摘要
目的 探讨自回归移动平均乘积季节(ARIMA)模型在基层血站单采血小板临床需求量预测中的可行性,为血站制定科学合理的单采血小板招募计划和库存管理提供科学依据。方法 应用SPSS23.0统计学软件对顺德区中心血站2012年1月~2022年6月的单采血小板临床使用数据构建ARIMA模型,用所建模型预测2022年7月~2023年3月的单采血小板使用量并与实际使用量进行比较,评价模型拟合效果。结果 建立的最优模型为ARIMA(2, 1, 0)(0, 1, 1)12,残差序列自相关函数(ACF)和偏自相关函数(PACF)落在95%CI内;Ljung-Box Q统计量为24.941,差异无统计学意义(P>0.05),说明残差是随机分布的,残差不存在相关性,为白噪声序列,模型检验通过。应用模型ARIMA(2, 1, 0)(0, 1, 1)12对顺德区2022年7月~2023年3月的单采血小板临床使用量进行预测,预测结果与实际值均在95%CI内,平均相对误差为7.06%,预测值与实际值的曲线趋势基本一致,模型拟合效果较好。结论 ARIMA模型可用于顺德区单采血小板临床需求量的短期预测,为单采血小板的招募和库存管理提供科学依据。
Objective To explore the feasibility of Autoregressive Integrated Moving Average(ARIMA)model in predicting the clinical demand of apheresis platelets in primary blood stations,so as to provide scientific basis for blood stations to formulate scientific and reasonable apheresis platelet recruitment plan and inventory management.Methods SPSS23.0 was used to construct the ARIMA model for the clinical use data of apheresis platelets from January 2012 to June 2022 in Shunde District Central Blood Station.The model was used to predict the use of apheresis platelets from July 2022 to March 2023 and compared with the actual use to evaluate the fitting effect of the model.Results The optimal model was ARIMA(2,1,0)(0,1,1)12,and the residual sequence auto correlate function(ACF)and partial autocorrelation function(PACF)fell within 95%CI.The Ljung-Box Q statistic was 24.941(P>0.05),and the difference was not statistically significant(P>0.05),indicating that the residuals were randomly distributed,and there was no correlation between the residuals.It was a white noise sequence,and the model test passed.The model ARIMA(2,1,0)(0,1,1)12 was used to predict the clinical use of apheresis platelets in Shunde District from July 2022 to March 2023.The predicted results and the actual values were within 95%CI,and the average relative error was 7.06%.The curve trend of the predicted value and the actual value was basically the same,and the model fitting effect was good.Conclusion ARIMA model can be used to predict the short-term clinical demand of apheresis platelets in Shunde District,and provide scientific basis for the recruitment and inventory management of apheresis platelets.
作者
吕艺通
刘志泉
莫巧频
王东
谢庆欢
刘曼丽
LYU Yi-tong;LIU Zhi-quan;MO Qiao-pin(Blood Supply Service Department,Foshan Shunde District Central Blood Station,Foshan 528300,China)
出处
《中国实用医药》
2023年第23期144-148,共5页
China Practical Medicine
基金
佛山市卫生健康局医学科研项目(项目编号:20220315)。
关键词
自回归移动平均乘积季节模型
基层血站
单采血小板
需求量预测
Autoregressive integrated moving average model
Primary blood stations
Apheresis platelets
Demand forecast
