Application of ARIMA Model to Drug Storeroom in Drug Purchasing Prediction
- VernacularTitle:自回归整合移动平均模型在医院药库采购预测中的应用
- Author:
Jin HAN
;
Qingguo ZHAO
;
Rongrong WU
- Publication Type:Journal Article
- Keywords:
Time series analysis;
ARIMA model;
Prediction;
Drug purchasing;
Drug storeroom
- From:
China Pharmacy
2007;0(31):-
- CountryChina
- Language:Chinese
-
Abstract:
OBJECTIVE:To explore the new drug purchasing mode using autoregressive integrated moving-average (ARIMA) prediction model for improvement of the working quality and efficiency in hospital drug storeroom.METHODS:Drug consumption data from week 1 to week 47 in 2008 were collected.According to ABC method,category A drugs were defined among which 10 kinds of drugs were sampled randomly.Based on the data of from week 1 to week 44 in 2008,software SPSS13 was applied for the modeling and fitting of ARIMA model.The established model was applied to predict the data of from week 45 to 47,with the predicated data compared with the actual consumption data.RESULTS:The predicted purchasing amount using ARIMA model were consistent with the actual consumption data,with prediction accuracy for quantity at 89.19% and prediction accuracy for whole unit of purchased drugs at 97.56%,respectively.CONCLUSIONS:Good fitting and high short-medium term predication accuracy were obtained in the prediction using ARIMA model,and which could provide scientific support for drug purchasing and help manage the drug stock reasonably without appearance of out of stock or overstock.
