In many complex processing systems with limited resources, fast response times are demanded, but are seldom delivered. This is an especially serious problem in healthcare systems providing critical patient care. In this paper, we develop a multiclass Markovian queueing network model of patient flow in the Accident and Emergency department of a major London hospital. Using real patient timing data to help parameterise the model, we solve for moments and probability density functions of patient response time using discrete event simulation. We experiment with different patient handling priority schemes and compare the resulting response time moments and densities with real data.
pubs.doc.ic.ac.uk: built & maintained by Ashok Argent-Katwala.