This page demonstrates a key Lean concept: Mura, or variation.
The simulator is based on a game described in Goldratt's
The Goal
(which is well worth reading). Click the help button to the right of
the spinner to get an explanation of how the simulation works.
Very nice, Allen. You always do good work. The Goal is an excellent book and this illustrates your point about work station averages nicely.
I am not convinced that you can say this represents 5 back to back sprints, though. Subsequent sprint velocity is not constrained by prior sprint performance since stories are not passed ‘down the line’ as in a production process.
I wonder if you will be doing a bottleneck simulator to demonstrate the effectiveness of WIP limits?
Bernie — The sprint issue depends on whether or not the stories are dependent (i.e. you can’t release until you’ve completed a whole set of them). When they’re fully independent (as they should be if you’re following the INVEST criteria), normal averages kick in. When you have to complete a story before you can do the next one, then we have the dependent-variable situation demonstrated in the simulation.
I’ve been pondering the notion of a WIP simulator, but I’m not sure what it would look like (other than a simulation of the coin-flip game). In the class, I was just going to get some friends to do the coin-flip exercise and record them. If anybody has a better idea, post it here!
Thanks alot – your answer solved all my problems after several days strnggliug
A project management trainer once sat my group down, gave us each 1 six-sided die, and had us roll the die to inject variability at each step of our simulated projects. I pocketed my die and keep it with me where ever I work. If asked about it I say that a project manager gave it to me as a schedule prediction tool. We used to have a joke about dice and estimation: “The first roll selects the number, the second roll selects the unit of time.”