基于深度强化学习的生鲜产品联合库存控制与动态定价研究

针对生鲜产品的易逝性特征以及复杂多变的现实环境导致生鲜产品的最优订货和定价策略难以获得的问题,提出了基于深度强化学习方法的生鲜产品联合库存控制与动态定价方法,结合生鲜产品特性对问题进行建模并定义为马尔可夫决策过程,然后基于深度强化学习设计了生鲜品联合库存控制和动态定价算法。实验结果表明,基于深度强化学习的联合库存控制和动态定价策略收益表现最佳,因此,基于深度强化学习的联合库存控制和动态定价研究能够提高企业收益,有效促进强化学习在收益管理领域的落地,具有实际应用价值。     

基于动态定价策略的数据中心能耗成本优化

当前,云数据中心的能耗问题已成为业界关注的热点.已有研究工作大多致力于从技术角度降低数据中心的能耗,或在能耗与性能之间寻求一种最佳的折衷.云计算作为一种商业计算模式,已有研究成果很少考虑到云定价策略对能耗管理机制的影响.文中提出了基于动态定价策略的数据中心能耗成本优化方案.建立起服务价格和能耗成本的统一模型,通过研究两者之间的关系.协同优化服务价格与能耗成本,使数据中心的收益达到最优.鉴于数据中心规模庞大、承载任务繁重等特点,论文采用基于重载近似的大规模排队系统来对数据中心建模,根据不同数据中心间的服务需求量和电价差别,设计了多数据中心间的负载路由机制,旨在削减数据中心的整体能耗成本.针对单个数据中心,定义了双阈值策略以动态调节服务器的各种状态(On/Off/Idle等),从而使数据中心能耗成本得到进一步优化.实验结果表明,论文提出的解决方案能够在满足用户QoS需求的前提下,较好地优化数据中心能耗成本,同时使数据中心的收益达到最优.     

基于Simulink的库存系统的动态建模与仿真

该文首先通过将随机的、不可控的需求控制问题转化为安全库存设置问题的方法,建立了库存系统的仿真模型。然后基于MATLAB／Simulink软件对多组策略值进行仿真,动态实时模拟了（t,s,S）策略的库存系统的入库、出库时机和数量。最后选择库存总费用的日平均值最低的一组为最优库存策略,从而使企业能够合理规划和安排仓库及各类设施、资源,文现库存成本的最小化。仿真结果清晰、直观,为现代经营者做出正确的库存决策提供了有价值的参考。     

不确定环境下的一种动态响应库存优化方法

针对随机需求提前期环境下的库存管理问题,提出一种启发式动态响应算法求取最优订货策略。建立最优订货策略的动态非线性优化模型,并设定客户需求和订货提前期分别为线性和高斯随机变量,通过变化形式、步长和变化频率的不同模拟实际经济运营过程;在微粒群寻优过程中引入柔性变异概率及动态更新响应方式,使微粒具有感知外界环境变化及对变化的响应能力,提高算法对复杂动态系统环境变化的适应性。实证分析结果证明了所提方法对最优订货量实时变化的动态响应能力。     

考虑体验服务和促销努力的时尚服装库存与定价优化决策

时尚服装受季节性、时尚流行性等因素的影响,其时尚度随时间衰减,致使市场价值不断降低,造成服装行业一直面临高库存问题.鉴于此,研究需求依赖于价格和时尚程度的服装库存模型,考虑体验服务投资可提高服装时尚度、促销降价可改善销售需求的作用,构建一般情形、体验服务投资情形、促销努力情形以及服务与促销联合4种情形下的优化订货和定价模型,分析体验服务投资和促销努力联合时对服装零售商优化决策的影响.研究结果表明:体验服务投资对边际效益的影响是先增后减;服装零售商通过决策折扣促销力度和延后折扣时间点,优化库存水平和定价,可降低库存成本,增加利润;体验服务和促销努力联合作用时,产生利润互补或替代效应,这取决于服装时尚度衰减因子,当其大于某临界值时,体验服务和促销努力两者为互补关系,反之则为替代关系.     

库存水平影响需求下变质品订购、定价和保鲜技术投资的联合决策

研究变质品的订购、定价和保鲜技术投资联合决策问题.假设产品需求同时受价格和库存水平的影响,系统不允许缺货并放松期末库存水平为零的约束,零售商拥有有限的货架空间或存储空间,同时考虑零售商可以通过投资保鲜技术减低产品的变质率,以平均利润最大化为目标构建库存水平影响需求下变质品的订购、定价和保鲜技术投资联合决策模型.首先证明最优策略的存在性和唯一性,并给出零售商建立期末库存的条件;然后利用最优解的相关性质设计一个求解模型的多阶段迭代算法;最后通过具体算例验证展示模型和算法的可行性和实用性,并完成相关参数的敏感性分析,获得一定的管理启示.     

基于连续性库存检查策略的库存系统仿真优化

库存能够保证企业的正常运作,但过多的库存又导致库存成本增加、占用大量流动资金．因此,有必要在库存持有量和成本之间寻求一个平衡点．利用AnyLogic软件构建了单一库存点的库存仿真系统．以连续性库存检查策略为例进行仿真与优化实验,得出最优库存控制策略,在保证客户服务水平的基础上实现了库存成本最低．     

时变需求下生产与库存系统的最优控制决策方法

针对供应链中库存随着需求的变化可能导致的积压和对生产(或采购)产生的不利影响,为了更好地协调生产(或采购)并减少产品库存,本文研究了以生产为中心的一类基于库存约束和动态线性时变需求下的多品种、多周期、多循环的生产与库存的最优控制模型,结合应用最优控制理论,给出了一种采用切比雪夫多项式逼近和高斯.切比雪夫数值积分对最优控制问题进行数值求解的方法.最后,对某一时变需求情况下的模型应用MATLAB软件进行了求解,得到了生产(采购)与库存的最优控制策略,有效地保证了供应链系统的持续稳定的循环.     

基于强化学习的动态定价策略研究综述

随着人工智能技术的不断发展,其应用领域也逐渐拓宽。为了能进一步将强化学习技术应用于动态定价领域,构建智能动态定价系统,对与动态定价相关的强化学习技术进行介绍,将已有研究从供应商数目(单供应商、多供应商)、所处环境模型(MDP、POMDP、Semi-MDP)、选用算法(Q-Learning, SARSA, Monte-Carlo)等方面进行综述。基于综述内容提出未来智能动态定价的研究方向。     

基于客户感知价值的云计算服务动态定价策略优化方法

论文参考借鉴客户感知价值理论和人工免疫系统的进化思想,提出了"客户感知价值导向"的定价策略,并给出不确定云计算服务复杂环境下的客户感知价值模型及评价方法;然后采用情感倾向性分析技术设计客户感知价值动态识别与抽取算法,实现对客户网络评价服务产品信息动态抽取和实时挖掘,并随时量化跟踪客户态度倾向趋势,为解决客户感知价值的动态识别、评价方法的随机鲁棒优化问题将提供一种新的思路和解决方案。     

A Link-Node Discrete-Time Dynamic Second Best Toll Pricing Model with a Relaxation Solution Algorithm

Dynamic congestion pricing has become an important research topic because of its practical implications. In this paper, we formulate dynamic second-best toll pricing (DSBTP) on general networks as a bilevel problem: the upper level is to minimize the total weighted system travel time and the lower level is to capture motorists’ route choice behavior. Different from most of existing DSBTP models, our formulation is in discrete-time, which has very distinct properties comparing with its continuous-time counterpart. Solution existence condition of the proposed model is established independent of the actual formulation of the underlying dynamic user equilibrium (DUE). To solve the bilevel DSBTP model, we adopt a relaxation scheme. For this purpose, we convert the bilevel formulation into a single level nonlinear programming problem by applying a link-node based nonlinear complementarity formulation for DUE. The single level problem is solved iteratively by first relaxing the strick complementarity by a relaxation parameter, which is then progressively reduced. Numerical results are also provided in this paper to illustrate the proposed model and algorithm. In particular, we show that by varying travel time weights on different links, DSBTP can help traffic management agencies better achieve certain system objectives. Examples are given on how changes of the weights impact the optimal tolls and associated objective function values.

一种需求率随时间变化且允许缺货的库存模型

针对库存管理和控制中存在的实际问题,在最大程度降低库存运营成本的前提下,综合考虑了货物损坏率以及缺货率引起的损失等多种影响,提出了一个需求率为一般连续函数的库存模型,并进一步建立了该模型的两阶段优化方法.该方法通过简单的求解过程,就能够找到最优订货策略,从而有效的降低了库存成本.最后运用一个简单的数据实验,证明了该模型和方法的有效性和准确性.该库存成本函数模型及其求解方法也可以适用于需求率是非单调的随时间变化函数的模型.     

Maximizing profits in an inventory model with both demand rate and holding cost per unit time dependent on the stock level

We study an EOQ inventory model with demand rate and holding cost rate per unit time, both potentially dependent on the stock level. The ordering cost, the holding cost and the gross profit from the sale of the item are considered. The objective is to maximize the average profit per unit time. We present the analytical formulation of the problem and demonstrate the existence and uniqueness of the optimal cycle time, giving a numerical algorithm to obtain it. Moreover, we provide two fundamental theoretical results: a rule to check when a given cycle time is the optimal policy, and a necessary and sufficient condition for the profitability of the system. Several EOQ models analyzed by other authors are particular cases of the one here studied. We present some numerical examples to illustrate the proposed algorithm and analyze the sensitivity of the optimal solution with respect to changes in various parameters of the system.     

随机供应中断和退货环境下库存问题的建模与控制

供应中断和退货的不确定性,导致企业库存控制异常困难.然而,由于系统存在多个随机因素,采用传统方法很难构建模型,所以,以往均将供应中断和退货分开考虑.为综合分析它们对库存的影响,本文在采用布朗运动描述库存水平动态变化过程条件下,利用连续时间Markov链、更新过程和鞅理论,构建了系统期望折扣总费用模型,并设计了交叉熵法确定最优库存策略.最后,通过仿真实验,分析了供应中断、净需求和系统参数,对最优控制策略和期望费用的影响.     

Exact formulations and algorithm for the train timetabling problem with dynamic demand

In this paper we study the design and optimization of train timetabling adapted to a dynamic demand environment. This problem arises in rapid train services which are common in most important cities. We present three formulations for the problem, with the aim of minimizing passenger average waiting time. The most intuitive model would consider binary variables representing train departure times but it yields to non-linear objective function. Instead, we introduce flow variables, which allow a linear representation of the objective function. We provide incremental improvements on these formulations, which allows us to evaluate and compare the benefits and disadvantages of each modification. We present a branch-and-cut algorithm applicable to all formulations. Through extensive computational experiments on several instances derived from real data provided by the Madrid Metropolitan Railway, we show the advantages of designing a timetable adapted to the demand pattern, as opposed to a regular timetable. We also perform an extensive computational comparison of all linear formulations in terms of size, solution quality and running time.     

The optimal integrated inventory policy with price-and-credit-linked demand under two-level trade credit

In this paper, we proposed a generalized, integrated, supplier–retailer inventory model using a trade credit policy. The trade credit policy adopted here is a two-level trade credit policy in which the supplier offers the retailer a permissible delay period M, and the retailer in turn provides customers a permissible delay period N. Cases where M > N and M ? N are explored thoroughly. In addition, the demand rate is assumed to be a function of both retail price and the customers’ credit period. Consequently, this paper deals with the problem of determining the optimal retail price, economic order quantity, and the number of shipments from the supplier to the retailer in one production run for an integrated inventory system under both two-level trade credit and price-and-credit-linked demand rate. Algorithms are developed in order to determine the joint optimal policies. Numerical examples are presented to illustrate the proposed models, as well as sensitivity analysis of key parameters.     

价格影响需求的易变质产品动态定价模型

在产品销售价格影响需求的条件下,利用最优控制理论建立了易变质产品的动态定价模型,目标是最大化产品销售周期内总的销售利润。利用Pontryagin 最大值原理得到了产品销售价格的最优性条件。通过对模型的理论分析得出如果产品销售价格介于单位产品购买费用和产品销售价格上限之间,且产品库存在销售周期结束之前始终为正时,销售周期内各时刻的产品最优销售价格一定大于与相应时刻变质率和产品单位库存成本有关的一个下界,销售周期内各时刻的产品最优库存水平一定小于与相应时刻变质率和产品单位库存成本有关的一个上界。     

Particle Swarm Optimization and Pure Adaptive Search in Finish Goods’ Inventory Management

Abstract

Inventory management deals with a tradeoff between the benefits of keeping stocks of goods that allows fulfillment of the customer’s demand, and the cost of carrying inventory. Inventory control techniques are very important components and the most organizations can substantially reduce their costs associated with the flow of materials. This paper presents new inventory management model based on particle swarm optimization and pure adaptive search global optimization algorithm in production-inventory system. The proposed model is focusing on planned level of demand for finished goods, production and raw materials cost, production capacity as the norm, change of the production cost and inventory capital cost, all of which are typical factors in automobile manufacture industry. The model determines different factors such as the minimizing inventory quantity, minimizing inventory value, and minimizing production cost based on demand for production items. The model is tested with original real-world dataset obtained from the automotive company Lear from US and its factory in Novi Sad, Serbia.     

考虑时变奖励的多摇臂算法在动态定价中的应用

考虑到动态定价是一个非固定性的多摇臂(Multi-Armed Bandit,MAB)问题,即厂商的利润会随时间变化,因此在相关研究基础上,研究了需求不确定情况下考虑时变奖励的置信区间上界(Upper Confidence Bound,UCB)算法在动态定价问题上的应用.将商品定价问题描述为一个多摇臂问题,并构建利润最大...