Medical material forecasting- An Example from a regional teaching hospital in Yunlin

• Main Authors: Jui-Hung Hsia, 夏瑞鴻
• Other Authors: Bo-Wen Cheng
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/80536233759945990230

碩士 === 國立雲林科技大學 === 全球運籌管理研究所碩士班 === 101 === NHI budget restrictions have an impact on hospital revenues that is easily affected by the proportion of inpatients to outpatients. Hospitals must strike a balance between hospital profitability and the quality of medical care. Medical material costs account for about 30% of hospital operating costs. Hospitals have to find a suitable inventory between inventory turnover and the stock-out rate. This research used ABC classification to classify the inventory. We chosed class A be the forecast object, and then used Holt exponential smoothing, multiple regression, and back-propagation neural network as the prediction methods. Finally, we used MSE, MAD, MAPE to assess the accuracy of the methods. The results show that the back-propagation neural network is a better method than the Holt exponential smoothing and multiple regression method and can serve as a demand forecasting model for managing hospital inventories.