Sunday, January 15, 2006

On time estimates

The good people at Reynolds Kitchens give us this tip on saving time in the kitchen:
To speed up food prep, chop, shred or grate similar ingredients for recipes at the same time. Wrap with foil and refrigerate. Foil keeps food odors from transferring to other foods.
Strictly speaking, this isn't going to save time at all. It's just going to rearrange time. You still have to chop, shred, or grate. Their argument is really that you don't want to be bothered with the duties of a sous chef when you are busy being a regular chef.

I noticed this when I was getting ready to make a batch of pancakes for the kids. The Bisquick box says that the prep time for pancakes is three minutes, and it takes three minutes to cook up a batch. That prep time makes some pretty generous assumptions:
  • that I know where the pans, measuring cups, and mixing bowls are (they get moved around sometimes)
  • that the last couple eggs in the carton are still unbroken (my son broke a couple recently)
  • that the milk hasn't gone off
  • that the kids get their own orange juice and silverware
...and so on. If you ever do any cooking, you can see that in order for the Bisquick time estimates to be valid, either they assume everything's perfectly in place before they start their timers, or they violate Murphy's inviolable law.

HellboyAt NASA we used to handle time estimates in this tongue-in-cheek manner: whenever someone tells you how long a job will take, you multiply by two, add one, and raise to the next higher unit. So when the Bisquick people tell you prep time is three minutes, you should allot seven hours.

There was a news Web site in Australia that published a mathematical formula for predicting the effects of Murphy's law. The page has expired, but others have salvaged the following:
...a panel of experts has provided the statistical rule for predicting the law of "anything that can go wrong, will go wrong" - or ((U C I) x (10 - S))/20 x A x 1/(1 - sin (F/10)). [...] In the calculation, five factors have to be assessed: urgency (U), complexity (C), importance (I), skill (S) and frequency (F), and each given a score between one and nine. A sixth, aggravation (A), was set at 0.7 by the experts after their poll.
There are two problems with the formula: first, it says nothing about predicting what will go wrong; second, it gives us false hope that we can avoid having anything go wrong. It's better to ignore the predicted prep time and set aside the seven hours. :-)

One more note about pancakes: Hellboy loves them. The day Hellboy (pictured) discovered his love for pancakes was truly Hell's darkest hour. Never mind the prep time. :-)

Update: Guy Kawasaki writes the following about the time estimates of entrepreneurs:
Generally, an entrepreneur has no idea what sales will be, so she guesses: "Too little will make my deal uninteresting; too big, and I'll look hallucinogenic." The result is that everyone's projections are $50 million in year four. As a rule of thumb, when I see a projection, I add one year to delivery time and multiply by 0.1.
If people are going to do this to our estimates anyway, why is it necessary for us to make them? :-)

Labels:

posted by Ron @ 9:35 AM

0 Comments:

Post a Comment

<< Home