Comparative study of holt-winters, lstm and prophetin forecasting hospital outpatient visits flow

Effective hospital outpatient visits flow forecasting is an important task for modern hospitals to implement intelligent management of medical resources. Since outpatient visits flow may be nonlinear and dynamic, we investigate a comparative study of three methods: Holt-Winters, LSTM and Prophetto find the most suitable model for forecasting hospital outpatient visits flow. Holt-Winters is a classical statistics model, LSTM is a deep learning model, and Prophet is a new hybrid model. The comparative experiments were conducted on the two datasets: the outpatient visits flow at General Hospital of Cu Chi Area (in Ho Chi Minh City, Viet Nam) and the outpatient visit flow at Ho Chi Minh City Hospital of Dermato-Venereology. The experimental results demonstrate that Holt-Winters is more effective than LSTM and Prophet model in this forecasting problem.