温特斯线性和季节性指数平滑法预测模型在医院管理中的应用探讨  

作　　者：杨建南[1] 李萍[2] 李世云[1] 钟鸣[3] 刘勇华[4] 何均华[5] 

机构地区：[1]成都铁路中心医院,成都市610081 [2]成都市妇产科医院,成都市610081 [3]成都市医学信息所,成都市610081 [4]成都市卫生局计财处,成都市610081 [5]成都铁路局科学技术研究所,成都市610081

出　　处：《中国卫生信息管理杂志》2010年第6期74-77,共4页Chinese Journal of Health Informatics and Management

基　　金：成都铁路局科技项目(CX0233)

摘　　要：目的探讨温特斯线性和季节性指数平滑法预测模型在医院管理中的应用价值。方法建立温特斯线性和季节性指数平滑法预测模型对成都市卫生局所属医院2009年1～4季度门急诊诊次进行季节预测,并评价该预测模型的预测精度。结果温特斯线性和季节性指数平滑法预测模型对成都市卫生局所属医院2009年1～4季度门急诊诊次的定量预测结果满意。通过计算机反复迭代试算,获得了一组最佳的平滑常数(α,β,γ),使预测模型的预测误差最小。该预测实例平均绝对百分误差(MAPE)为3.5703%<10%,预测精度较高。结论温特斯线性和季节性指数平滑法在医院管理中有实用价值。在建立预测模型时,结合计算机进行迭代运算,可建立最佳预测模型,进一步提高预测精度。Objective To construct Winter's exponential smoothing and seasonal exponential smoothing (WESSES) method and evaluate its application in hospital management.Method The prediction model of WESSES method was constructed and used to predict the outpatient visits of clinic and emergency of the hospitals governed by Chengdu Health Bureau from First to 4th quarter in 2009.The forecast accuracy of WESSES method was evaluated.Resulths The forecast accuracy of WESSES method was satisfied. The smooth constant(α,β,γ) with the smallest prediction error were calculated through iterating the pilot calculation repeatedly with computer.The mean absolute percentage error(MAPE) of this forecast example(3.5703%) was less than 10%presented high forecast accuracy.Conclusion It is valuable of WESSES method in hospital management.The forecast accuracy of WESSES method could be improved through iterating the pilot calculation repeatedly with computer while this method was constructed.

关 键 词：温特斯线性和季节性指数平滑法预测模型 平滑系数 预测精度 预测误差 门急诊诊次 

分 类 号：R197.3[医药卫生—卫生事业管理]