Comments on _Critical Chain_, ch 6

The scene is the classroom where the professor and students meet again to discuss their project management reports where they interviewed the people involved looking for official reasons and unofficial reasons for budget/time overruns.

So the first project management report identifies that the official reasons are all external factors (outside the company) and the unofficial reasons are all internal factors (inside the company). 

So here are the official reasons:

“Even better," I say, and continue, "One: Particularly bad weather conditions that delayed construction. Two: Unforeseeable difficulties experienced by the vendors who supply the machines. Three: Longer than expected negotiations with the Malaysian government concerning employment terms."

And here are the unofficial reasons:

“Unofficial reasons given by the project leader," I read. "One: Corporate forced an unrealistic schedule to start with. Two: It was dictated that we choose the cheaper vendors, even though it was known that they are less reliable. Three: In spite of repeated warnings, efforts to recruit and train plant personnel and workers started too late."

Interesting. Each of these unofficial reasons is connected to one of the official reasons. 

Official reason #1 was caused by unofficial reason #1. Bad weather delayed construction because corporate forced unrealistic schedule [that didn't properly account for expected bad weather].

Official reason #2 was caused by unofficial reason #2. Vendors didn't supply machines in time because corporate forced us to use vendors that we thought weren't good enough to supply machines in time.

I'm not clear on the relationship for the #3 reasons.

... "There is a pattern here. The lower the level of the person, the more the finger points internally, rather than externally. You'll find the same thing in my report."

The discussion turns to the idea that not accounting for uncertainty is the major problem. Then somebody says...

“So there's nothing anybody can do," is Charlie's conclusion. "You cannot force certainty on a situation that contains major uncertainties."

That's like saying that we can't learn about nature cuz we might be wrong, or we can't fix mistakes because we can't be sure that any particular mistake was fixed, or we can't do things to help ensure getting the right answer on a math problem because there's always the possibility that we get something wrong despite our best efforts. I'm reminded of numerical methods class in college; a big idea was that we can't remove all error but we can do things to reduce error.

So like say I regularly go to work at 9am and I know it takes 30 minutes to drive there. Lets say I don't want to be late. Let's say I know that sometimes there'll be a train that takes up upto 10 minutes, let's say 10% of the time. So I could leave my house 40 minutes before the meeting such that if I get stuck from the train, I'm still not late to work. So like it's uncertain on any given day whether or not I'll lose 10 minutes due to a train. But I'm reasonably certain that if I factor in 10 minutes for the train situation, I'll never be late to work due to losing 10 minutes from the train.

The rest of the chapter presents the problem of figuring out the amount of safety (time accounting for uncertainty) in the steps of a project and working out how that should be factored in to create a whole project estimate.