Improving the efficiency of the operating room environment with an optimization and machine learning model.

Area of Science:

• Healthcare Management
• Operations Research
• Machine Learning

Background:

• Operating rooms (ORs) are critical financial centers for hospitals, necessitating efficient management.
• Post-anesthesia care unit (PACU) capacity often acts as a bottleneck, causing OR delays and procedure cancellations.
• Inefficient scheduling leads to patient waiting times in ORs and impacts downstream surgical workflows.

Purpose of the Study:

• To develop a generalizable method for sequencing OR procedures to minimize delays.
• To reduce patient wait times caused by PACU unavailability.
• To enhance overall hospital operational efficiency and patient throughput.

Main Methods:

• Machine learning was used to predict the PACU time required for different surgical procedures.
• Two integer programming models were developed to schedule OR procedures, minimizing peak PACU occupancy.
• Discrete event simulation was employed to compare the optimized schedule against existing practices.

Main Results:

• The proposed scheduling system significantly reduced OR delays attributed to PACU congestion.
• Simulations indicated a potential 76% reduction in total PACU holds during the second half of 2016.
• High operating room utilization was maintained, demonstrating the system's efficiency.

Conclusions:

• The developed optimization and machine learning approach effectively minimizes OR delays caused by PACU bottlenecks.
• This scheduling system offers a practical solution for improving hospital resource management.
• Implementation of this system can lead to substantial improvements in patient flow and operational efficiency.