Prediction and analysis of red blood cells clinical demand based on the ARIMA model
YANG Xu
LIU Shuqi
HUANG Hui
CHEN Xiulan
LOU Yiling
CAO Shiyi
JIANG Qingqing
Abstract：Objective To construct an ARIMA multiplicative seasonal model suitable for predicting the clinical demand of red blood cells in Wuhu City,and provide a scientific basis for blood collection organizations to formulate red blood cell collection and supply balance programs and recruitment plans.Methods The ARIMA model was constructed based on the clinical use of red blood cells in Wuhu City Central Blood Station from January 2012 to December 2021.The data were processed through time series stabilization,model identification,and parameter verification to determine the optimal model.The red blood cell clinical use demand from January 2022 to August 2022 was predicted using the optimal model,and the prediction effect was verified using actual values.Results The optimal model was ARIMA(0,1,0)(1,1,0)12,with BIC=12.162.The ACF and PACF of the residual sequence were basically within the 95%confidence interval,and the Ljung-Box Q value was 15.265,with P=0.576>0.05,which met the requirements of white noise sequence,and the model fitting was validated.Except for April and May,the actual values of each month were within the 95%confidence interval of the predicted value,and the average relative error between the ARIMA model prediction value and the actual value was-0.00375,and the mean absolute percentage error(MAPE)was 7.087%,with good prediction effect.Conclusion The ARIMA(0,1,0)(1,1,0)12 model can be used to predict the clinical demand of red blood cells,and it can provide a reference for blood collection,non-remunerated blood donation recruitment and inventory management.
Keywords：Red blood cellARIMA multiplicative seasonal modelDemand forecastingHealth management
Publication Date：2025-11-25
Online Publishing Date：2025-12-16(First online date of this platform, not the publication date of the document)
Pages：7( 1272-1278 )