Solving the dynamic ambulance relocation and dispatching problem using approximate dynamic programming.

Area of Science:

• Operations Research
• Emergency Medical Services Optimization

Background:

• Effective ambulance dispatch and relocation are critical for timely emergency patient care.
• Real-time decision-making is essential for both initial dispatch and post-service vehicle repositioning.

Purpose of the Study:

• To propose and evaluate a model for optimizing ambulance deployment using approximate dynamic programming (ADP).
• To demonstrate the impact of ADP on reducing emergency response times compared to traditional methods.

Main Methods:

• Development of an optimization model leveraging approximate dynamic programming (ADP).
• Empirical testing using real-world data from emergency service providers in Vienna.
• Comparison of ADP-based dispatching with classical dispatching rules.

Main Results:

• The ADP model reduced average response times from 4.60 to 4.01 minutes (a 12.89% improvement).
• Explicit consideration of time-dependent factors like travel times and request volumes is crucial for optimal performance.
• Deviating from classical dispatching rules significantly enhances efficiency.

Conclusions:

• Approximate dynamic programming offers a superior approach to ambulance dispatch and relocation compared to conventional methods.
• Incorporating time-dependent data in optimization models is vital for minimizing emergency response times.
• The proposed ADP model enhances the efficiency of emergency medical services.