首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于MDP的血小板库存最优订货策略和使用策略研究
引用本文:李猜,耿娜,王春鸣.基于MDP的血小板库存最优订货策略和使用策略研究[J].运筹与管理,2019,28(7):108-117.
作者姓名:李猜  耿娜  王春鸣
作者单位:1.上海交通大学 工业工程与管理系,上海 200240; 2.上海交通大学 中美物流研究院,上海 200030; 3.上海交通大学 医学院附属仁济医院,上海 200127
基金项目:国家自然科学基金项目(71471113,71671111,71432006)
摘    要:血液是一种特殊的稀缺资源,临床上通过输血可以治疗一些血液疾病甚至挽救生命。目前血液尚不能人工制造,只能通过献血者捐献的方式获得,因此需要合理的库存管理来有效利用有限的血液资源。近期研究表明,血小板生命周期较短,且其输注效果会随着储存时间的延长而降低。因此,本文针对血小板的库存控制问题,考虑短缺、浪费和输血新鲜度之间的平衡,以效用最大化为目标建立有限时域的马尔科夫决策过程模型,探讨最优的血小板订货策略和使用策略。根据最优策略数值解的主要特征提出了启发式策略,数值实验结果表明所提启发式控制策略性能接近于最优控制策略,且优于传统策略。

关 键 词:血小板  库存控制  马尔科夫决策过程  阈值策略  配额策略  
收稿时间:2017-09-21

Optimal Ordering and Issuing Policy for Platelet Inventory ControlProblem Based on Markov Decision Process
LI Cai,GENG Na,WANG Chun-ming.Optimal Ordering and Issuing Policy for Platelet Inventory ControlProblem Based on Markov Decision Process[J].Operations Research and Management Science,2019,28(7):108-117.
Authors:LI Cai  GENG Na  WANG Chun-ming
Institution:1.Department of Industrial Engineering and Management, Shanghai Jiao Tong University, Shanghai 200240, China; 2.Sino-US Logistics Institute, Shanghai Jiao Tong University, Shanghai 200030, China; 3.Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 200127, China
Abstract:Blood is a special and rare resource, and transfusion of blood can be used clinically to treat patients for some blood diseases and may even save lives. Since blood cannot be manufactured, and can only be donated, effective inventory management is needed to make efficient use of the limited blood resource. Recent researches show that the shelf life of platelets is limited and the utility of transfusion platelet declines with the increasing of storage time. This paper studies the inventory control problem of the platelets. To balance the shortage, the wastage and the freshness of the transfused platelets, this paper proposes a finite horizon Markov Decision Process(MDP)model with the objective of maximizing the total utility. By solving and observing the optimal ordering and issuing policy, this paper proposes a heuristic policy, and numerical results show that this heuristic policy is close to the optimal policy and superior to the traditional policy.
Keywords:platelet  inventory control  markov decision process  threshold policy  quota policy  
本文献已被 CNKI 等数据库收录!
点击此处可从《运筹与管理》浏览原始摘要信息
点击此处可从《运筹与管理》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号