Scheduling and Optimization of Appointment System in Hospital using Machine Learning

Abstract

            One of the biggest challenge hospitals facing is efficient patient queue management to reduce patient waiting time and patient overcrowding. Long-term waiting contributes to considerable waste of human resources and time, and raises patient dissatisfaction. The cumulative waiting time of all the patients is the time he has to wait for each patient in the queue. If patients could access the most appropriate care plan and know the approximate waiting period, it would be simple and attractive that will be updated to them dynamically. The Patient Time Prediction (PTTP) algorithm is proposed to predict a patient's wait time for each analysis phase. For each prophecy, practical patient data from various hospitals are acclimatized and a patient-treatment time model is proposed. The duration of treatment for each patient is estimated in the prevailing queue. A Hospital Queuing-Recommendation (HQR) framework is built on the basis of the expected waiting time. A patient's prescribed effective and easy-care plan is estimated and predicted by HQR. HQR model is built using JAVA frames and random forest algorithm is implemented. Experimental results show that there is optimal reduction in waiting time of the patient for the given schedule of treatment plans.