An intelligent real-time scheduler for out-patient clinics: A multi-agent system model

Area of Science:

• Operations Research
• Healthcare Management
• Computer Science

Background:

• Outpatient clinic scheduling is often a long-term, push-based system relying on average demand.
• Uncertainty and variability in patient demand lead to prolonged waiting times and resource under-utilization.
• Current systems struggle to adapt to real-time fluctuations in patient arrivals and resource availability.

Purpose of the Study:

• To develop and evaluate an intelligent real-time scheduler for outpatient clinics.
• To improve clinic efficiency by dynamically matching patient flow with resource availability.
• To transform outpatient clinics from open-loop to closed-loop systems.

Main Methods:

• Modeling outpatient clinics as a multi-agent system.
• Implementing a real-time scheduler with predictive resource allocation and path-optimized patient scheduling.
• Utilizing an auction-bidding coordination mechanism for real-time resource rescheduling.
• Conducting simulation studies followed by real-world implementation at Aravind Eye Hospital.

Main Results:

• The intelligent real-time scheduler significantly reduced patient waiting times and cycle times.
• Resource utilization was substantially improved compared to isolated scheduling methods.
• The system effectively matched resources with stochastic patient demand in real time.
• The scheduler successfully converted the clinic system into a pull system and a closed-loop operation.

Conclusions:

• Intelligent real-time scheduling is effective in enhancing outpatient clinic performance.
• Dynamic rescheduling based on actual system status optimizes patient flow and resource allocation.
• This approach offers a significant improvement over traditional push-based scheduling systems in healthcare settings.