Understanding Constraints-at 6:00am on a Saturday
On Saturday, June 5, my wife and I flew to visit family. We booked a 7:00am flight out of Indianapolis. We arrived at the airport at 5:30am, only to find a line of people, six abreast, well over 1,000 feet long at security! How can this be?? In Indianapolis?? On Saturday morning?
We spent nearly an hour in line, which gave me plenty of time to a) get over my disappointment of not getting a Cinnabon for breakfast, b) chat with an old acquaintance of ours who happened to be in line behind us and c) analyze what was wrong with the system that such a huge line would develop.
When I see a slowdown like this, I instinctively look for the constraint, the single limiting factor that decreases total system throughput. And here's where I learned something new.
You are probably thinking, as I was, "The constraint is at the security section." Well, yes and no. When we finally got to a place in line where we could see the metal detectors, I was stunned to see so few people walking through them. In fact, my simple timing observations measured only one person moving through each detector every 30 seconds, when it appeared that the pace could well have been one every 10 seconds. Which meant throughput could have been doubled or even tripled!!! Tell that to the folks who were late for their flight! Security people watching the detectors actually appeared bored!!
This excess capacity was a clue...the metal detectors must be downstream from the real constraint. So, I looked upstream one step. And there it was.
Security staff was instructing all passengers to take off their shoes before moving through the metal detectors. And so, in the cramped space created in the Disneyland-like serpentine lines, people were trying to juggle luggage, small children and boarding passes while bending over, unlacing and removing their Nikes and Birkenstocks. This was taking much longer to do than to walk through the metal detectors. And so the entire 1,000 foot line was being hung up, two feet at a time (sorry, I couldn't resist the pun).
Having identified the constraint, what could be done about it?? Several things could happen... now...on virtually no budget...raising no security concerns.
- Put two rows of benches or chairs along side each line leading to a single metal detector. If people could sit, they could more safely and quickly take off their shoes. It took people twice as long to take off their shoes as to walk through the metal detector. Thus, two sets of chairs for each line.
- Put signage up, 40 or 50 feet sooner telling people to plan on taking off their shoes. No surprises...quicker compliance.
- Put entry carpets on top of the normal carpet and then vacuum it once an hour, to demonstrate concern for patrons' bare/socked feet. At least make people feel like someone cared.
In so doing, this would shift the constraint back to the most capital-intensive portion of the security position-the human and luggage metal detectors.
Where the analysis could happen again.
Alas, I went through the same security site again last Friday afternoon. This time, no line at all. But also no improvement in the throughput. Still, one person every 30 seconds. They were just lucky. Friday, at 3:30pm, had nowhere near the outbound passenger demand that early Saturday morning did.
Will we ever learn?
The tools of Lean and Theory of Constraints are phenomenal. TOC tells you where...Lean tells you how. Why don't more people embrace it? Why not use it in government? Why don't the airlines (who depend on both rapid and secure passenger checks) demand it? Why do we think each of our situations is so unique we can't learn from other arenas? Seth Godin wrote well on the foolishness of thinking everything is unique last week...please ponder his conclusion.
Thanks for listening. Apply this today. I hope it is helpful.