First time here? Check out the Help page!
1 | initial version |
Aaron (as usual) has done a masterful job of expanding on the topic :) As is usual for this sort of challenge, the modeler knows "enough to be dangerous" and probably enough to model a typical scenario, but that's not enough for the "real" situation. For example:
For a production facility, these things probably have a high degree of variance, so the goal of the model is to understand the general story, tinker with some key variables to assess how important they are, then figure out how to best represent results in a manner that provides useful understanding to the client. In other words, ie, "all models are wrong, but some are useful." Try to make is as useful as possible, but don't get bound up with exactness / perfection.
2 | No.2 Revision |
Aaron (as usual) has done a masterful job of expanding on the topic :) As is usual for this sort of challenge, the modeler knows "enough to be dangerous" and probably enough to model a typical scenario, but that's not enough for the "real" situation. For example:
For a production facility, these things probably have a high degree of variance, so the goal of the model is to understand the general story, tinker with some key variables to assess how important they are, then figure out how to best represent results in a manner that provides useful understanding to the client. In other words, ie, "all models are wrong, but some are useful." Try to make is as useful as possible, but don't get bound up with exactness / perfection.