question

The key to this is to add a parameter as an Option type and then put text for the Empirical distribution name in the edit box BELOW the values table view - if you try to edit the table cell it will not currently accept it and will revert to numeric.

Then there are a number of ways to use this model parameter of which one is to use it to set a value somwhere in the model and another is to read the parameter directly from somewhere in the model. Both will work with scenarios in the experimenter, and with the parameters dashboard.

So for this example I created a ColorPalette and some Empirical Distributions so that we might use those in the example to change interarrival time and to set the color of the arriving items by choosing the experiment scenario.

InterArrival Distributions:

Color Distributions and Palette:

Setting Labels

The first approach puts labels on the source for us to reference in the arrival type and creation triggers:

So then the inter-arrivaltime is set to: Empirical(current.intArrivalEmp).get(getstream(current))

And the creation trigger sets the items color to:

Color.fromPalette(Empirical(current.colorEmp).get(getstream(current)),"ColorPalette1")

Note that this will work without the Model Parameters table – as it’s just reading the distribution names from the label.

To get the Model Parameters to set this via the Experimenter and on Reset. We need to add them as Options. Open the model parameters tool found in the toolbox:

Click on the arrow to the right of the value cell on the new parameter and select the Option type form the pull down which should then show this:

Now for the interarrival distribution names we need to add two options to press the + sign against Options. Click on each new value cell and enter the distibutions name BELOW the table in the edit box. Then for the reference use the sample to click on the source and choose the label “intArrivalEmp” we created (only if setting labels -see below ‘Direct Parameter Acces’s). Then rename the rows to tell FlexSim what you want shown in the experimenter view – I’ve used the same names as the distribution:

You can then rename the parameter so that you know what they are in the experimenter view:

Next set up the experimenter scnarios by checking the pararmeter you just added and adding scenarios.

Note: At the moment you need to type in the row number of the option value you want – so 1 or 2 in this case – and the value for that option will be displayed as shown.

By selecting the column you can then set the current scenario for the model when running interactively. Resetting the model will set the label value and so the source will use FastArrivals or SlowArrivals depending on the scenario chosen.

The same process goes for the empColorSetter parameter which sets the colorEmp label on the source to “RGBColors” or “ManyColors”

Direct Parameter Access

A second approach is to use the model parameters directly. In this approach the labels are not used and instead of entering the distribution names we for the parameter’s option values we enter the Empirical sampling expression which we will then evaluate from the source.

So instead of “FastArrivals” and “SlowArrivals” we’ll have Empirical("SlowArrivals").get(param(1)) and Empirical("FastArrivals").get(param(1))

There’s no need to sample or add code to the OnSet so it can be removed. I added this as a separate parameter in the model so you could see both methods in parallel and named it EmpiricalInterArrivalSampler along with the same technique for color sampler:

Then in the source inter-arrival time field the expression becomes:

Model.parameters["EmpiricalInterArrivalSampler"].evaluate(getstream(current))

And the expression for setting the item color is:

Color.fromPalette(Model.parameters["EmpiricalColorSampler"].evaluate(0),"ColorPalette1")

Model with both approaches attached so the Scenarios look like this:

Model attached in the next comment.