Stochastic programming for outpatient scheduling with flexible inpatient exam accommodation

Area of Science:

• Operations Research
• Healthcare Management
• Applied Mathematics

Background:

• Hospital appointment scheduling is complex due to variable procedure times and patient arrivals.
• Inpatient exams can be cancelled, unlike outpatient exams, adding scheduling complexity.
• Existing scheduling methods may not fully account for stochasticity and optimization.

Purpose of the Study:

• To determine optimal appointment schedules for outpatient-inpatient hospital systems.
• To minimize costs associated with patient waiting times, machine idle times, and exam cancellations.
• To evaluate the impact of optimizing slot size and inpatient block placement.

Main Methods:

• Formulation and solution of two-stage stochastic programming models.
• Development of a computational method for simultaneous optimization of inpatient block placement and slot size.
• Simulation using FlexSim Healthcare for evaluating schedule performance.
• Analysis of commonly used sequencing rules and development of a Pareto frontier.

Main Results:

• Optimizing slot size reduced overall weighted costs by 11.6%.
• Simultaneous optimization of slot size and inpatient block placement further reduced costs by an additional 12.6%.
• An ALTER variant sequencing rule, distributing inpatient blocks evenly, performed best in the absence of optimization tools.

Conclusions:

• Stochastic programming models effectively optimize hospital appointment scheduling.
• Simultaneous optimization of slot size and inpatient block placement yields significant cost reductions.
• The ALTER variant sequencing rule offers a robust alternative when optimization tools are unavailable.