Time Series Neural Network-based Analysis of Hospital Material Prediction under Infectious Disease Epidemic

Abstract

In order to overcome the inaccuracy of traditional time series and linear models for predicting short-time hospital supplies under infectious disease epidemics, this paper combines time series and neural network models, establishes a neural network-based time series prediction model, preprocesses hospital supplies data, and uses the model to train the data for prediction, and uses a tertiary class A sentinel hospital in Guizhou province for 24 days in February 2020 for medical. The study shows that the absolute value of error is mostly less than 0.2, the mean absolute error is less than 0.001, and the root means the square error is less than 0.1, indicating that the time series neural network model has high prediction accuracy.

A study on modified SEIR model considering quarantine status and the prediction of material demand during epidemics

“…Wang et al, with the consideration of the delay caused by the latent period of an epidemic, constructed a multi-objective stochastic programming model with time-varying demand for the emergency logistics network based on the epidemic diffusion rule 10 . Wang et al combined time series and neural network models, established a neural network-based time series prediction model for hospital material under infectious disease epidemics 11 . Pan et al formulated a demand forecasting model with a general demand forecasting function based on the last-period demands, extra demand caused by the last-period unfulfilled demand, and uncertain demand 12 .…”