Perform experiment with factor variation and optimization
In an experiment with factor variation, you can vary the attribute values for the objects included in the simulation. Depending on the configuration, many scenarios are created. You can specify an optimizing function to find the optimal configuration of processes and resources without having to simulate all kinds of possible scenarios.
The optimization becomes active only from 101 configurations.
Procedure
Activate the Evaluate tab bar.
Click Start simulation experiment. The Experiment Wizard opens.
Select the Experiment with factor variation experiment type.
Select the New experiment or Existing experiment option. If you want to simulate an existing experiment, select the relevant simulation experiment by clicking the button next to the File box.
Click Next.
Select the database to be used for the experiment. To browse, click the Browse button next to the field. The Select database dialog opens. Select the relevant database, enter your user data if necessary, and select the database settings.
If you selected Existing experiment in the previous step, the appropriate database is displayed automatically.
You can only include models from the same database in a simulation experiment.
Click Next.
Click Add models, select at least one simulation-relevant model, and click Next.
In the Select factors step, add objects for the factor variation. To vary different attributes for an object, add the object as many times as required by clicking Add factor, and select each required attribute in the Attribute box.
Specify the lower and upper limits (Low/High) for the factors and the increments of the intermediate steps. To do this, click in the corresponding box and enter the required number or time.
Select the number of replications. The default is 1.
Click Next.
In the Select responses step, add the objects and models for which you want to save responses. To save various experiment results for a model or object, add the model or object to the list as many times as required by clicking Add model/Add object, and select the required result for each in the Result box. A particular combination of object and attribute can be used only once. This means that an object can be used multiple times only if it has differing attributes. If an object was added with all attributes it is no longer available on the list of factors.
Click Next.
Select whether an optimization is to be performed and, if required, if the target function is to be maximized or minimized.
Specify the weighting of individual results in the target function. Enter a corresponding positive or negative floating point number. If a result has no influence on the target function enter zero or no value.
Specify the relevant settings and click Finish.
The models are loaded and the simulation experiment is started. Depending on how the output options are configured, the output file created is displayed automatically. The run, configuration, and replication numbers are displayed in separate rows for each simulation run. The factor and result values used, as well as the value of the target function (optimization) are also shown. By comparing target values, you can identify the best configuration for your optimization target and optimize your process using this data.