基于多Agent的Supply-Hub库存策略建模与仿真研究

Supply-Hub是一种能较好地解决装配式供应链协同运作问题的供应链运作模式,其库存参数设置对于平衡上游供应商与下游制造商之间的利益,吸弓l供应商加入Supply-Hub非常关键.建立了由两个供应不同零部件的供应商向制造商补货时的供应商多周期成本模型,采用多Agent仿真方法,研究了三种情形下(s,S)库存策略参数的变动对供应商及制造商的影响,以期为Supply-Hub运营商合理确定库存策略参数提供一定参考.     

基于多Agent的Supply-Hub库存策略建模与仿真研究北大核心

Supply-Hub是一种能较好地解决装配式供应链协同运作问题的供应链运作模式,其库存参数设置对于平衡上游供应商与下游制造商之间的利益,吸弓l供应商加入Supply-Hub非常关键.建立了由两个供应不同零部件的供应商向制造商补货时的供应商多周期成本模型,采用多Agent仿真方法,研究了三种情形下(s,S)库存策略参数的变动对供应商及制造商的影响,以期为Supply-Hub运营商合理确定库存策略参数提供一定参考.     

快速补货模式下的动态定价与库存联合策略

在供应有限的情况下,研究常规补货和快速补货下商品动态定价问题.首先,建立了动态规划模型,理论证明了最优库存策略是基于(s,S)策略下改进的基本库存策略.其次,提出了一种启发式策略求复杂系统的最优策略,启发式算法能够求出最优价格和最优库存水平.最后,数值算例研究表明,库存管理中采用快速补货提高了零售商的利润;初始库存水平越高零售商的利润越高.     

具有定价和开机成本的生产-库存系统的最优(s,d,S)策略

考虑一个具有有限容量和开机成本的连续盘点生产-库存系统, 其控制策略为(s,d,S)策略. 未被满足的需求都会丢失. 当机器处于关闭状态时,库存产品可以两个不同的价格进行销售. 当机器处于开机状态时,库存只能以较高的价格进行销售. 研究了如何发现该系统下的最优(s,d,S)策略,并开发了用于计算最优控制参数的有效算法.     

基于ERP的(s,S)策略下库存优化控制决策支持系统

针对当前ERP软件系统无法动态地给出优化订购策略并对历史数据进行有效的分析等不足,以最小化库存费用为目标建立起折扣准则下库存优化数学模型,对ERP软件中导出的各类历史数据进行模型化分析,动态地得出各类产品(s,S)结构形式的优化订购策略.基于该模型设计并开发了库存优化控制决策支持系统,为用户提供决策支持,很大程度降低了企业库存费用.     

一类跳扩散需求存贮系统(s,S)库存控制策略研究

考虑的是连续检查库存,需求为一个常时间函数和-个复合Poison跳扩散随机过程的和的存贮系统最优库存控制问题.基于期望折扣成本最小建立了无穷时间区间具有固定订购成本的最优库存模型,确定可采用(s,S)策略进行库存控制,给出了最优(s,S)策略的充要条件--HJB方程Ⅰ、Ⅱ.我们采用"猜测"的方法确定了最优(s,S)策略对应的值函数形式,建立了确定库存参数的最优化模型.     

具有不耐烦顾客的生产服务库存系统最优生产策略

研究了基于(s,S)策略有不耐烦顾客的生产服务库存系统.建立了水平相依的拟生灭过程,通过截尾近似的方法求出系统稳态概率向量,并给出系统一些性能指标.最后,设计遗传算法并通过数值算例求解最优生产策略.     

带有服务员休假的库存系统

研究了带有服务员多重休假、损失销售和(s,S)补货策略的库存系统,其中顾客到达为泊松过程、休假时间及系统前置补货时间都服从指数分布.利用拟生灭过程方法对系统进行稳态分析,给出了带有休假的库存系统的稳态分布、平均库存和平均损失率,并将带有休假库存系统的性能指标与经典无休假库存系统的性能指标做了比较.最后通过数值算例说明服务员休假对库存系统的影响.     

具有不耐烦顾客和多重工作休假的排队库存系统

研究具有不耐烦顾客和多重工作休假的M/M/1/N排队库存系统模型,分别考虑了系统中库存为零时服务员休假和系统中顾客数为零时服务员休假两种休假方式,基于(s,S)库存策略,运用矩阵迭代方法得到了系统稳态概率分布,并给出系统相关性能指标,进而建立系统平均库存费用函数.通过数值算例对比分析了两种休假方式下的系统主要参数变化对系统重要性能指标的影响,并在最优费用的层面对两个模型的优劣进行了对比分析.     

考虑疫情风险与双重时效性的生鲜品供应中断库存控制策略研究

为解决重大疫情引发的供应中断导致生鲜品库存水平振荡,在市场需求随机变化的情形下,建立一个由供应商、配送中心和零售商组成的三级生鲜供应链库存系统,考虑由疫情引起的三种风险情景,引入系统动力学模型对零售商动态库存系统运营进行仿真分析。研究发现：受疫情风险传导系数和变质率两个序参量影响,供应链库存呈现振荡趋势;通过确定不同供应中断时长下的疫情风险情景提出优化保鲜投入策略、安全库存策略、以及共享库存联合提前转运策略有效降低零售端库存水平振荡并使其呈现渐稳趋势,实现产品在交付过程中的双重时效性,达到供需匹配,缓解疫情风险带来的影响,为相关零售企业提供决策支持。     

VMI条件下具有复合二项随机需求的销售商库存策略研究

考虑一个典型的单一产品的二级供应链系统:单供应商对单销售商,假定系统中销售商的需求分布为复合二项分布,未满足的需求机会损失;补货间隔时间为一随机变量.本文采用概率方法对销售商的需求分布、期望缺货、期望库存周期及库存的稳定性分布进行研究的基础上,构建了使单位时间内销售商的期望库存成本费用最小的库存模型,由此模型便可确定VMI模式下供应商对销售商的库存补货参数s和S,并且给出了在补货响应时间为泊松分布的情况下模型的求解算法,还给出了及时补货响应情况下的5个算例.为补货策略的实施提供了一种简单易于控制的思路和方法.     

存货质押融资模式下的供应链协调

针对供应商、零售商和银行组成的供应链金融系统,研究了存货质押融资模式下的供应链协调。零售商作为中小企业,资金有限,需要向银行申请贷款,然而零售商信用较低难以获得融资。供应商作为供应链的核心企业,采用含有回购价格和回购比例的回购契约为零售商提供担保,使得零售商融资得以实现。分析了不确定需求下供应链金融系统,供应链以及零售商的最优决策,给出了融资情况下供应链协调的条件,指出了在供应链协调情况下批发价格和回购策略对供应链收益分配的影响,说明了回购的作用:一方面使零售商易于获得银行融资,另一方面使供应链协调得以实现,指出了质押率、贷款利率和零售商自有资金等参数对最优决策的影响。数值算例验证了理论分析结果。     

区块链技术影响下的供应链系统动态响应性研究

为解决供应链系统中信息时滞和不对称问题,加快推动区块链技术在供应链管理中的创新发展,本文运用控制理论构建由分销商和零售商组成的二级供应链系统控制模型,引入区块链技术影响下的信息校正因子,推导系统的传递函数,通过MATLAB仿真不同需求信号下的订单可变性和库存波动。仿真结果表明:(1)区块链技术的应用提高了库存系统的精准性和稳定性;(2)高库存可变性伴随着高订单可变性;(3)指数平滑系数和区块链技术影响下的信息校正因子有效抑制供应链中的牛鞭效应;(4)信息延迟时间越长,区块链技术影响下的信息校正因子对控制系统的抑制作用越显著。本研究量化了区块链技术对供应链中牛鞭效应的影响,提高了供应链系统的精准性和稳定性,丰富了区块链技术在供应链管理中的应用,为企业管理者提供新的研究思路。     

A serial inventory system with supplier selection and order quantity allocation

Given the prevalence of both supplier selection and inventory control problems in supply chain management, this article addresses these problems simultaneously by developing a mathematical model for a serial system. This model determines an optimal inventory policy that coordinates the transfer of items between consecutive stages of the system while properly allocating orders to selected suppliers in stage 1. In addition, a lower bound on the minimum total cost per time unit is obtained and a 98% effective power-of-two (POT) inventory policy is derived for the system under consideration. This POT algorithm is advantageous since it is simple to compute and yields near optimal solutions.     

A robust optimization approach to reduce the bullwhip effect of supply chains with vendor order placement lead time delays in an uncertain environment

Supply chain management is important for companies and organizations to improve their business and enhance competitiveness in the global marketplace. The bullwhip effect problem of supply chain systems with vendor order placement lead time delays in an uncertain environment is addressed in this paper. Among the numerous causes of bullwhip effect, we focus on uncertainties with respect to demand, production process, supply chain structure, inventory policy implementation and especially vendor order placement lead time delays. Minimizing the negative effect of these uncertainties in inducing bullwhip effect creates a need for developing dynamical inventory policy that increases responsiveness to demand and decreases volatility in inventory replenishment. First, a dynamic model of supply chain with above uncertainties is developed. Then, a novel uncertainty-dependent robust inventory control method using inventory position information is proposed. Additionally, the maximum allowable vendor order placement lead time delay that ensures the stability of supply chains and the suppression of bullwhip effect under the proposed inventory control policy is explored and measured. We find that vendor order placement lead time delays do play important role in supply chain dynamics and contribute to its turbulence and volatility. The effectiveness and flexibility of proposed method is verified through simulation study.     

JIT环境下工厂线边库存容量投资协调策略研究

在JIT环境下,制造商要求长期合作的供应商在其工厂周边租用仓库,而自己只维持较少的线边库存。本文以此为背景,假定制造商投资建设工厂线边库存的费用与其容量大小成正比,分别建立了供应链分散决策和集中决策模型,并提出了制造商工厂线边库存投资建设协调模型。研究结果表明,相比分散决策,供应链集中决策时制造商投资建设的最优工厂线边库存容量更大,从而有利于供应商和整个供应链,而不利于制造商。算例分析证明提出的基于不对称博弈的线边库存投资费用分担策略能够实现整个供应链利益的帕累托优化。     

Analysis of supply chains with quantity based fixed incentives

This paper examines the use of quantity based fixed incentives to coordinate inventory decisions in a decentralized supply chain. We consider a two stage supply chain of autonomous supplier and distributor and prove that the optimal ordering policy for the newsvendor distributor under fixed incentives is an (s,S)$(s\text{,}S)$ type policy. We further show that external and internal quantity based incentives can restore channel coordination in single period and channel members can benefit through arbitrary splitting of the resulting additional chain profit. The single period results are extended to multiple periods and the impact of fixed incentives on the distributor’s optimal stocking policy and channel efficiency are examined under three different multi-period supplier strategies. Numerical examples are used to compare the multi-period strategies and to provide additional managerial insights. The results show that contrary to common belief, incentive plans developed and maintained based only on current inventory data perform poorly in long term and that such incentive plans must be periodically updated to enhance their efficiency. Furthermore, we show that high level of incentives designed to push too much inventory downstream of the supply chain can actually reduce the chain’s efficiency.     

The impact of stock-dependent demand on supply chain dynamics

This paper presents an investigation on the dynamics of a supply chain system under stock-dependent demand. Considering the feature of piecewise linearity, a switched linear model composed of three subsystems is developed. Based on the switched model, some analytical stability results are derived. Simulation experiments are designed to verify the stability results and observe nonlinear dynamics. We show that stock-dependent demand not only leads to different stability results but also makes nonlinear dynamics more complicated. We also reveal that the nonlinear dynamics of the switched model, such as chaotic and periodic fluctuations of inventory and order, are essentially caused by switching frequently among subsystems due to uncertainties of inventory status. The results obtained in this paper help us understand the dynamic complexities of supply chain system and provide guidelines for selecting decision parameters to improve overall performance.     

Managing the order pipeline to reduce supply chain volatility

The bullwhip effect in particular, and supply chain volatility in general, has been the subject of much analytical and empirical investigation by researchers. One goal of this work has been to determine supply chain designs and policies that minimize volatility. Using a system dynamics approach, we use three distinct supply chain volatility metrics to compare the ability of two alternative pipeline inventory management policies to respond to a demand shock. The results indicate that no one policy dominates on all three metrics of supply chain volatility. A simplistic static pipeline policy minimizes the bullwhip effect and lessens the likelihood of on-hand inventory oscillations, while a more sophisticated dynamic pipeline policy may converge more rapidly to the new equilibrium. In addition, simulation results suggest that the dynamic policy provides better customer service through fewer stockouts and backorders.