Advance Multi-Priority, Multi-Appointment Patient Scheduling With Dependent Demand and Lead Times.

Area of Science:

• Operations Research
• Healthcare Management
• Health Informatics

Background:

• Existing healthcare scheduling models often overlook appointment dependencies and lead times.
• These factors are critical for quality of care, especially when specialist assessments rely on timely diagnostic tests.

Purpose of the Study:

• To develop a dynamic scheduling model that incorporates appointment dependencies, lead times, and patient heterogeneity.
• To minimize patient wait times and ensure diagnostic results precede consult appointments.

Main Methods:

• Utilized Approximate Dynamic Programming (ADP) to create an Approximate Optimal Policy (AOP).
• Derived a closed-form solution for optimal approximation parameters, offering computational advantages.
• Evaluated AOP performance via simulation against benchmark policies and current clinical practices.

Main Results:

• The AOP consistently outperformed benchmarks in reducing wait times and ensuring timely diagnostic completion.
• A practical heuristic, MSP, derived from AOP, showed comparable performance with high interpretability.
• The approach was validated using a case study from a Stroke Prevention Clinic (SPC).

Conclusions:

• The proposed dynamic scheduling model effectively addresses complexities in healthcare scheduling.
• The AOP and MSP offer efficient solutions for optimizing patient flow and improving care quality.
• The model has broad applicability across various clinical settings with interdependent appointments.