具有多元马氏需求的易变质多产品的EOQ模型  

The EOQ Model with Multivariate Markov Demand for Deteriorating Items

作　　者：陈杰[1,2] 陈志祥[1] 

机构地区：[1]中山大学管理学院,广东广州510275 [2]琼州学院理工学院,海南三亚572022

出　　处：《管理工程学报》2016年第4期93-101,共9页Journal of Industrial Engineering and Engineering Management

基　　金：国家自然科学基金资助项目(71372154);海南省自然科学基金资助项目(20151008);海南省高等学校科学研究基金资助项目(HJKJ2012-42)

摘　　要：本文研究一类新的易变质多产品库存控制策略,即具有多元马氏需求特征的多产品的最优(Q,T,?t)策略,该策略综合考虑了市场需求在不同产品之间具有马氏转移特征和变质率因素。论文首先建立了多产品多周期的多元马尔可夫需求预测模型,并通过该模型确定了各种产品在订货周期内的需求之间的关系。同时,在该关系的理论基础上,进一步给出了多产品多周期库存的最优(Q,T,?t)策略,进而结合算例给出模型的最优策略的数值解。The deteriorating item is a special one in the commodity market because a significant characteristic of its quality is time dependent. However, the suppliers of perishable products in China rely on their experiences to make ordering decisions instead of scientific methods. This has resulted in high average wastage rate of perishable products in China. The problem of perishable inventory control under the condition established by objective factors is a theoretical issue with practical significance. There are many references about joint decision of perishable multi-product inventory under stochastic term. However, the existing literature on inventory control policy considers the random demand state that is independently distributed. There is a lack of study on inventory control policy with mutual transfer characteristics of the multi-product demand. Based on multi-product demand with mutual transfer characteristics, the multivariate Markov demand model is established to forecast demand state of perishable multi-products for the inventory system in the next period. The relationships of demand for various products are determined. Meanwhile, the model of total expected cost per unit time for multi-product is proposed. The optimal control policy for multi-product inventory has been studied systematically under the total expected cost model. In this study, we propose a new class of control policy for perishable multi-product inventory which is the optimality of（Q,T,t） policy for multi-product inventory with multivariate Markov demand. How to calculate the time of the inventory level when it is equal to zero, the length T of ordering period and ordering quantity Q are extremely important for the process of optimizing control system to manage perishable inventory. When the three variable values are too large, it means that the more deteriorated products, the higher cost of the loss. The optimality of（Q,T,i）policy not only comprehensively considers the market demand with Markovian transfer characteristics and det

关 键 词：易变质产品 库存 多元马尔可夫模型 最优(Q T t)策略 

分 类 号：O211.62[理学—概率论与数理统计] F253.4[理学—数学]