Optimizing Spinal Surgery OR Scheduling: Using Machine Learning Models for Predicting Operative Time and Improving Schedule Design

Introduction: The OR schedules are primarily based on estimations of operative times by the surgical team based on experience, without considering several important factors related to the patient demographics or surgeon characteristics. Estimation of operative time with low accuracy and precision leads to suboptimal scheduling of available operating rooms. We aimed to develop a predictive ML model for operative time using our institutional data, to improve the scheduling of neurosurgical ORs to minimize waste and maximize resource utilization.

Methods: We developed a predictive model using our institutional which included diverse neurosurgical cases across three tertiary hospitals. We applied multiple ML techniques, including Linear Regression, Support Vector Regression, Deep Neural Networks, and Extreme Gradient Boosting Regression. The model was further optimized using a First Fit Decreasing (FFD) bin-packing algorithm for practical OR scheduling, considering operational constraints such as the duration and type of procedures.

Results: he Extreme Gradient Boosting Regression model exhibited the best performance, with lower mean absolute errors (RMSE: 52.24, MAE: 37.25) and higher R2 (0.76) values compared to other models. Implementing this model in a simulated OR scheduling scenario using institutional data, we successfully scheduled 30 random procedures within minimal residual time, such that the model ran out of procedures for the week with 86 and 171 minutes remaining in the cranial and spinal ORs, respectively, demonstrating potential efficiency gains. The cranial procedures were scheduled over four days and spinal procedures over six days, optimizing the use of available OR time.

Conclusion : The utilization of multiple ML models in a strategic combination for predicting operative times with OR scheduling algorithms can significantly improve the accuracy and efficiency of surgical schedules. This approach minimizes wasted OR time and enhances resource allocation, potentially improving patient outcomes and satisfaction by reducing wait times.