Tagging Variables as Results

You can add a variable to the results included in the Results Window by right-clicking on the variable editing area and selecting Results Window Tag as Result. A small yellow medal on the variable input mode icon indicates that the variable has been tagged as result: Input tagged as result Excel input tagged as result Function tagged as result.

Both system- and user-defined variables can be tagged as results, including numeric values, Booleans, strings, and dates.

Variables defined as Excel Input can only be tagged as results if they point to a single-cell range in the Excel worksheet. (See Input Modes.) Tables cannot be tagged as results, nor can arrays that do not include functions.

After the simulation is run, variables tagged as results are listed in the Tagged variables Tagged variables node of the Deterministic Results Deterministic Results.

After running Monte Carlo, the Results Results node of the Monte Carlo results Probabilistic Results shows all those variables that have been both tagged as results from PetroVR Plan and selected as results from the Monte Carlo Control Window with its Units, Mean, Min, Max, SD, etc. values when possible.

Correlation rank results Correlation ranks and Cross Plots results Cross Plots also include variables tagged as results.

Once you have run Monte Carlo, the Data results Data node of the Probabilistic Results show all selected results and the values those variables took for each iteration. Click on the column header to arrange iterations by the values of the selected variable.

If a single-value variable is defined as a function of one or more runtime variables and tagged as result, it will be reflected in the results as an array, showing the deterministic value it took at the end of each period (rather than the mean value for the period). The Plan project must have been simulated at least once for period values to be available.

The values of variables that describe events or states that are unique and are not calculated again after they are first verified (such as Perform Task or Duration of a job, Initial Rate and Initial Reserves), will be frozen from that point - for example, after a job has been completed, its Perform Task value will evaluate to true for the rest of the simulation even if it was defined as a function whose condition ceased to be met at a later point in the simulation. When referenced from other functions, they will always return the frozen value during the rest of the simulation. See Runtime-Dependent Variables.

A Runtime-Dependent Variables that is tied to a nonce event (such as an Abandonment job) and so will yield only two series of values (one before and one after the event) will not be reported as an array but as a single value. That is to say, if it is tagged as result it will show the value it had at the end of the simulation.

Using Tagged Variables for Cluster Identification

One interesting use of the ability to tag variables as results is the possibility to do further analysis on results yielded by Monte Carlo and identify clusters of iterations that share a given result. This can be done in the Data results Data node of the Probabilistic Results by sorting iterations based on which values those variables took in each.

For example, you can create a Boolean variable and divide iterations by whether the Boolean has been true or false in each iteration. Or you can define variables in such a way that they return as result a string subject to certain conditions. After simulating, sorting iterations by that string will allow you to differentiate and group iterations at a simple glance.