动态更新决策理论、模型、算法及应用

Dynamic Replenishment: Theories,Models,Algorithms,and its Applications

物流动态更新决策理论已经有近百年的历史,它既是现代物流管理的核心,又是管理科学的起源,经过多年的补充、衍变、提高和完善,更新决策理论已经成为管理科学—运筹学应用的典范。追溯不同的经典管理理念,不同的更新决策模型都殊途同归地获得了经济订货量公式EOQ,展现了EOQ所特有的"原始之美"。而由于制造、市场、财务和运营之间的管理冲突,又产生了成本中心说、利润中心说、商物分离说和服务中心说四大物流管理学说,显露出"混沌之美"。在需求呈现出线性变化和二次非线性变化的条件下,可以给出具有"和谐之美"的统一方程和解析算法,对应上述决策原则的著名算法PPT、Silver Meal、LUC和Ritchie均为其特例。在复杂的、动态的、非线性的需求情况下,杨氏方程提出了具有"统一之美"的数学结构,构造出了具有普适性的解析算法,这是对管理科学之科学统一性的诠注。

This paper attempts to review dynamic replenishment theories, models, algorithms and its applications since last century. Firstly, the original beauty of EOQ model is exhibited by examining four schools of classical management thoughts. Secondly, taking the conflicts among manufacturing, marketing, financing and operating functional areas as background, the beauty of chaos is being exposed referring to four kinds of management principles of physical distribution. Thirdly, a kind of algebraic equation and related analytic heuristic algorithms for linear and quadratic time varying demand patterns,are presented here,while four well-known heuristics such as PPT, Silver-Meal, LUC and Ritchie's are merely typical cases of the equation. Finally, a beautifully unified equation for dynamic replenishment is then developed with heuristic algorithm and conjecture work for complicated and non-linear demand patterns,which is a footnote for management science as the unity of science.