论文部分内容阅读
目的应用灰色GM(1,1)模型及其残差修正GM(1,1)模型预测门急诊人次,为医院科学化管理服务。方法收集某综合三级甲等医院2002年-2014年每月门急诊人次数据,其中2002-2013年的月数据用于建立GM(1,1)模型和残差修正GM(1,1)模型,2014年各月数据用于验证两个模型的预测效果,用平均相对误差绝对值(mean absolute percentage error,MAPE)评价模型的预测精度,并用两个模型预测2015年数据。结果 GM(1,1)模型和残差修正GM(1,1)模型的MAPE值分别为:10.92%,8.31%,残差修正GM(1,1)模型的预测精度高于GM(1,1)模型,两个模型的预测结果都显示,2015年门急诊人次将继续保持增长。结论两个模型适用于医院门急诊人次预测,可为医院管理提供科学依据。
Objective To forecast the number of emergent and emergency patients using gray GM (1,1) model and its residual GM (1,1) model to serve the hospital scientific management. Methods Monthly outpatient and emergency department attendance data of a comprehensive Grade A hospital from 2002 to 2014 were collected. The monthly data from 2002 to 2013 were used to establish GM (1,1) model and residual error correction GM (1,1) model The data of each month in 2014 were used to verify the predictive effect of the two models. The mean absolute percentage error (MAPE) was used to evaluate the prediction accuracy of the model. Two models were used to predict the 2015 data. Results The MAPE values of the GM (1,1) model and the residual GM (1,1) model were 10.92% and 8.31%, respectively. The prediction accuracy of the residual GM (1,1) model was higher than that of GM (1,1) 1) model, the predictions of both models show that the number of outpatient and emergency services will continue to grow in 2015. Conclusions The two models are suitable for predicting the number of outpatient and emergency departments in hospitals and providing a scientific basis for hospital management.