Periodic review inventory management with contingent use of two freight modes with fixed costs

We study a stochastic inventory model of a firm that periodically orders a product from a make‐to‐order manufacturer. Orders can be shipped by a combination of two freight modes that differ in lead‐times and costs, although orders are not allowed to cross. Placing an order as well as each use of each freight mode has a fixed and a quantity proportional cost. The decision of how to allocate units between the two freight modes utilizes information about demand during the completion of manufacturing. We derive the optimal freight mode allocation policy, and show that the optimal policy for placing orders is not an (s,S) policy in general. We provide tight bounds for the optimal policy that can be calculated by solving single period problems. Our analysis enables insights into the structure of the optimal policy specifying the conditions under which it simplifies to an (s,S) policy. We characterize the best (s,S) policy for our model, and through extensive numerical investigation show that its performance is comparable with the optimal policy in most cases. Our numerical study also sheds light on the benefits of the dual freight model over the single freight models. © 2011 Wiley Periodicals, Inc. Naval Research Logistics, 2011     

基于人工蜂群算法优化的改进高斯过程模型

高斯过程(GP)的非线性特征导致其对大样本的训练时间复杂度过高,而且其超参数的选取是否适当直接影响高斯过程回归模型的预测精度。提出采用人工蜂群(ABC)算法优化改进GP以减小时间复杂度和提高预测精度。改进GP通过选取训练样本的子样本进行模型学习,以降低训练过程的时间复杂度。ABC通过优化改进GP的超参数,提升预测精度。选取训练样本的子样本构建改进GP回归(GPR)模型,采用ABC算法搜寻改进GPR的最优超参数,并用得到的超参数构建最优的改进GPR模型,输入测试样本进行预测并输出预测精度。将该模型应用于解决海上远程精确打击(LPSS)体系作战效能评估问题中,通过MATLAB仿真实验,与常见的多种优化方法相比较,验证了该模型的有效性。     

The impact of operational decisions on the design of salesforce incentives

When facing high levels of overstock inventories, firms often push their salesforce to work harder than usual to attract more demand, and one way to achieve that is to offer attractive incentives. However, most research on the optimal design of salesforce incentives ignores this dependency and assumes that operational decisions of production/inventory management are separable from design of salesforce incentives. We investigate this dependency in the problem of joint salesforce incentive design and inventory/production control. We develop a dynamic Principal‐Agent model with both Moral Hazard and Adverse Selection in which the principal is strategic and risk‐neutral but the agent is myopic and risk‐averse. We find the optimal joint incentive design and inventory control strategy, and demonstrate the impact of operational decisions on the design of a compensation package. The optimal strategy is characterized by a menu of inventory‐dependent salesforce compensation contracts. We show that the optimal compensation package depends highly on the operational decisions; when inventory levels are high, (a) the firm offers a more attractive contract and (b) the contract is effective in inducing the salesforce to work harder than usual. In contrast, when inventory levels are low, the firm can offer a less attractive compensation package, but still expect the salesforce to work hard enough. In addition, we show that although the inventory/production management and the design of salesforce compensation package are highly correlated, information acquisition through contract design allows the firm to implement traditional inventory control policies: a market‐based state‐dependent policy (with a constant base‐stock level when the inventory is low) that makes use of the extracted market condition from the agent is optimal. This work appears to be the first article on operations that addresses the important interplay between inventory/production control and salesforce compensation decisions in a dynamic setting. Our findings shed light on the effective integration of these two significant aspects for the successful operation of a firm. © 2014 Wiley Periodicals, Inc. Naval Research Logistics 61: 320–340, 2014     

基于知识的模型库管理系统

模型库管理系统是决策支持系统区别于其它计算杌信息系统的一个重要特征,着重介绍了如何将知识的表达技术(网络图)和推理机制(深度优先搜索)应用于模型库管理系统,从而实现模型的自动集成和选择,并给出了具体实现过程。     

五参数寿命分布S(α,β,λ,μ,δ)

在四参数寿命分布的基础上增加了一个非中心参数,进而提出五参数寿命分布S(α,β,λ,μ,δ),并给出了定义、基本性质及其证明。     

带噪声的 AR 模型的参数估计及定阶

本文讨论了带噪声情形下 AR 模型的参数估计及定阶问题,还给出了噪声模型的参数估计,并证明了它们的强相容性。     

自由后坐式自动武器运动系统优化设计研究

本文介绍自由抢机式自动武器运动系统有关结构参量的一种优化设计方法。优化的目标函数为自动机后坐到位速度和弹丸的初速。优化的结构参数有自动机的质量,弹簧的抗力等。     

Cρ 准则下选择最优子集的并行算法

C_p准则是目前颇受重视的一种变量选择准则。本文针对大型线性回归模型,推导了从所有可能子集中用C_p准则选择最优子集的(乘除法) 运算次数,提出了C_p准则下变量选择的并行算法。给出了在YH-1和YH-2向量巨型计算机上运行该算法的模拟结果且获得了15倍左右的向量加速比s/v,体现了该算法的优越性。     

弹道数据处理融合算法的并行计算

针对参考文献[1]中提出的融合多信源信息的融合算法,讨论了其中大计算量的测元遴选问题,并给出了它的并行算法。最后详细地分析了此并行算法的高效性和可扩展性,给出了加速比的仿真结果     

纯引力轨道验证质量辐射计效应与气体阻尼耦合模型的数值分析

辐射计效应和气体阻尼是纯引力轨道验证质量的重要干扰力,是影响纯引力轨道构造水平的重要因素.在纯引力轨道飞行器中,这两种力分别描述了由腔体中温度梯度和验证质量相对运动引起的气体分子作用,两者从不同角度描述了气体分子作用力,均是气体分子作用力的一部分,而两者的耦合模型则可以反映验证质量受到的气体分子作用力总和.针对耦合模型形式复杂的特点,本文以内编队系统为例,利用数值方法分析了耦合模型中的影响因素,这些因素包括内卫星相对运动速度、内卫星半径、外卫星腔体半径、腔体平均温度、腔体温差和腔体平均压力等.对大量计算结果进行了数据拟合,给出了内卫星气体分子作用力与各物理参数关系的拟合公式,和原始计算结果相比,拟合误差在20％以内.