. We have to, in cases like this, ask “why” nurses would feel pressured to work around these systems. Hint — it's not because they're “lazy” or they “don't care.” It's due to problems in the system.
Likewise, why do pilots “forget” to set the flaps? Their life is on the line — we need to help them through proper work design and standardized work.
“Multi-tasking” is often a cause of errors or forgetfulness. The pilots are multitasking in the cocktpit. We all know medical professionals are forced to “do two things at once.” But multitasking is a myth. In fact, there's a good little book on this subject (look for a full review later) called The Myth of Multitasking. The book rightfully points out that humans can, at best, switch quickly between two different tasks. And that switching causes downtime in the brain (the cost of switching tasks and trying to get back on track).
We have to do a better job of designing jobs so that interruptions don't occur as often. In hospital labs, I've been involved in redesigning work so that the highly skilled Medical Technologists are no longer interrupted by phones ringing right into the work area. These calls interrupt their work and, ironically, would slow down their testing work when the calls are often of a “where's my test result?” nature.
By improving flow (and turnaround time), the number of calls go down, but when work is redesigned, front desk staff take the calls, since they're equally capable of looking up result status. They can transfer calls back to an MT if really required. Even then, the MT's work responsibilities are divided so ONE can always focus on the main testing flow. In the old design, “everybody” took calls which meant everybody could be pulled away from the main flow of work. Better work design leads to fewer interruptions and better results for all.
I see the same thing in a doctor's office front desk. The front office staff all “do it all”, from answering phones to working with patients in person. We're experimenting with different breakdowns of roles and responsibilities, so one person can focus on the patients who are physically in front of them, without all of the frustrating interruptions.
We can fix many of these problems with Lean and engineering principles. But not today, for me… I'll be relaxing and enjoying the rest of my birthday. But back to it on Monday…