历史订单信息对牛鞭效应的影响分析

客户需求信息的失真是导致牛鞭效应存在的原因,基于零售商的历史订单数据对其需求进行预测可以部分消除牛鞭效应。论文基于零售商-分销商二级供应链视角,分析了在零售商的需求为线性自回归模式的二级供应链中,分销商利用零售商历史订单数据和现有订单数据进行需求预测时自身库存成本的变更以及整个供应链的牛鞭效应的缓解程度。结果表明:分销商利用历史订单数据进行库存的决策可以显著地降低自己的平均库存和需求的波动,这种降低程度在零售商的订货提前期较大的情况下比较明显,但是零售商的需求预测相关系数对它影响不大。     

基于SARMA需求模式的牛鞭效应分析

在自回归移动平均(ARMA)模型的基础上,建立需求过程为季节性自回归移动平均(SARMA)的时间序列,零售商采用最小均方差(MMSE)预测技术预测市场需求,库存采用补充订货至目标库存(order-up-to)策略的简单两级季节性供应链牛鞭效应量化模型,并对模型牛鞭效应的大小及其影响因素进行理论分析和实例验证,不仅刻画出各种情形下牛鞭效应存在的辨别条件和属性,而且实证结果表明煤炭供应链采用SARMA模型度量牛鞭效应优于ARMA模型.     

ARMA（1,1）需求下的信息延迟对供应链最优决策的影响

在假定顾客需求满足ARMA（1,1）过程的前提下,考虑了由一个零售商和一个供应商所组成的两级供应链系统最优订购决策问题.分别建立了需求信息不延迟与延迟这两种情形下零售商和供应商的最优订购决策模型,通过比较得出：当需求呈正相关时,需求信息延迟不仅可以减小牛鞭效应,而且可以降低供应链系统的平均总成本.     

基于产品需求替代的两级供应链牛鞭效应分析

考虑两种产品需求替代下的自回归移动平均时间序列,零售商采用最小均方差技术预测市场需求,以订货点法确定订货量的两级供应链牛鞭效应量化模型,并对该模型牛鞭效应的大小及其影响因素进行理论分析和算例验证,研究表明:1)需求替代情形下牛鞭效应的表达式对有无需求替代的情况都适用;2)替代系数对被替代产品的牛鞭效应无影响,对替代产品的牛鞭效应有影响,并且替代系数同替代产品的牛鞭效应变化方向一致;3)需求替代情形下缩短提前期并不一定能减少替代和被替代产品的牛鞭效应.     

供应链零售商预测技术研究—基于牛鞭效应的视角

一阶自回归(AR(1))序列模拟需求过程是传统文献采用的经典模型,然而上述文献关于需求过程参数(如需求自回归系数)对牛鞭效应的影响分析缺乏实践意义,为了更符合企业的实际决策过程,本文建立了需求依赖于价格、而以AR(1)序列模拟价格过程的需求函数模型,分析了最小均方差、移动平均和指数平滑预测下的牛鞭效应,确定了零售商的预测技术选择条件。研究表明:(1)产品市场规模不影响零售商预测技术的选择;(2)当产品价格敏感系数较小或价格自回归系数较小时,零售商应选择最小均方差预测技术;(3)当产品价格敏感系数和价格自回归系数均较大时,零售商应选择移动平均预测技术。     

牛鞭效应弱化与反牛鞭效应强化在完美信息下联合作用的博弈

在供应链运作过程中,同时存在牛鞭效应与反牛鞭效,若仅考虑到供应链的成本、需求偏差等问题,这种存在会因有限理性的驱使使得牛鞭效应弱化与反牛鞭效应强化.因此,认为供应链的上下游在周期内会表现出牛鞭效应弱化与反牛鞭效应强化的联合作用,联合作用使得单个企业达到低平均库存成本,也意味着供应链的整体库存最低且整体市场需求偏差最低,间接地、自动地从整体上消除牛鞭效应或反牛鞭效应,使得整条供应链不管是短期的还是长期来看是最佳的,若是长期,还会给供应链企业带来显著的战略优势.     

需求与库存水平相关的供应链量折扣协调模型

考虑由一个供应商和一个零售商构成的供应链系统,零售商的市场需求依赖于该产品的库存水平,研究了利用批量折扣实现供应链完美协调的问题.首先,我们得出了在分散式系统下供应链无法实现完美协调,讨论了供需双方的最优决策.其次,作为stackelberg的主导方,供应商提供批量折扣计划,得出了此折扣计划实现系统完美协调的条件,分别给出了实现完美协调和不能实现完美协调时的批量折扣计划.最后分析了需求函数中的参数和产品的成本对实现供应链系统完美协调的影响.     

季节性供应链零售商补货策略研究——基于牛鞭效应视角

通过建立含有季节性自回归移动平均需求过程的供应链,零售商采用最小均方差预测技术预测提前期需求,分析(R,D)、(R,S)、(R,βS)、(R,γO)和(R,γO,βS)五种补货策略下的牛鞭效应.研究结果表明:(R,γO)补货策略是弱化牛鞭效应的最优补货策略,然而(R,γO)补货策略时出现了反牛鞭效应,无法保证供应链的安全供给.实践中当库存量调节系数和订货量调节系数较大时,(R,βS)补货策略能有效弱化牛鞭效应,当库存量调节系数和订货量调节系数较小时,(R,γO,βS)补货策略能有效弱化牛鞭效应;对于(R,βS)和(R,γO,βS)补货策略,牛鞭效应随着库存平滑系数的增大而增大;对于(R,γO)和(R,γO,βS)补货策略,牛鞭效应随着订货平滑系数的增大而增大;对于(R,S)、(R,βS)和(R,γO,βS)补货策略,牛鞭效应随着订货提前期的增大而增大;对于(R,γO)和(R,γO,βS)补货策略,牛鞭效应随着时刻t的增大而增大,但时刻t增大到一定程度时,牛鞭效应值基本不变.     

提前期需求为模糊随机条件下的连续库存补货策略

考虑提前期内需求为模糊随机变量且提前期为可缩短情形下,建立由购买商和供应商所组成的简单供应链连续库存补货策略优化模型,其中订单量、再订货点和提前期为决策变量.首先推导出模糊随机需求条件下购买商和供应链的成本函数,然后,进一步考虑总需求为三角模糊数,推导出供应商、购买商和供应链的模糊成本函数.在此基础上分别从购买商成本最小和供应链成本最小角度对模型进行求解,结合具体算例对模型进行应用分析和比较分析,结果表明模型具有有效性和实用性,并得出如下结论:从购买商本身角度考虑订购策略所产生的供应链成本总是大于从供应链整体角度考虑订货策略所产生的供应链成本,同时从购买商本身角度考虑订货策略所产生的最优订购量、购买商成本低于从供应链整体角度考虑订货策略所产生的最优订购量、购买商成本.     

需求信息更新下资金约束供应链贸易信贷定价及订购决策

研究了需求信息更新情形下,单一制造商和单一资金约束零售商组成的两级供应链的定价与融资订购决策问题。研究发现当零售商初始资金水平越低或更新的需求估计值越大时,制造商将会采取更为激进的定价策略;零售商初始资金水平越低,其订购数量越大,仅当更新的需求估计值低于某一阈值时,零售商订购数量关于更新的需求估计值递增。此外,数值模拟显示,相对于制造商,需求信息更新更容易为零售商带来额外价值;在一定条件下,制造商和零售商可同时从需求信息更新中获得额外价值。     

随机需求下有产能约束的技术创新策略研究

在随机需求和技术变革的环境下,基于有产能约束的单供应商-单零售商的供应链结构,研究供应商分销价格决策和技术创新策略以及零售商订货决策。建立了三阶段Stackelberg博弈模型,通过逆推方法求得了供应商最优分销价格和技术创新策略以及零售商最优订货量,深入探讨了供应商产能、新技术出现概率以及市场需求期望与波动分别对供应商、零售商和供应链整体利润的影响。结果表明当供应商产能不足时进行技术创新会提高供应商和供应链的利润,但零售商因间接承担供应商技术创新的投资成本而利润下降;当供应商产能过剩时进行技术创新则会降低供应商及供应链的利润,而零售商的利润增加。新技术出现概率增加会提高供应链各成员的利润;提高市场需求期望并减小市场波动对供应商及供应链有利,但可能会降低零售商的利润。     

需求随价格变化的具有折扣的易变质物品的库存模型

本文研究了在需求随价格变化及物品易变质的条件下，当供应商给予数量折扣时的库存问题。证明了当供应商给予数量折扣时，零售商的需求量是增大的，并给出了供应商给予数量折扣时零售商的订货量和订货周期的计算方法。对物品变质率和需求价格敏感系数对零售商的订货量、订货周期、出售价格和单位时间利润的影响进行了数值分析，并给出了数值算例。     

非对称信息下供应链批发价格契约应对突发事件

针对两个制造商与一个零售商组成的二级供应链,建立了制造商一方与零售商联盟,制造商另一方与之竞争、零售商类型为其私人信息下的博弈模型,从系统优化的角度研究了供应链系统批发价格如何应对突发事件.首先证明稳定条件下,零售商成本信息为非对称信息时,批发价格契约能实现集中控制与分散控制下-该供应链的协调.当突发事件造成零售商的成本分布变化时,通过求解突发事件下供应链系统最优利润的Karush-Kuhn-Tucker(KKT)条件,提出了供应链应对突发事件的最优策略,进一步证明了改进的批发价格契约可以协调突发事件下分散化决策的供应链系统.数值仿真表明,当零售商的成本在一个较小的范围内扰动时,原有契约机制具有一定的鲁棒性,而当扰动较大时,需要对原来的生产计划进行调整.     

随机需求条件下的延迟发运策略模型及性质

本针对随机需求条件下物流配送中心的库存和运输联合决策问题，在基本库存和自身运输能力不足的情况下，提出对剩余客户订货需求采取部分延迟到下一期与部分利用第三方物流立即发运两相结合的策略，并在具有一般惩罚(损失)费延迟发运量限制的条件下，建立运输和库存相关总成本数学期望最小的优化模型，论证了该模型的主要性质，在此基础上很容易构造求解该类问题的优化方法。     

The use of advance demand information in a project-based supply chain

In a project environment, a manufacturer is confronted with two types of demand: regular demand from many small orders and very irregular, lumpy demand from infrequent, large orders. Manufacturers who build to stock must carry large safety stocks to meet the lumpy demand. As part of the project engineering process, however, project engineers and implementers (e.g. installers) typically have developed information about material requirements well in advance of placement of orders. We analyze the inventory reduction that could be achieved if the installer were to communicate advance demand information (ADI) to the manufacturer. We look at it in particular when the bid is placed. We focus on the following characteristics of available ADI in project environments: First, ADI information is uncertain, because decisions on installer and manufacturer selection have not yet been finalized. Second, information is detailed, available at the item level. We show that ADI is particularly valuable when potential demand for large projects is irregular and when proposals for potential projects have a high probability of leading to orders.     

Managing Dynamic Inventory Systems with Product Returns: A Markov Decision Process

This paper presents a Markov decision process for managing inventory systems with Markovian customer demand and Markovian product returns. Employing functional analysis, we prove the existence of the optimal replenishment policies for the discounted-cost and average-cost problems when demand, returns, and cost functions are of polynomial growth. Our model generalizes literature results by integrating Markovian demand, Markovian returns, and positive replenishment lead times. In particular, the optimality of the reorder point, order-up-to policies is proved when the order cost consists of fixed setup and proportional cost components and the inventory surplus cost is convex. We then make model extensions to include different cost components and to differentiate returned products from new ones. Finally, we derive managerial insights for running integrated closed-loop supply chains. At the aggregate level, returns reduce effective demand while many structural characteristics of inventory models are intact. A simple heuristic for managing systems with returns is to still utilize literature results without returns, but effective demand is lower than customer demand.     

On the value of customer information for an independent supplier in a continuous review inventory system

We consider the inventory control problem of an independent supplier in a continuous review system. The supplier faces demand from a single customer who in turn faces Poisson demand and follows a continuous review (R, Q) policy. If no information about the inventory levels at the customer is available, reviews and ordering are usually carried out by the supplier only at points in time when a customer demand occurs. It is common to apply an installation stock reorder point policy. However, as the demand faced by the supplier is not Markovian, this policy can be improved by allowing placement of orders at any point in time. We develop a time delay policy for the supplier, wherein the supplier waits until time t after occurrence of the customer demand to place his next order. If the next customer demand occurs before this time delay, then the supplier places an order immediately. We develop an algorithm to determine the optimal time delay policy. We then evaluate the value of information about the customer’s inventory level. Our numerical study shows that if the supplier were to use the optimal time delay policy instead of the installation stock policy then the value of the customer’s inventory information is not very significant.     

Random Demand Satisfaction in Unreliable Production–Inventory–Customer Systems

A method for calculating the probability of customer demand satisfaction in production–inventory–customer systems with Markovian machines, finite finished goods buffers, and random demand is developed. Using this method, the degradation of this probability as a function of demand variability is quantified. In addition, it is shown by examples that the probability of customer demand satisfaction depends primarily on the coefficient of variation, rather than on the complete distribution, of the demand.     

Analysis of optimal opportunistic replenishment policies for inventory systems by using a (s,S) model with a maximum issue quantity restriction

The analysis of optimal inventory replenishment policies for items having lumpy demand patterns is difficult, and has not been studied extensively although these items constitute an appreciable portion of inventory populations in parts and supplies types of stockholdings. This paper studies the control of an inventory item when the demand is lumpy. A continuous review (s,S) policy with a maximum issue quantity restriction and with the possibility of opportunistic replenishment is proposed to avoid the stock of these items being depleted unduly when all the customer orders are satisfied from the available inventory and to reduce ordering cost by coordinating inventory replenishments. The nature of the customer demands is approximated by a compound Poisson distribution. When a customer order arrives, if the order size is greater than the maximum issue quantity w, the order is satisfied by placing a special replenishment order rather than from the available stock directly. In addition, if the current inventory position is equal to or below a critical level A when such an order arrives, an opportunistic replenishment order which combines the special replenishment order and the regular replenishment order will be placed, in order to satisfy the customer's demand and to bring the inventory position to S. In this paper, the properties of the cost function of such an inventory system with respect to the control parameters s, S and A are analysed in detail. An algorithm is developed to determine the global optimal values of the control parameters. Indeed, the incorporation of the maximum issue quantity and opportunistic replenishment into the (s,S) policy reduces the total operating cost of the inventory system.