SimRunner takes your existing
When you conduct an analysis using SimRunner, you build and run projects. With each project, SimRunner runs sophisticated optimization algorithms on your model to help you optimize multiple factors simultaneously. For each project, you will need to provide SimRunner with a model to analyze or optimize, identify which input factors to change, and define how to measure system performance using an objective function. SimRunner can conduct two types of tests: Pre-Analysis (Statistical Advantage) and Simulation Optimization.
Using SimRunner will help you find accurate solutions for your modeling needs. No longer must you sit and experiment with what you think might work, only to find that your solution actually interferes with productivity in other parts of the model. SimRunner will help you locate true solutions by monitoring how changes affect each part of the model. In other words, SimRunner will not let you improve one area of your model at the expense of another—the results you get are beneficial to the entire model.
The following describes how to start and prepare a project for analysis. Remember that you must create and validate your model prior to analyzing or optimizing it in SimRunner.
How to select a model:
How to define the input factors:
Please Note: SimRunner will optimize only single value macros defined as scenario parameters.
How to define the objective function (output variables):
With the model built, the input factors selected, and an objective function defined, you are ready to conduct a Pre-Analysis. Also known as Statistical Advantage, the Pre-Analysis runs several tests to identify the initial bias (warm-up period), determine appropriate run-length to reach steady-state, find the number of replications necessary to ensure that each event occurs at least once, and locate the model averages.
Simulation Optimization is a multi-variable optimization that tries different combinations of input factors to arrive at the combination that provides the best objective function (output) value. Depending on the number of selected input factors and the complexity of the solution space, this process can take a long or short time. Optimizations with many factors and complex solutions take longer to run.
How to run an optimization:
© 2012 ProModel Corporation • 556 East Technology Avenue • Orem, UT 84097 • Support: 888-776-6633 • www.promodel.com |