需求信息部分已知下生鲜品的低碳订购策略研究

为了研究需求信息缺失和低碳减排下的生鲜品企业运营决策问题,基于单周期随机库存系统,分别建立受碳限额与交易政策规制和无碳约束的分布式鲁棒优化模型,通过极大极小期望利润准则和最优化方法求解出两种情形下生鲜品的最优订购量,并运用数值算例检验需求随机因子的标准差系数和碳排放权交易价格对生鲜品订购量、利润和碳排放的影响。结果表明,存在唯一的最优库存因子使得生鲜品企业在最坏分布下的期望利润取得最大;与不受碳政策规制情形相比,生鲜品企业在碳限额与交易政策下能够实现高利润和低排放;需求信息缺失对生鲜品企业在碳限额与交易政策下期望利润的影响小于不受碳政策规制情形。     

基于随机替代关系的单周期可替代品库存模型

基于确定型的可替代产品单周期库存模型已有研究,它假定替代数量关系是确定的。而事实上,替代过程中由于消费者需求的个性化因素影响,替代的数量关系往往是随机的。为此,假定替代数量关系符合随机分布条件下,提出了两类可替代产品的单周期库存模型,并设计了基于遗传算法的仿真优化算法。算例分析的结果表明,考虑随机替代关系可显著地提高决策者的收益;该算法是有效的。     

基于延期支付的非立即变质物品的库存模型

一类短质期变质产品,如生鲜蔬菜、水果等,增加其库存展示量能为顾客提供更多的拣选机会,从而刺激顾客增加购买量.对此,在供应商给予零售商固定延期支付期限且在综合分析已有变质物品的库存模型的基础上,建立了一个更能准确反映当前存货水平和需求率的库存模型.模型中以零售商的年总费用最小为目标讨论了模型的最优解,通过模型的分析求解,得到零售商的最优订货周期及最优订货量的简单判定方法.通过具体算例,结合灵敏度分析,分别分析了物品的固定保鲜期、物品的变质率及供应商给与零售商的延期支付期限对零售商最优订货策略的影响.     

基于初始新鲜度的生鲜农产品订货策略

目的以初始新鲜度为基准,结合保鲜包装成本构建订货模型来优化生鲜农产品零售商的订货策略。方法以单个销售周期内零售商利润最大为目标,在易腐商品库存模型中引入初始新鲜度,结合保鲜包装成本,构建单周期生鲜农产品的订货模型,借助Matlab对模型进行优化求解。结果通过算例可知,当销售周期为30 d,保鲜包装成本为0.15元,初始新鲜度为0.87时,取得最大利润值311.99元,且初始新鲜度由0.6上升到0.95期间,最优订货周期由2.86d上升到23.66d,最优订货量由74.76kg上升到120.60 kg,再回落到101.37 kg;而保鲜包装成本由0.05元上升到0.25元,利润最大值由287.44元上升到327.13元,且初始新鲜度越大,利润上升幅度越高。结论初始新鲜度较高时,零售商可以通过提高保鲜包装成本及采取少批次多批量的订货策略来提高企业的利润;而初始新鲜度较低时,在食品安全的范围内,零售商可以通过降低保鲜包装成本和采取多批次少批量的方式来提高利润。     

基于设备使用周期的备件库存控制模型研究

由于设备系统的复杂性、设备故障的随机性,备件的需求难以准确确定,相关费用也难以做出准确评估,从而使得备件这一类库存控制问题变得相当复杂.基于备件的需求量与设备使用周期密切相关这一关键点,从设备的整个使用周期入手,结合设备的使用情况,在备件需求率随时间变化的情况下,对备件库存控制问题进行了研究,建立了在设备整个使用周期内使备件总订货成本和存储费用最低的数学模型,给出了备件最佳订货次数、订货时间和订货批量的计算公式.最后给出了一个算例以证明模型的实用性.     

供应链环境下的供应商管理库存模型研究

基于文献[8]提出的供应商管理的订货和派送库存模型存在的参数设计过于复杂、储存成本计算有欠商榷的缺点,重新确定储存成本,并删除了对库存没有实质影响的部分参数,提出了既考虑顾客需要,又考虑供应商运输成本规模经济的固定周期检查派送和存货补足库存模型,给出了该模型的全局最优解计算,最后给出算例分析,并同经济定货模型进行比较。     

需求更新条件下零售商双渠道订货策略研究

在快速反应系统下,要求零售商可以最大化满足市场的需求。本文考虑零售商存在2种订货方式,即线上订货和线下订货并假设线下订货可以立即送达;分别建立了这2种方式下的利润模型。分析了双渠道订货模式下零售商可以满足市场最大化需求但是会使自身利润降低,并通过需求更新模式分析研究了单渠道和双渠道订货模式的选择问题来实现零售商的利润最大化。最后通过算例验证模型的正确性。     

单周期随机需求下供应链库存协调机制设计

研究了单周期随机需求条件下两级供应链的库存协调机制设计问题.建立了基于弹性订货策略的供应链库存风险共担机制,并通过算例验证了弹性订货策略在供应链库存协调中的有效性.研究结果表明,在供应链上下游无合作时,需方承担全部库存风险,其个体理性的决策无法实现供应链的整体优化,而如果双方实施弹性订货策略,分担库存风险,则可以实现供应链的Pareto优化.     

空间需求聚集下安全库存模型及其实证

用定量的分析方法,对安全库存模型进行了改进.讨论了当空间需求的相关系数变动时,安全库存的成本是如何变化的.当各地区需求不是呈线形正相关的情况下,建议零售商建立中心仓库.安全库存管理不再是各自为政的独立运营过程,而是利用聚集经营去给零售企业带来更大的利润.     

供应链上游层面VMI模式的利益分配机制

选取供应链上游层面VMI模式为研究对象,考虑实施VMI后,订货主动权的转移导致订货策略的变化,以及制造商对各零部件需求速率的不同导致其对各零部件的订货周期也不一致的情况,通过建立经济效果模型,探讨上游层面VMI对供应链整体利益的影响.在此基础上,根据供应商与制造商在合作中经常出现的3种情境——平等、供应商占优、制造商占优,建立VM1模式下不同情境的利益分配机制.该机制最终可体现在供应商所售产品的重新定价上,此定价既能弥补供应商因管理制造商库存而增加的成本,又可使制造商的利润得到保障,达到共赢的目的,为上游层面VMI模式的顺利推广提供理论依据.     

多级库存控制的连锁经营企业配送网络设计

研究了随机需求条件下连锁经营企业配送网络设计及其库存决策的联合优化问题.详细分析了基于POT(power of two)多级库存控制策略的连锁企业多级工作库存及订货成本,给出了门店及配送中心在满足给定服务水平条件下的安全库存成本.在综合考虑运输成本和配送中心选址成本的基础上,建立了以系统总成本最小为目标的配送系统总成本优化模型,并采用遗传算法求解该优化模型,在得到最优配送网络设计方案的同时,确定了配送中心订货周期及门店配送周期.通过算例验证了模型及算法的有效性,并分析了需求、运输距离和选址成本等因素的变化对系统总成本的影响,为连锁经营企业的物流配送网络设计及库存控制提供决策支持.     

基于需求依赖库存的变质性商品最优库存-配送模型

超市型销售终端中商品展示水平的“选择效应”和“广告效应”,使得商品需求在很大程度上依赖其库存展示水平。针对这一实际情形,假设销售商在此类终端销售某一变质性商品,其中需求线性依赖于该商品当前库存水平,当库存水平下降到一定量时,立即从配送中心运送商品对超市库存进行补充。以单位平均利润最大化为目标构造相应的库存-配送模型,使用最优化理论讨论了模型最优解的存在性及唯一性,并提供寻求模型整体最优解的算法,以此求出最优的库存与配送方案。最后给出了数值例子,分析了模型参数变化对系统最优策略和利润的影响。     

A stochastic inventory model for an immediate liquidation and price-promotion decision under price-dependent demand

We consider a periodic review stochastic inventory system where the current on-hand inventory exceeds the maximum supply needs in the future. Consequently, one must make an immediate inventory liquidation decision on the liquidation quantity and promotional price with the goal of maximising the overall profit where the demand during the liquidation period (DDLP) is a random variable whose distribution depends on the promotional price. We develop a price-dependent DDLP model and an inventory model for optimising the liquidation quantity and unit promotional price. The model is applicable for general distributions of the DDLP and regular demand (i.e. demand during the future periods following the promotion period). We also investigate four special cases where the DDLP and regular demand are assumed to be either exponential or uniform random variables. The two models that assume the exponential distribution for regular demand can be examined analytically and simplified using the mathematical properties we derive. The additional two models that assume the uniform distribution for regular demand do not have closed-form expressions but can be solved numerically. Some numerical examples are presented for further elaboration of the models and to demonstrate their practical use.     

Inventory performance of some supply chain inventory policies under impulse demands

This paper attempts to study the impact of impulsive demand disturbances on the inventory-based performance of some inventory control policies. The supply chain is modelled as a network of autonomous supply chain nodes. The customer places a constant demand except for a brief period of sudden and steep change in demand (called demand impulse). Under this setting, the behaviour of each inventory policy is analysed for inventory performance of each node. It is found that the independent decision-making by each node leads to a bullwhip effect in the supply chain whereby demand information is amplified and distorted. However, under a scenario where the retailer places a constant order irrespective of the end customer demand, the inventory variance was actually found to decrease along the supply chain. The variance of the inventory remained constant along the chain when only the actual demands are transmitted by each node. The results also showed that the inventory policy which is best for one supply chain node is generally less efficient from a supply chain perspective. Moreover, the policy which performs poorly for one node can be most efficient for the supply chain. In a way, our results also provide a case for coordinated inventory management in the supply chain where all members prepare a joint inventory management policy that is beneficial for all the supply chain nodes. The results have significant industrial implications.     

Decision support system for vendor managed inventory supply chain: a case study

Vendor-managed inventory (VMI) is a widely used collaborative inventory management policy in which manufacturers manages the inventory of retailers and takes responsibility for making decisions related to the timing and extent of inventory replenishment. VMI partnerships help organisations to reduce demand variability, inventory holding and distribution costs. This study provides empirical evidence that significant economic benefits can be achieved with the use of a genetic algorithm (GA)-based decision support system (DSS) in a VMI supply chain. A two-stage serial supply chain in which retailers and their supplier are operating VMI in an uncertain demand environment is studied. Performance was measured in terms of cost, profit, stockouts and service levels. The results generated from GA-based model were compared to traditional alternatives. The study found that the GA-based approach outperformed traditional methods and its use can be economically justified in small- and medium-sized enterprises (SMEs).     

需求受相关性影响的设备备件库存优化

备件多级库存模型通常基于备件需求相互独立的假设，但随着库存系统层次的增加以及协同管理方式的应用，备件需求的相关性将显著影响库存优化决策。针对需求具有相关性的备件库存问题，以服务响应时间为约束条件，以库存成本及缺货成本最小化为目标建立备件两级库存决策模型。引入Nataf概率变换法，利用已知的备件需求边缘概率密度函数构造满足特定相关性条件的随机需求样本，并将蒙特卡洛仿真与遗传算法相结合求解最优库存分配方案。仿真算例证明，设备备件库存的最优决策随着需求相关性系数的增大而发生变化，根据需求相关性的变化适当地调整库存决策，有利于降低备件库存系统总成本，提高库存系统对顾客需求的响应能力。     

Analysis of production-inventory system for deteriorating items with demand disruption

We develop a production-inventory model for deteriorating items with demand disruption. This differs from the conventional inventory model which is affected only by either deterioration or disruption. In real-life production-inventory systems, deterioration of products and demand disruption are inevitable. The objective of this paper is to address these issues and to be able to derive the optimal production run time and replenishment policy for spot market purchases. We divide the problem into different scenarios according to disruption's time and magnitude. In each scenario the optimal production and inventory plans are provided so that the manufacturer can satisfy the new demand and decrease the loss. Then a numerical example is used to illustrate the model.     

需求正态分布下包装延迟策略对库存影响的研究

探讨了在需求正态分布下不同包装策略下的库存量计算方法,并对相同的需求不确定性、需求相关系数、包装分布下包装延迟策略对库存的影响进行了分析比较.通过比较发现包装延迟策略在客制化需求日益增加的情况下,包装延迟策略能在一定程度下降低库存,为企业带来一定的收益.     

Inventory lot-sizing with supplier selection under non-stationary stochastic demand

Inventory lot-sizing and supplier selection problem has been studied in the literature considering mainly time-varying deterministic demand. However, in real life, most of the products exhibit non-stationary stochastic demand. In this context, we propose an integer linear programming model for inventory lot-sizing and supplier selection problem under non-stationary stochastic demand with all-units quantity discounts and fill rate constraints. Through detailed analysis of experimental results, we show the impacts of fill rate requirements and demand coefficient of variation on costs, inventory levels and order allocations.     

Innovative vendor-managed inventory strategy exploiting interconnected logistics services in the Physical Internet

Traditional supply chain networks are often designed in the interests of a company. Once the network has been defined, the storage and distribution of goods are usually fixed and restricted within the network. This is assumed to be an inherent limit of current inventory control research. Instead of specialised hierarchical storage networks, this paper proposes an innovative vendor-managed inventory strategy exploiting the Physical Internet (PI), which is an open, universal, interconnected logistics system. In such a system, facilities and means of transport are shared and can be allocated according to demands of users. As a result, the PI allows users to stock anywhere in the network and also provides open multisourcing options for orders with on-demand warehousing services within the PI. Inventory decisions can be made dynamically by each player to minimise networkwide inventory levels. A non-linear, simulation-based optimisation model was developed for the vendors’ inventory decision-making when confronted with stochastic demands. A metaheuristic using simulated annealing was applied to solve the problem, and then, the optimised inventory decisions were validated using simulation. The results suggest that the proposed PI inventory model can reduce the total logistics cost while maintaining a comparable or better level of end customers’ services.