Blood demand forecasting-An Example from a regional teaching hospital in Yunlin

• Main Authors: SHIN-HAO CHEN, 陳信豪
• Other Authors: Bor-Wen Cheng
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/d9sbp6

碩士 === 國立雲林科技大學 === 工業工程與管理系 === 103 === Blood is scrapped by medical staffs every year. There many problems about blood inventory and order quantity need to improve in the hospital. The goal of this study is to reduce the rate of scrapping. Thus, this study expects to forecast the demand of long-term blood. The orders of blood depend on medical staffs experiences currently. Blood is scrapped for many reasons. One of reasons is obsolescence. This study uses different forecast methods, regression, back-propagation neural network and exponential smoothing to forecast the demand of blood and expects to construct the forecast model of demand. Additionally, MSE, MAE and MAPE used to evaluate model. The results show that the single exponential smoothing is a better method than the double exponential smoothing, back-propagation neural network, multiple regression method, winters method, and addwinters method. This model could offer some suggestions for managing the blood.