Simulation and Machine Learning Based Real-Time Delay Prediction for Complex Queuing Systems

Real-time delay predictionReal-time Delay Prediction involves providing entities arriving at a queue with an estimate of their wait time at the time of their arrival at the queue. This wait-time estimate is generated as a function of the state of the system at the time of the entity’s arrival to the queue. In this chapter, we present a hybrid simulationHybrid simulation and machine learningMachine learning-based approach toward real-time delay predictionReal-time Delay Prediction for complex queuing systems. The approach involves usingDiscrete-Event Simulation a discrete-event simulationSimulation of the queuing system under consideration to generate system state data that is in turn used to train machine learningMachine learning methods that generate real-time delay predictionsReal-time Delay Prediction. We provide a case study illustrating this approach that involves generating real-time delay predictionsReal-time Delay Prediction for end-stage renal disease patients registering on kidney transplantationKidney transplantation waitlists.