I've enjoyed this discussion and would like to add two points.
First, before I was became an MBA/Management Ph.D, I was an R&D engineer for
Boeing working primarily on NASA programs and (among other things) built
computer programs to simulate the performance of a particular aerospace
vehicle on a particular mission. Prior to the mission, we used the
simulation to vary system parameters to optimize the performance of the
simulated vehicle. During a mission, we received telemetry data from the
actual mission and checked telemetry against the simulation. One of the
most important lessons that I learned about simulations occurred when a
mission failed badly (at the cost of a billion dollars to the U.S.
government). For the next six months to a year, we worked full time to
figure out what was wrong with the simulation and to use that information to
correct the simulation. (After we made corrections, we could match the
telemetry data exactly.) The corrected simulation worked much better in
future missions.
Personally, I don't think that management simulations can reach this level
of accuracy for reasons that people have pointed out. That said, I think
that it is important for teachers to talk to students about the purpose,
assumptions, and limitations of the simulations and to try to debrief them
after the exercise. Students can learn a lot from simulations. But they
shouldn't be lulled into thinking that they've learned everything or not
being alert for "second-order effects" that are not captured by the
simulation.
The second observation is that I think that simulations can be useful for
looking at "cross functional" as well as "cross cultural" situations. (It
is not that hard to program the behavior of a stereotypical accountant and a
stereotypical engineer in a particular situation. (-: ).