基于R语言时间序列和ARIMA模型预测新建综合医院麻醉科耗材领用的研究

Research on Predicting Consumables Requisition of Anesthesiology Department in Newly-Built General Hospital Based on R Language Time Series and ARIMA Model

目的针对新建医院,运用自回归积分滑动平均(Auto Regressive Integrated Moving Average,ARIMA)模型建立合适的领用支出的医学经济学模型,借此预测麻醉科耗材支出需求的变化趋势,合理的控制科室预算。方法利用R软件对我院麻醉科2016年6月至2020年6月耗材支出数据建立ARIMA模型,将2020年7月至2021年6月耗材支出的实际值与预测值分别进行比较,评价模型的预测性能。采用统计学方法比较前4年麻醉耗材支出金额与第5年金额的差异性。结果建立ARIMA(1,0,2)模型对麻醉科耗材需求进行预测,对预测模型行ADF检验,P<0.01;使用LB统计量对残差序列进行检验时,P>0.05,建模成功;前4年和最后1年的支出金额比较,差异无统计学意义(P>0.05)。结论 ARIMA(1,0,2)模型能较好地拟合新建综合医院麻醉科的耗材领用需求,制定更合理的备货计划,降低库存积压成本,同时为计算科室预算提供理论依据。

Objective For newly-built hospital, auto regressive integrated moving average(ARIMA) was used to establish an appropriate medical economics model for expenditures, so as to predict the trend of changes in the expenditures of anesthesiology consumables and reasonably control department budget. Methods The R software was used to establish an ARIMA model for the consumable expenditure data of the department of anesthesiology in our hospital from June 2016 to June 2020. The actual value of the consumable expenditure from July 2020 to June 2021 was compared with the predicted value, and the model’s performance was evaluated. Statistical methods were used to compare the difference between the amount of anesthesia consumables spent in the previous four years and the amount in the fifth year. Results The ARIMA(1,0,2) model was established to predict the demand for consumables in the anesthesiology department. The ADF test of the prediction model was P<0.01. When the LB statistic was used to test the residual sequence, P>0.05, the modeling was successful. There was no statistically significant difference in the amount of expenditure in the first four years and the last year(P>0.05). Conclusion The ARIMA(1,0,2) model can better fit the consumables requirements of the anesthesiology department of a newly-built general hospital, formulate a more reasonable stocking plan, reduce the cost of inventory backlog, and provide a theoretical basis for calculating department budgets.