Click on the following link to read a full description of the complete Automated Simulation Output Analyser (AutoSimOA) built during this 3 year research project:  AutoSimOA - The finished analyser.   This paper includes full descriptions of each component of the analyser: Warm-up analyser, Replications Calculator and Single run analyser.  It provides full details of the heuristics created around each algorithm and the overall framework of AutoSimOA.

 

Click here to view a flow diagram of how we interpret the decision process involved in output analysis.  AutoSimOA tackes the ‘HOW’ part of this diagram.  The ‘WHAT’ part is currently left up to the user.   In order to create a completely automated system the decisions and interpretations captured in the ‘WHAT’ part of the diagram would need to be fully automated.  This would undoubtedly prove to be a challenging and complex task.  Currently, cyclic data cannot be input directly into AutoSimOA.  The cycle would need to be taken out of the data by averaging over the cyclic periods.  Also, the Warm-up Calculator cannot currently detect the initial bias in transient data.  Warm-up methods assume steady-state and the question of initial bias in transient data has not yet been adequately addressed in the research literature.  This work has concentrated on estimating the mean and variance of output data.  There are however other measures of performance that may be of interest to the user e.g. proportions, quantiles…   It would be useful to extend AutoSimOA to provide decisions for these measures.