Predictive Analytics on Scheduled Surgery Cases in Forecasting the Operating Theatre Utilisation

The scheduled surgery cases, allocation of clinical provider, machine, equipment, preparation time, surgery performance, and patient recovery give the big impact on OT utilization. Low OT utilization due to no show patient and scheduling bottleneck interrupt patient flow in clinical process. It also decreases the admission to the OT and wastage of resources. In order to improve the capacity of OT, the logical solution need to be carried out is utilization audit. The trend of scheduled surgery cases has identified, and element affect the OT capacity have used to predict the OT optimization for future planning. The purpose of this study is to investigate efficiency of operating theatre (OT) utilization in healthcare institution and the application predictive analytics on its daily operational data. OT contribute to the revenue for the hospital and workload.  This project use machine learning in identify which model can use to predict the decision of admission to which facility after the surgery. The model has been going through a few mathematical reasoning to getting the usage efficiency on the dataset. The result shows that Support Vector Machine (SVM) got the highest accuracy in test data rather than Logistic Regression (LR) and Random Forest Classifier. SVM used to predict the admission decision, which contribute to the surgery scheduling in OT. This project can be extended to the admission decision with the factor or severity of patient condition to undergo any intervention in outpatient.