Research on Hospital Operation Index Prediction Method Based on PSO-Holt-Winters Model
暂无分享,去创建一个
The prediction1 of hospital operation indicators is of great significance and can provide an important basis for hospital operation and management, so as to assist managers to make decisions such as resource allocation and task planning. In order to solve this problem, a novel Holt-Winters model based on particle swarm optimization (PSO) is proposed, aiming at the accurate prediction of hospital operating indicators. In the process of model construction, according to the characteristics of time series data of hospital operation indicators, a time decay mean square error function is constructed as an optimization function of particle swarm optimization algorithm, which enables particle swarm optimization algorithm to better fit recent historical data and grasp the characteristics of recent time series, so as to improve the prediction accuracy. An example is given to analyze the hospital operation index data of a third-class hospital from 2014 to 2017. By initializing the parameters of the model and optimizing the parameters, the improved PSO-Holt-Winters model of TDMSE-1 is established, which can accurately predict the outpatient, inpatient, emergency, discharged and surgical cases.
[1] Arash Ghanbari,et al. Developing a hybrid artificial intelligence model for outpatient visits forecasting in hospitals , 2012, Appl. Soft Comput..
[2] Yong Li,et al. Predictive Analysis of Outpatient Volumes of a First-class Grade A General Hospital through ARIMA Models , 2014 .
[3] C. Chatfield,et al. Prediction intervals for the Holt-Winters forecasting procedure , 1990 .
[4] Xuhua Yang,et al. 基于深度信念网络的医院门诊量预测 (Forecasting of Hospital Outpatient Based on Deep Belief Network) , 2016, 计算机科学.