Predicting ambulance offload delay using a hybrid decision tree model rights and content


Offload delay is common problem at the interface of the ambulance service and emergency department.

Predicting this delay is essential to be proactive in mitigating it.

We present a framework and case study that uses a Naive Bayes classifier to remove the noisy training observations before decision tree induction.


Ambulance offload delay (AOD) is a growing health care concern in Canada. It refers to the delay in transferring an ambulance patient to a hospital emergency department (ED) due to ED congestion. It can negatively affect the ability of the ambulance service to respond to future calls and reduce the efficiency of the system when the delay is significant. Using integrated historical data from a partnering hospital and an Emergency Medical Services (EMS) provider, we developed a decision-support tool using a hybrid decision tree model to predict the severity of AOD occurring within 1–5 h in an EMS system. The primary objective of this study is to provide an AOD prediction model based on the current system status, hour of the day, and day of the week. With this information, decision-makers can be proactive with efforts to mitigate AOD. Various prediction models are developed with different focuses and forecast periods. This research demonstrates a novel hybrid decision tree method applied with administrative data in a health care setting. A naïve Bayes classifier is first used to remove noisy training observations before decision tree induction. This hybrid decision tree algorithm was tested against the basic classification and regression tree (CART) algorithm, using classification accuracy, precision, sensitivity, and specificity analysis. The results indicate that the hybrid algorithm shows improvements in performance in the classification of the real-world problem. It is anticipated that the prediction model for AOD produced from this study will be directly transferable. It can be generalized to other EMS systems, where predicting AOD is important for efficient operations.


Hybrid decision trees
Ambulance offload delay
Classification and regression tree
Decision support
Machine learning

Mengyu Li is an Instructional Assistant Professor in the Department of Industrial and Systems Engineering at the University of Florida. She teaches numerous undergraduate and graduate courses, including: Supply Chain Management, Senior Design Project, Systems Design, and Systems Architecture. Dr. Li received her Ph.D. in Industrial Engineering from Dalhousie University. Her research focuses on designing processes to improve ambulance deployment and emergency resource allocations. In 2018, her research received the Top EMS Research Project Award in Nova Scotia, Canada, and continues to be used to reduce wait times by the provincial health authority.

Peter T. Vanberkel's primary research involves improving healthcare operations using stochastic operational research methods. He is a Professor in the department of Industrial Engineering at Dalhousie University and a Staff Scientist at the IWK Health Centre.

Xiang Zhong received her B.S. from the Department of Automation, Tsinghua University, Beijing, China, in 2011, and her M.S. in Statistics and Ph.D. in Industrial Engineering from the University of Wisconsin-Madison in 2014 and 2016. Currently, she is an Assistant Professor of the Department of Industrial and Systems Engineering at the University of Florida. Her research interests include stochastic modeling and control, and data analytics with the application in healthcare, service and production systems. She is a member of IEEE, IISE and INFORMS.

View full text