Selective Parametric Analysis

Revision as of 20:04, 21 January 2016 by Bbecane (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Needs to be filled in.

General idea:

When you have a lot of inputs that you want to be able to easily perform parametric analysis on, don't run them all at once -- the complexity is multiplicative. Instead select a small subset of inputs.

To make this convenient, define each potential parametric input as a Choice, with the All option. Select a single value for most of your inputs, but for a small set, select All. These are the variables for your parametric analysis.

Make Input for each of your inputs and cluster these on an input panel so they are conveniently accessed.

When using your model, select a small subset, view your results, then to "pivot", you select a new subset from your input panel and recompute.

The main trick in making this work is dealing with edit tables. If a normal edit table is based on an choice index, then its result will still contain that index. The trick is to change your edit tables to DetermTables. Then the DetermTables return only the slice selected.

See Also


You are not allowed to post comments.