Online Appointment Scheduling of Patients in a Resource Constrained Facility

An online scheduling problem for a scanning facility in a hospital is described. The facility operates in a resource constrained and high-demand scenario in which machine idleness is undesirable. Other practical constraints dictate that several patients are scheduled in fixed time slots. The objective is to balance the competing goals of minimizing patient waiting time and negligible idle time of the machine. Based on the data available from the facility, patients are divided into three types, and two different policies are proposed for scheduling: one fills all slots one by one, while the other uses the best-fit variant of online bin-packing. To account for variability in the scan times, some slots are overbooked. The proposed appointment policies are simulated and compared. Simulations are based on data received from the MRI department of a cancer-care hospital. The second policy trades off between lesser waiting time for patients and machine idleness.