Please enable JavaScript to view this site.

Vensim Help

You can set up and run sensitivity simulations using the Sensitivity Control dialog as described above.  You can also enter the control parameters directly into a file.  The format for this file is:

#simulations , method={U|M|L|G|F} , seed , filename , warnings={0|1}



#simulations  is a number specifying how many simulations to perform.  If this number is not included, 200 simulations will be performed. This option is ignored when the F (File) method is active.

method is a character indicating the method to use; Univariate, Multivariate, Latin Hypercube, Latin Grid, or File

seed is the random noise seed to use. Normally this should be an integer between 1 and 231 (about 2E9), or 1 and 223 in single precision. The default is 1234. If seed is negative, the legacy LCG random number generator will be used. See NOISE RNG. If seed is 0, the seed will be nonreproducibly randomized (useful if you want to perform distinct samples on multiple computers without editing the control file for each). This option is ignored when the F (File) method is active.

filename is used with the F method option to specify a file containing the sample to be evaluated (see below).

warnings can be set to 0 or 1 to control the display of runtime warnings.





This would make 250 multivariate simulations, with warnings off, randomly changing both PRICE OF STEEL and BUILDING TIME.

File option

The F (File) method requires a file containing the sample to be evaluated. This file must be a tab delimited text file (UTF8 for international characters). The first row should contain a list of model constant names (tab delimited), with the corresponding values on each successive row.  If a value is missing from this file the previously used value is repeated.


Filename: "myScenarios.txt"

price of steel<tab>building time<tab>NOISE SEED<return>




This would generate three simulations with the given values for price of steel and building time, and varying the dynamic noise realizations of and RANDOM functions in the model through NOISE SEED.