I was watching a documentary on Discovery the other day about future cities. One of the things mentioned was a smart building that would sense your presence as you enter and start your computer to avoid time wasted on waiting for it to boot. Cool idea. But that’s where it stops. My question is: Why can’t we design a computer that starts instantly?
This takes me to another one of my pet peeves – complex solutions for even simple problems. When I started as an engineer I wasn’t immune to this. Things changed when I got my hands on “In Pursuit of Elegance”. A whole world of simple solutions opened up since I read the book. Somehow we have come to believe that complex problems need complex solutions. Half of this comes from not focusing on the real reason why the problem exits. With more tech savvy workforce this will be a huge challenge for a manager. These days, everyone seems to have a database fetish. Create a database and run another report to put more band-aids on the problem. One of the teams that I have been involved with created a system that manages over 100 SKU/1.2 million units per annum with a simple card system that would beat all the ERP systems. If I want to change the system all we need is a day. What if I’m hooked up to a complex ERP system? Ever tried pulling teeth out of someone. This is close.
I have come across a lot of situations where the engineers would come up with PLC controlled automated solutions to be trumped by a simple system that uses siphon and gravity. Think of the ‘dabbawallahs’ of Mumbai. They’ll beat any sophisticated logistics company with their effectiveness. All they need to run the system are color chalks.
This again goes back to “Respect for people’s skills”. They are more valuable than any machine.
These days I have established a rule for all my kaizen teams: If you create a system that is held hostage to a couple of experts then it’s not a solution at all.
Coming back to the equation. That is what I call as elegance -- Very simple but powerful. Even the equation for a circle is more complex than this.