基于多摇臂赌博机的产品定价算法

针对在线零售商在不完全需求信息下的单产品定价问题,提出了一种基于多摇臂赌博机的产品定价算法。为了提升多摇臂赌博机算法在定价问题中的效果,该算法利用了需求曲线的单调性,并加入了消费者偏好识别。对消费者的保留价格进行分析得到消费者购买概率,将在线零售商的定价问题建模为多摇臂赌博机模型,给出了相应的定价算法并进行了理论分析,最后通过仿真实验比较了相关算法的定价效果。仿真结果表明该算法提高了在线零售商的收益。     

时变系统最小均方算法的性能分析

在无过程数据平稳性假设和各态遍历等条件下,运用随机过程理论研究了最小方算法（LMS）的有界收敛性,给出了估计误差的上界,论述了LMS算法收敛因子或步长的选择方法,以使参数估计误差上界最小。这对于提高LMS算法的实际应用效果有着重要意义。LMS算法的收敛性分析表明：（1）对于确定性时不变系统,LMS算法是指数速度收敛的;（2）对于确定性时变系统,收敛因子等于1,LMS算法的参数估计误差上界最小;（3）对于时变或不变随机系统,LMS算法的参数估计误差一致有上界。     

模糊测试变异策略优化方法

基于覆盖引导的模糊测试技术是当前研究的热点,AFL是该领域的代表性工具。文章在AFL基础上进行了改进,将Havoc的变异策略建模为一个多臂赌博机问题,并提出一种结合ε-greedy算法和置信区间上界（Upper Confidence Bound,UCB）算法的动态调整变异策略的方法,实现了工具EnAFL的设计。通过与AFL的对比实验分析得出,EnAFL在代码覆盖率和测试效率方面表现更出色。     

考虑交通时变特性的城乡快递无人机配送优化研究

针对农村快递网点运营成本高、网点建设滞后导致的电商物流配送成本高问题,提出了城乡客运班车+无人机的快递配送模式。在考虑了配送过程中路网交通的时变特性的情况下,以无人机—车辆配送系统总成本最小为优化目标,建立了时变网络下带时间窗的无人机—车辆路径问题（TDVRPDTW）模型,并提出一个由基于最近邻思想的改进CW算法和动态规划启发式算法构成的两阶段启发式算法来求解TDVRPDTW。最后,通过算例求解验证构建模型的合理性和求解算法的有效性,为制定农村物流配送的城乡客运班车+无人机快递配送方案提供决策支持。     

易腐商品定价模型及其粒子群解法

论文研究了易腐商品的定价问题,并基于随机需求分布假设和利润最大化原则提出了一种新的易腐商品最优定价模型。多个随机分布的存在使该模型比传统模型更加复杂,因此难以使用常规函数极值法获得解析解,论文尝试将粒子群优化算法引入该模型进行演化求解。最后给出的算例分析表明:该模型和算法可以快速有效地解决随机需求情况下的易腐商品最优定价问题。     

量子粒子群算法在易逝品多目标定价中的应用

对易逝品的多目标定价问题进行了研究。从利润最大化角度建立易逝品多目标最优定价模型。模型中涉及复杂的需求函数,常规函数极值法不易获得问题解析解,因此引入量子粒子群算法,结合惩罚函数对模型进行演化求解。根据给出的算例分析表明,利用量子粒子群算法,可以快速有效地得到不同订货量下的最优定价与折扣价组合。     

模糊随机环境下考虑残次品处理的库存配送问题研究

针对考虑残次品的多生产商选择多商品多阶段库存配送问题,建立了一个基于动态规划的双层库存配送模型。高端物流服务集成商以整个供应链网络成本最小为目标制定采购决策;库存配送服务商以运营成本最小为目标,在集成商决策下制定库存和配送决策。设计了模糊随机环境下基于动态规划的双层全局-局部-邻域粒子群算法（Bi-DPGLNPSO）对模型进行求解。并通过算例验证模型和算法的有效性和合理性。通过参数测试和算法对比检验算法的优越性。     

动态多路径选择的混合演化算法

动态路径诱导系统（DRGS）是智能运输系统（ITS）研究的一个重要内容,动态路径诱导算法要考虑到全局最优和实时性问题。因此建立了一种包含实时路网信息而且可以针对时间进行离散化处理的路网模型,同时提出了一种用改进的Ford最短路径算法来初始化种群的演化算法,并设计了一组特定的演化算子（选择、交叉、变异）,来求解动态路径诱导系统中的“多准最优路径”。最后,通过数值实验表明了此算法的可行性和有效性。     

基于连续时间模型的多智能体系统采样数据一致性

讨论了一般线性模型的多智能体系统具有时变采样间隔的采样数据一致性问题.首先基于连续时间模型,利用采样数据的离散时间特性分析时变采样间隔允许的上界.由于不考虑采样间隔之间的状态,Lyapunov函数仅需要在每个采样时刻保证递减.由此得到了一个利用线性矩阵不等式求解更低保守性的时变采样间隔上界的方法.接着通过参数化矩阵变量得到了基于线性矩阵不等式的控制器设计方法.最后数值仿真展示了理论结果的正确性.     

基于贝叶斯网络理论的时变混合盲源分离算法

以状态空间模型作为信道的变化模型，研究了时变混合情况下非平稳信号的盲分离问题。首先将隐马尔可夫模型（HMM）和混合高斯（MOG）模型结合起来对具有动态结构和复杂分布的非平稳源信号进行建模，然后运用贝叶斯网络理论处理信道时变情况下独立成分分析（ICA）模型中各变量和参数之间的关系，提出了一种基于贝叶斯推断的可同时完成混合矩阵盲估计及源信号盲分离的算法，通过采用逼近方法有效地减小了算法计算量。计算机仿真试验证明本文算法的有效性。     

Dynamic pricing,promotion and replenishment policies for a deteriorating item under permissible delay in payments

We study the problem of dynamic pricing, promotion and replenishment for a deteriorating item subject to the supplier's trade credit and retailer's promotional effort. In this paper we adopt a price- and time-dependent demand function to model the finite time horizon inventory for deteriorating items. The objective of this paper is to determine the optimal retail price, the promotional effort and the replenishment quantity so that the net profit is maximized. We discuss the properties and develop an algorithm for solving the problem described. The numerical analyses show that an appropriate promotion policy can benefit the retailer and that the promotion policy is important, especially for deteriorating items. Furthermore dynamic decision-making is shown to be superior to fixed decision-making in terms of profit maximization. Some special cases, such as with no credit period and for non-deteriorating items, are discussed as is the influence of the time-varying demand, the rate of deterioration and the credit period on the retailer behavior.     

动态图像特征空间的立体视觉匹配算法

提出图像特征空间概率分布参数时变的立体视觉匹配问题。立体视觉匹配算法通常是针对时不变提出的。因此,对于概率分布时变的图像特征空间而言,这些方法均不能有效地实现动态立体视觉匹配。针对这一问题,提出的方法是对图像分割边缘构建动态贝叶斯网,结合贝叶斯学习并充分利用其学习所获得的图像空间概率模型变化演进的规律,得到较准确、平滑地图分割的动态结果,以此作为基元间的对应性,在边缘区域进行置信传播,实现图像的动态立体视觉匹配。     

云环境下基于组合双向拍卖的动态资源定价

云环境下的市场交易机制缺乏灵活性,且在某些情况下定价不合理。为此,提出一种基于组合双向拍卖的动态资源定价模型,给出云资源分配与定价算法,用户通过响应时间出价,资源提供商根据负载情况要价。仿真实验结果表明,该算法与固定比例的定价算法相比,能提高18%的用户利益与9%的资源提供商利益。     

CR系统中基于NPGP的功率控制算法

为解决认知无线电（cR）系统中认知用户对主用户的干扰及系统功率控制问题,在介绍认知无线电原理及其基本在线认知任务循环框图的基础上,将DavidGoodman等人提出的基于非合作带竞价博弈的功率控制算法应用到认知无线电系统中。与传统的功率控制算法相比,该算法不仅降低了认知无线电系统中认知用户的发射功率,而且降低了认知用户之间的干扰,提高了系统的整体性能。最后通过MATLAB仿真分析比较了带价格因子和不带价格因子两种情况下的性能。     

Pricing and production lot-size/scheduling with finite capacity for a deteriorating item over a finite horizon

Although the lately evolved manufacturing technologies such as enterprise resource planning (ERP) provide a unified platform for managing and integrating core business processes within a firm, the decision-making between marketing and production planning still remains rather disjoint. It is due in large parts to the inherent weaknesses of ERP such as the fixed and static parameter settings and uncapacitated assumption. To rectify these drawbacks, we propose a decision model that solves optimally the production lot-size/scheduling problem taking into account the dynamic aspects of customer's demand as well as the restriction of finite capacity in a plant. More specifically, we consider a single product that is subject to continuous decay, faces a price-dependent and time-varying demand, and time-varying deteriorating rate, production rate, and variable production cost, with the objective of maximizing the profit stream over multi-period planning horizon. We propose both coordinated and decentralized decision-making policies that drive the solution of the multivariate maximization problem. Both policies are formulated as dynamic programming models and solved by numerical search techniques. In our numerical experiments, the solution procedure is demonstrated, comparative study is conducted, and sensitivity analysis is carried out with respect to major parameters. The numerical result shows that the solution generated by the coordinated policy outperforms that by the decentralized policy in maximizing net profit and many other quantifiable measures such as minimizing inventory investment and storage capacity.Scope and purposeWe consider a manufacturing firm who produces and sells a single product that is subjected to continuous decay over a lifetime, faces a price-dependent and time-varying demand function, shortages are allowed and a completely backlogged, and has the objective of determining price and production lot-size/scheduling so as to maximize the total profit stream over multi-period planning horizon. We develop a tactical-level decision model that solves the production scheduling problem taking into account the dynamic nature of customer's demand which is partially controllable through pricing schemes. As analogous to the sales and operations planning, the proposed scheme can be used as a coordination center of the APS system within a generic enterprise resource planning framework which integrates and coordinates distinct functions within a firm.This paper differs from the existing works in several ways. First, we propose a dynamic version of the joint pricing and lot-size/scheduling problem taking into account the capacitated constraint. Second, several key factors being considered in the model, such as the demand rate, deteriorating rate, production rate, and variable production cost are assumed time-varying that reflect the dynamic nature of the market and the learning effect of the production system. A third difference between the past research and ours is that the price can be adjusted upward or downward in our model, making the proposed pricing policy more responsive to the structural change in demand or supply.     

云环境下基于SLA的优化资源分配研究

针对云计算环境下如何高效分配资源,实现资源供应者利润最大化这一难题,提出了一种基于服务级别协议（SLA）的动态云资源分配策略。该策略通过将SLA中的计算力、网络带宽、数据存储等属性作为优化参数,构造了一种服务请求与资源的映射模型,同时设计相应的效用函数,并结合改进的与模拟退火算法相融合的混合粒子群算法（SA-PSO）,实现云环境下的优化资源分配。实验分析结果表明,基于SLA参数的SA-PSO算法具有更好的全局最优值,在给定虚拟资源相同情况下,调用该算法完成用户任务实现的利润更高。     

Optimal dynamic pricing for deteriorating items with reference-price effects

In this paper, a dynamic pricing problem for deteriorating items with the consumers’ reference-price effect is studied. An optimal control model is established to maximise the total profit, where the demand not only depends on the current price, but also is sensitive to the historical price. The continuous-time dynamic optimal pricing strategy with reference-price effect is obtained through solving the optimal control model on the basis of Pontryagin's maximum principle. In addition, numerical simulations and sensitivity analysis are carried out. Finally, some managerial suggestions that firm may adopt to formulate its pricing policy are proposed.     

多类扰动下电子侦察卫星动态调度问题研究

针对电子侦察卫星在执行初始侦察计划过程中各种扰动发生的情况,分析研究各类扰动的特点,建立资源和任务之间的优先映射关系,引入扰动测度来度量动态调度规划结果与原规划结果之间的变化程度;以最大化完成任务优先级之和,以及扰动发生后对原始计划调整最小为目标,建立了具有两级优化目标的动态约束满足模型,提出动态调度的启发式信息计算方法及基于启发式信息的动态调度方法,通过实例仿真,表明该扰动测度和链式效应影响计算方法,以及模型和算法的可行性,对解决实际问题具有一定理论意义和现实意义。     

Dynamic joint pricing and production policy for perishable products

A joint dynamic pricing and production problem for perishable products without shortages is considered. The demand rate is price‐dependent and time‐varying. This paper constructs an optimal control model to maximize the total profit under a general nonlinear production cost function. The feature of the optimal joint dynamic pricing and production policy is analyzed by solving the corresponding optimal control problem on the basis of improved Pontryagin's maximum principle. Then, an effective algorithm is designed to obtain the optimal joint policy. The case of the joint static optimal policy is also investigated and compared with the dynamic one. Finally, numerical examples are presented to illustrate the effectiveness of the proposed methods, and some managerial implications are provided for the management of perishable items.     

Price-Based Residential Demand Response Management in Smart Grids: A Reinforcement Learning-Based Approach

This paper studies price-based residential demand response management (PB-RDRM) in smart grids, in which non-dispatchable and dispatchable loads (including general loads and plug-in electric vehicles (PEVs)) are both involved. The PB-RDRM is composed of a bi-level optimization problem, in which the upper-level dynamic retail pricing problem aims to maximize the profit of a utility company (UC) by selecting optimal retail prices (RPs), while the lower-level demand response (DR) problem expects to minimize the comprehensive cost of loads by coordinating their energy consumption behavior. The challenges here are mainly two-fold: 1) the uncertainty of energy consumption and RPs; 2) the flexible PEVs’ temporally coupled constraints, which make it impossible to directly develop a model-based optimization algorithm to solve the PB-RDRM. To address these challenges, we first model the dynamic retail pricing problem as a Markovian decision process (MDP), and then employ a model-free reinforcement learning (RL) algorithm to learn the optimal dynamic RPs of UC according to the loads’ responses. Our proposed RL-based DR algorithm is benchmarked against two model-based optimization approaches (i.e., distributed dual decomposition-based (DDB) method and distributed primal-dual interior (PDI)-based method), which require exact load and electricity price models. The comparison results show that, compared with the benchmark solutions, our proposed algorithm can not only adaptively decide the RPs through on-line learning processes, but also achieve larger social welfare within an unknown electricity market environment.