부산대학교 도서관

The majority of medical facilities base their assessment of how congested their emergency rooms (ERs) are on the length of time patients must wait in queues. Exorbitant wait times in many ER departments make it harder to serve patients effectively and drive up overall costs. In system queuing applications, contemporary techniques including deep learning (DL) or machine learning (ML) have been frequently employed. Additionally, the flawless collaboration and interaction among healthcare practitioners is made possible by the pairing of the a forementioned prediction technology into a hospital's electronic health records (EHR), and scheduling systems, leading to a more efficient and focused on patients' approach to emergency care. This work creates an empirical research of predicting patients' wait times with multiple approaches and achieves the best operating model to more effectively prioritise patients within the queue. Utilising genuine ER data further contributes to the value of this effort. In addition, the proposed approaches anticipate individuals' wait times more precisely instead of a conventional mathematical strategy. The proposed method can be easily adapted by the healthcare industry's queue system by integrating information from electronic health records (EHRs).