信息共享受限条件下的供应链网络系统牛鞭效应控制策略

研究在信息共享受限条件下供应链网络库存系统的牛鞭效应控制问题,建立了包括市场需求、信息可获得性、信息及时性等不确定性因素的库存网络系统状态转移模型,从系统内部动力学机制的角度分析了牛鞭效应的成因,提出了动态库存控制策略,并给出了策略参数设计的线性矩阵不等式组算法.运用系统稳定性理论,深入分析了信息共享对牛鞭效应的影响,并通过仿真结果验证了库存控制策略的有效性和实用性.     

供应商管理库存对牛鞭效应的影响

采用基于z变换的离散传递函数和测量牛鞭效应的控制工程方法,计算了一个由一个供应商和一个用户组成的、使用指数平滑预测的供应商管理库存供应链和传统供应链的牛鞭效应,并比较了这两种供应链对牛鞭效应的影响.通过仿真证实,应用供应商管理库存策略对供应链的物理过程进行再造是一种有效的减少牛鞭效应的方法.     

PI补充策略下供应链牛鞭效应的$H_infty$控制

牛鞭效应是指供应链管理订单制定环节中因信息扭曲造成的需求逐级放大的一种现象.针对供应链中的节点企业,在订货点法的基础上加以改进构建新的订单制定规则,并构建$H_\infty$控制器达到抑制牛鞭效应的目的,从而降低供应链整体成本.订单制定环节由企业订单规则和需求预测两个部分组成,为应对需求持续上升使安全库存发散的情况,在订货点法的基础上设计PI补充策略下的新订单规则,并以系统$H_\infty$范数与供应链牛鞭效应的指标定义相同为基础,引入$H_\infty$控制器代替预测函数.仿真结果表明,所设计的PI 补充策略下的$H_\infty$控制器法与传统订单制定算法相比,可有效削减牛鞭效应,并且使企业库存始终维持在一个安全稳定的状态.     

具有Markov跳变参数的闭环供应链系统切换控制

研究具有 Markov 跳变参数的闭环供应链(Closed-loop supply chain, CLSC)切换系 统建模以及具有抑制牛鞭效应的H∞控制问题. 针对再制造过程中的不确定性问题, 在考虑库存衰减因素的条件下, 根据库存水平的不同状态将系统建模为切换系统, 子系统间的切换服从 于一个Markov过程. 基于输入滞后的控制策略, 应用Markov切换思想对 系统进行控制器设计与性能分析, 在保证闭环供应链系统稳定的情形 下有效抑制牛鞭效应. 仿真例子说明所得结果的有效性.     

基于Anylogic的物流服务供应链牛鞭效应仿真分析*

针对当前牛鞭效应研究集中于产品供应链的现状,以及服务的无形性等众多与产品不同的特性,提出以服务能力模拟产品库存进行牛鞭效应研究的可能;构建了物流服务供应链的概念模型,归纳出其牛鞭效应的四个成因;以上下游服务能力调配策略为基础,构造了三阶物流服务供应链中的成本函数和等待时间函数,运用仿真软件Anylogic进行了实证分析和优化改进。结果表明牛鞭效应确实也存在于服务型供应链中,造成能力利用率偏低和资源浪费,同时验证了以服务能力模拟产品库存进行牛鞭效应研究的有效性。     

自适应控制对多级库存延迟问题的优化研究

针对在随机需求下交货延迟所导致供应链多级库存系统库存积压、缺货和牛鞭效应等问题,建立了基于自适应控制算法的多级库存动态优化模型。通过泰勒展开和拉布拉斯变换建立了基于APIOBPCS策略考虑延迟的动态多级库存控制模型;由Lyapunov渐进稳定性定理设计了一种适用于多级库存的模型参考自适应控制算法,其中以无交货延迟的参考库存模型作为目标,通过调节线性补偿函数和自适应控制率,逐渐缩小实际库存模型与参考库存模型间的输出误差,以此削弱交货延迟对多级库存模型的影响;通过实证数据验证了模型参考自适应控制对一个三级供应链库存系统的动态优化效果。仿真结果表明,自适应控制下的无信息共享多级APIOBPCS库存系统缺货全部归零,牛鞭效应下降40.7%。在不增加企业运营投入的前提下,通过自适应控制算法,优化资源配置,动态削弱了交货延迟对多级库存的影响,提升了供应链运营效率。     

基于鲁棒优化的供应链联合补货策略

供应链管理中的客户需求不确定性会导致需求信息偏差逐级放大的"牛鞭效应",但目前常用的供应链管理策略为供应商管理库存,没有考虑需求不确定的影响。为此,在单个供应商、多个零售商需求不确定的情况下,结合鲁棒优化法提出一种联合补货策略进行库存管理。构建一个非线性混合整数规划模型以计算两级供应链的总成本,通过总成本的变化来反映供应链系统的性能,采用鲁棒优化法求解供应链系统的最小总成本,并使用外部和内部两层迭代算法获得供应商和零售商的补货周期及补货数目。实验结果表明,与传统的供应链策略ERI和AR相比,该策略可有效降低供应链系统的总成本。     

考虑缺货策略的闭环供应链牛鞭效应系统动力学分析

为考察回收物流对供应链牛鞭效应的影响程度,建立了由需求方、制造商和原材料供应商、回收商组成的闭环供应链系统动力学模型。模型建立在供应能力受限并禁止退货的约束条件上,使系统成为分段线性复杂系统。分别在允许延期交货和不允许延期交货条件下进行实验,仿真实验结果表明,在不允许延期交货策略下,回收物流会放大牛鞭效应;而在允许延期交货策略下,若回收比例合理,则可以弱化牛鞭效应。     

神经网络在解决某企业供应链牛鞭效应问题中的应用

牛鞭效应是供应链运营管理中客观存在的现象。企业为了减少由实际需求和计划数量的偏差造成的生产不稳定,提高安全库存数量从而保证正常的生产活动,在此情况下需求逐级放大引发了牛鞭效应。精准预测是缓解牛鞭效应的重要手段,但是传统的时序预测在复杂的环境中并没有很好的预测效果。基于以上问题,从理论层面论证了需求预测、安全库存、牛鞭效应之间的关系,提出能够优化预测结果的ARIMA-BP模型。以某制造商企业近两年的产品订单数据为研究对象,分别用不同的预测模型对订单进行预测分析,再与该企业原预测模型下的牛鞭效应仿真结果进行对比。结果表明,ARIMA-BP的模型预测精度更高,能够有效地缓解牛鞭效应。     

多级集群式供应链跨链库存合作及鲁棒优化算法

以四级两单链的集群式供应链为研究对象,在两个单链相巨竞合条件下,建立了两单链问两零售商跨链问库存协作的时变鲁棒控制系统模型,以及牛鞭效应的弱化方法,并用H∞优化算法来求解动态最优决策.通过实例分析表明,集群式供应链跨链间协调及鲁棒优化能有效抑制牛鞭效应、降低库存水平,在减小订货波动的同时可提高对集群市场的响应速度.     

基于VMI&TPL 供应链的单供应商两制造商补货策略

供应商与第三方物流合作实施供应商管理库存,采用共同补货转运策略( , , )为地理位置相邻的两制造商供应零部件,并通过两制造商初期库存之差 来分配库存,允许制造商相互转运.首先分析随机需求下两制造商总库存的变化情况,求解共同补货转运策略( , , )下的库存、缺货数量和转运数量及其随各控制值变化的性质;然后,通过算例将提出的共同补货转运策略与供应商单独为各制造商补货的策略进行比较;最后,考察不同参数对共同补货转运策略效果的影响.     

Bullwhip effect measure in a seasonal supply chain

In this study, the bullwhip effect in a seasonal supply chain was quantified by considering a two echelon supply chain which consists of one supplier and one retailer. The external demand occurring at the customer was assumed to follow a SARMA (1, 0) X (0, 1) _s scheme, a seasonal autoregressive-moving average process, while the retailer employed an base-stock policy to replenish their inventory. The demand forecast was performed with a SARMA (1, 0) X (0, 1) _s using the minimum mean-square error forecasting technique. In order to develop the bullwhip effect measure in a seasonal supply chain, the lead time demand forecast, forecast error, and the optimal inventory policy at the retailer were derived in sequence. The variance of order quantity based on these results was obtained. Then, various properties were derived by analyzing the bullwhip effect measure. Specifically, it was determined that the seasonal cycle plays an important role in bullwhip effect under a seasonal supply chain. The findings also point out that the replenishment lead time must be less than the seasonal cycle in order to reduce the bullwhip effect. Therefore, the lead time needs to be reduced through collaboration between the retailer and supplier.     

Controller design and reduction of bullwhip for a model supply chain system using z-transform analysis

In this work, a discrete time series model of a supply chain system is derived using material balances and information flow. Transfer functions for each unit in the supply chain are obtained by z-transform. The entire chain can be modeled by combining these transfer functions into a close loop transfer function for the network. The model proves to be very useful in revealing the dynamics characteristic of the system. The system can be viewed as a linear discrete system with lead time and operating constraints. The stability of the system can be analyzed using the characteristic equation. Controllers are designed using frequency analysis. The bullwhip effect, i.e. magnification of amplitudes of demand perturbations from the tail to upstream levels of the supply chain, is a very important phenomenon for supply chain systems. We proved that intuitive operation of a supply chain system with demand forecasting will cause bullwhip. Moreover, lead time alone would not cause bullwhip. It does so only when accompanied by demand forecasting. Furthermore, we show that by implementing a proportional intergral or a cascade inventory position control and properly synthesizing the controller parameters, we can effectively suppress the bullwhip effect. Moreover, the cascade control structure is superior in meeting customer demand due to its better tracking of long term trends of customer demand.     

基于不同交货期策略的供应链库存优化与协调模型

根据供应链的实际运作情况建立了交货策略1,即基于时间窗的协议交货期模型,分析了以制造商为主的Stackelberg对策博弈中各方的最优决策;接着建立了交货策略2,即基于费用分担的可控交货期模型,分析了以分销商为主的Stackelberg对策博弈中各方的最优决策;然后对交货策略1和策略2进行了比较,给出了成本分担系数范围,提出了一种基于改进理想点的Nash协商交货期费用分担模型.数值分析比较了两种策略,表明费用分担模型是有效的.     

A study on inventory replenishment policies in a two-echelon supply chain system

Inventory control plays an important role in supply chain management. Properly controlled inventory can satisfy customers’ demands, smooth the production plans, and reduce the operation costs; yet failing to budget the inventory expenses may lead to serious consequences. The bullwhip effect, observed in many supply chain management cases, causes excessive inventory due to information distortion, i.e. the order amount is exaggerated while a minor demand variation occurs, and the information amplified dramatically as the supply chain moves to the upstream. In this paper, one of the main causes of bullwhip effect, order batching, is considered. A simplified two-echelon supply chain system, with one supplier and one retailer that can choose different replenishment policies, is used as a demonstration. Two types of inventory replenishment methods are considered: the traditional methods (the event-triggered and the time-triggered ordering policies), and the statistical process control (SPC) based replenishment method. The results show that the latter outperforms the traditional method in the categories of inventory variation, and in the number of backlog when the fill-rate of the prior model is set to be 99%. This research provides a different approach to inventory cost-down other than the common methods like: information sharing, order batch cutting, and lead time reduction. By choosing a suitable replenishment policy, the number of backorder and the cost of inventory can be reduced.     

复杂产品供应链质量控制方案递阶决策模型

提出一类供应链质量控制方案递阶决策模型,辅助复杂产品主制造商更好地推广和落实质量控制方案,以实现供应链整体质量的最优改善.首先搭建了一类新型的供应链质量屋分析平台,并将其表征为多目标规划模型,确定最优方案组合;然后,基于复杂产品供应链多层级结构设计质量屋网络,以控制资源为纽带构建多级多目标递阶决策模型,实现了质量控制方案的继承和衔接;最后,针对某型号商用飞机供应链开展案例研究,验证了所提出模型和方法的可行性和有效性.     

考虑多类零售商的库存与运输VMI 集成策略

研究由单批发商多零售商构成的供应商管理库存(VMI)供应链,批发商向外部供应商订货,并为价格和缺货成本不同的零售商补货.首先构造解析模型以分析如何为多类零售商确定不同的库存分配策略;然后将这一最优的库存分配策略与最优的先到先服务(FCFS)策略进行比较,同时比较相应简化的库存分配策略和简化的FCFS策略.解析和算例结果表明,简化的库存分配策略总是优于简化的FCFS策略,而最优的库存分配策略大部分情况下优于最优的FCFS策略.     

Quantifying and mitigating the bullwhip effect in a benchmark supply chain system by an extended prediction self-adaptive control ordering policy

An undesired observation known as the bullwhip effect in supply chain management leads to excessive oscillations of the inventory and order levels. This paper presents how to quantify and mitigate the bullwhip effect by introducing model predictive control (MPC) strategy into the ordering policy for a benchmark supply chain system. Instead of quantifying the bullwhip effect with commonly used statistical measure, we derive equivalently the expression of bullwhip metric via control-theoretic approach by applying discrete Fourier transform and (inverse) z-transform when the demand signal is stationary stochastic. A four-echelon supply chain is formulated and its dynamical features are analyzed to give the discrete model. An extended prediction self-adaptive control (EPSAC) approach to the multi-step predictor is implemented in the development of MPC formulation. The closed-form solution to MPC problem is derived by minimizing a specified objective function. The transfer function for MPC ordering policy is then obtained graphically from an equivalent representation of this closed-form solution. A numerical simulation shows that MPC ordering policy outperforms the traditional ordering policies on reducing bullwhip effect.     

Fuzzy knowledge-based token-ordering policies for bullwhip effect management in supply chains

The “Bullwhip Effect” is a well-known example of supply chain inefficiencies and refers to demand amplification as moving up toward upstream echelons in a supply chain. This paper concentrates on representing a robust token-based ordering policy to facilitate information sharing in supply chains in order to manage the bullwhip effect. Takagi–Sugeno–Kang and hybrid multiple-input single-output fuzzy models are proposed to model the mechanism of token ordering in the token-based ordering policy. The main advantage of proposed fuzzy models is that they eliminate the exogenous and constant variables from the procedure of obtaining the optimal amount of tokens which should be ordered in every period. These fuzzy approaches model the mentioned mechanism through a push–pull policy. A four-echelon SC with fuzzy lead time and unlimited production capacity and inventory is considered to survey the outcomes. Numerical experiments confirm the effectiveness of proposed policies in alleviating BWE, inventory costs and variations.