Class TimePredictorPlugin


  • public class TimePredictorPlugin
    extends java.lang.Object
    This is a simple time predictor based on a stochastic model. Restrictive assumption: Trace is replayable directly by model to get to the current marking state. (TODO: Leverage this by using a (stochastic) alignment of the observations to the model)

    Working procedure: - Go to current marking by simple replay in the model. - Simulate a number of traces to see end result

    Author:
    Andreas Rogge-Solti
    • Method Summary

      All Methods Instance Methods Concrete Methods 
      Modifier and Type Method Description
      java.lang.Double computeRisk​(org.processmining.contexts.uitopia.UIPluginContext context, StochasticNet model, org.deckfour.xes.model.XTrace observedEvents, java.util.Date currentTime, java.util.Date targetTime)  
      java.lang.Double predict​(org.processmining.contexts.uitopia.UIPluginContext context, StochasticNet model, org.deckfour.xes.model.XTrace observedEvents, java.util.Date currentTime)  
      • Methods inherited from class java.lang.Object

        clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
    • Constructor Detail

      • TimePredictorPlugin

        public TimePredictorPlugin()
    • Method Detail

      • predict

        public java.lang.Double predict​(org.processmining.contexts.uitopia.UIPluginContext context,
                                        StochasticNet model,
                                        org.deckfour.xes.model.XTrace observedEvents,
                                        java.util.Date currentTime)
      • computeRisk

        public java.lang.Double computeRisk​(org.processmining.contexts.uitopia.UIPluginContext context,
                                            StochasticNet model,
                                            org.deckfour.xes.model.XTrace observedEvents,
                                            java.util.Date currentTime,
                                            java.util.Date targetTime)