Optimizing stochastic models

If your model is stochastic, the results of the single model run may not be representative. The OptQuest Engine supports objective values that are based on experimentation through the General Replication Algorithm. This feature allows the user to provide the OptQuest Engine with results of multiple replications per simulation/iteration. The OptQuest Engine allows you to run a fixed number of replications per simulation or a varying number of replications per simulation.

When running a varying number of replications, you will specify minimum and maximum number of replications to be run. The OptQuest Engine will always run the minimum number of replications for a solution. OptQuest then determines if more replications are needed. The OptQuest Engines stops evaluating a solution when one of the following occurs:


Optimization experiment's properties view. Replications section

 To schedule fixed number of replications
  1. In the Projects view, select the optimization experiment.
  2. In the Replications section of the Properties view, select the Use Replications check box. 
  3. Choose the Fixed number of replications option.
  4. Specify the number of Replications per iteration in the edit box.
 To schedule varying number of replications
  1. In the Projects view, select the optimization experiment.
  2. In the Replications section of the Properties view, select the Use Replications check box. 
  3. Choose the Varying number of replications (Stop replications after minimum replications, when confidence level is reached) option.
  4. Specify the minimum number of replications in the Minimum replications edit box.
  5. Specify the maximum number of replications in the Maximum replications edit box.
  6. Define the Confidence level to be evaluated for the objective.
  7. Specify the percent of objective for which the confidence level is determined in the Error percent field.

Related topics

How-To video: Optimization experiment