The specific research questions to be
answered are, (i) does classification and sequencing of patients improve
the performance of an appointment system?, (ii) do the wellknown
scheduling rules that have been found to perform well in
earlier studies, also perform well in the situations considered in
the present paper?, (iii) if classification of patients improve the
performance, do simple classification significantly improve the
performance or more sophisticated statistical techniques for classification
further improve the performance? Considering the complexities
involved in the patient flows due to multiple classes of
patients and non-homogeneity in punctuality, no-shows and service
times, discrete-event simulation is used to model the patient
flows and evaluate the performance of various appointment systems.