Quick Changeover for American Airlines Center
Even though I'm still a pretty die hard Red Wings fan, I live in Stars country now. Don't ask me what I'll do if the Wings and Stars meet in the second round. I went to the Stars/Canucks game last night, which started at 8:30 PM central. I was surprised when, while putzing around the house Sunday, I saw the Spurs/Mavericks game on TV and realized it was also being played at the American Airlines Center. That left them only about 3.5 hours between the end of the NBA game and the start of the NHL game. Somebody had to hustle!
They showed a stop-action video of them taking down the basketball court and taking the cover pieces off of the hockey floor (I wish that video was online). This link shows a timeline for the conversion from hockey to basketball.
The article I linked to above had some details, including:
When the teams cleared the court, Waugh and a group of 60 went to work at precisely 5 p.m., peeling away 4×8 pieces of hardwood, taking down goalposts and removing spools of ABC TV cables in preparation for Game 3 of the Canucks-Stars playoff series.
“Normally our changeover is 40 people but we wanted more because it's a playoff game,” Waugh said.
After the court was off, the ice deck was on with 15 minutes.
The crew finished at 6:28 p.m., making the total time of the job 1 hour 28 minutes — in plenty of time for the 8:30 p.m. start.
“That's pretty good. Our fastest time last year was 1:55,” Waugh said.
While we don't always want to “throw people at the problem,” adding people CAN be an appropriate “quick changeover” strategy in any context. I'm not suggesting the Stars or the arena management are trying to be “Lean,” but I'm trying to draw a parallel for the manufacturing world (or healthcare, for that matter).
60 people should be able to do the changeover work faster than 40, assuming you haven't reached the point where they are starting to get in each others' way. Taking the standard work for the floor/rink changeover (assuming there is so) and adding 50% more people should cut the time by dividing the work content into smaller pieces (assuming it can be divided evenly and can be done in parallel). There certainly would be a “bottleneck” or constraint step in the changeover process that would limit the changeover speed.
Other than adding people, what could you do to shorten the changeover time?
- “External Setup” — do as many prep activities BEFORE the end of the NBA game as possible. Are all tools and equipment ready to go? You want to externalize setup activities so you don't waste time looking for things (or moving things unnecessarily) while the changeover is actually taking place (time during the changeover is called “internal setup.”
- Identify the bottleneck step and speed it up, through some method
- Practice, practice, practice: Move up the learning curve through repetition and continuous improvement of the process (like a NASCAR pit crew) — after each changeover, have the team debrief and discuss what went well and what didn't (and what to improve). I guess that's really the “PDCA” cycle at work.
Any other suggestions?
Is 88 minutes a “world class” time? It met customer expectations for last night (although I'm sure the ice WAS horrible).
Here is an article about the Pepsi Center (Denver) process for changeover. They always play hockey first when they have a “double header” out of concern for the quality of the ice. I guess Denver is more of a hockey town and Dallas is more of a basketball town (ignoring football, since it's April).
They have some standard work methods for putting the cover over their ice:
As they bring out the pieces of the cover, it looks like the conversion crew is working on a giant jigsaw puzzle. The crew, however, has a cheat sheet of sorts. The rounded pieces that cover the outer perimeter of the ice are numbered and stored in order. Once those are laid down, the rest of the pieces of the cover are the same size.