Enhancing Political Cohesion in NATO during the 1950s or: How it Learned to Stop Worrying and Love the (Tactical) Bomb

This article argues that the perceived need by NATO to nurture political cohesion within the Alliance during the 1950s resulted in the adoption of strategic concepts that were out-of-step with the military environment in which it was operating. It maintains that the Alliance acquiesced to American leadership on nuclear issues which led to the development of tactical nuclear capabilities at the expense of conventional war-fighting capabilities for the defence of the European Central Front. This resulted in a strategic concept that enhanced political cohesion but was militarily unviable.     

可靠度和费用约束下的加速老炼试验时间优化

为了提高产品实际使用可靠度,减少老炼试验和保修费用,基于效用函数研究了具有早期失效的产品在可靠度和费用约束下的加速老炼试验时间优化方法。考虑产品寿命分布参数的不确定性,利用自助法估计参数的先验分布,然后通过仿真方法获得实际使用可靠度和费用的联合效用函数。以联合效用函数值最大为优化目标建立加速老炼试验时间优化模型,得到最优的加速老炼试验时间。结合对数正态分布产品在温度应力下的加速老炼试验数值示例,阐述上述方法的应用,并对参数进行敏感性分析,结果表明提出的加速老炼试验方法不但能缩短产品的试验时间,还可以最大程度同时满足费用和可靠性的要求。     

武器装备体系演化研究

体系动态演化是武器装备体系建设发展、体系作战能力和体系对抗效能评估优化的基本认识前提。鉴于当前武器装备体系概念理论基于多问题思维视角现状,研究层次、问题形式和技术方法手段复杂多样,以体系演化为切入点,围绕装备体系动态演化的涵义,从体系结构优化设计、体系协同作战过程优化和体系作战效能评估优化3个层次综述了装备体系演化研究进展。在此基础上,讨论了基于大数据、超网络、认知粒计算的装备体系多学科研究需求,对可能性的技术思路给出可供借鉴建议。     

Flexible capacity strategy in an asymmetric oligopoly market with competition and demand uncertainty

This article studies flexible capacity strategy (FCS) under oligopoly competition with uncertain demand. Each firm utilizes either the FCS or inflexible capacity strategy (IFCS). Flexible firms can postpone their productions until observing the actual demand, whereas inflexible firms cannot. We formulate a new asymmetrical oligopoly model for the problem, and obtain capacity and production decisions of the firms at Nash equilibrium. It is interesting to verify that cross‐group competition determines the capacity allocation between the two groups of firms, while intergroup competition determines the market share within each group. Moreover, we show that the two strategies coexist among firms only when cost differentiation is medium. Counterintuitively, flexible firms benefit from increasing production cost when the inflexible competition intensity is sufficiently high. This is because of retreat of inflexible firms, flexibility effect, and the corresponding high price. We identify conditions under which FCS is superior than IFCS. We also demonstrate that flexible firms benefit from increasing demand uncertainty. However, when demand variance is not very large, flexible firms may be disadvantaged. We further investigate the effects of cross‐group and intergroup competition on individual performance of the firms. We show that as flexible competition intensity increases, inflexible firms are mainly affected by the cross‐group competition first and then by the intergroup competition, whereas flexible firms are mainly affected by the intergroup competition. Finally, we examine endogenous flexibility and identify its three drivers: cost parameters, cross‐group competition, and intergroup competition. © 2017 Wiley Periodicals, Inc. Naval Research Logistics 64: 117–138, 2017     

Global strategic studies: a manifesto

This contribution reflects on the state of strategic studies today and the criticism it has received in recent years, as being outdated and irrelevant. The authors formulate some premises for reinvigorating this field of inquiry by widening its scope and research agenda to do more justice to the wide variety of actors, perspectives and practices observable in the enterprise of strategy in our contemporary globalised world.     

分布式计算环境下的栅格数据存储策略

针对传统的栅格数据存储策略不能满足分布式计算环境下粗粒度数据访问需求,应对海量栅格数据计算时效率低下的问题,结合分布式文件系统的存储特点,同时考虑地图代数算子在Map/Reduce阶段以栅格瓦片为单位的计算特点,提出一种基于Hadoop分布式文件系统的栅格瓦片存储策略。围绕栅格数据瓦片分割、压缩瓦片数据组织与存储、分布式文件输入输出接口改进等方面对该存储策略加以实现,并使用基于该存储策略的地图代数局部算子的分布式计算流程加以验证。理论分析与实验结果表明,该策略能够显著提高分布式计算环境下空间分析算子的运算速度。     

基于压缩传感的PSO-OMP重构算法的研究

由于粒子群算法具有解决寻优问题的能力,将其应用于信号处理领域,提出了一种新的基于PSO-OMP的信号重构算法。为了降低计算复杂度,把粒子群算法运用在正交匹配追踪算法的匹配过程,以此来确定最优原子。实验结果表明,所提出的新的基于PSO-OMP的信号重构算法具有计算复杂度低和重构成功概率高等特点。     

广义逆指数分布元件的可靠性分析

在II型混合截尾样本下,得到了广义逆指数分布未知参数的最大似然估计。利用最大似然估计的渐近正态性构造了参数的渐近置信区间,运用Lindley's逼近方法和TierneyKadane's逼近方法计算出了参数的Bayes估计。最后,运用Monte-Carlo方法对上述估计方法结果作了模拟比较。     

基于有限条件的多波束雷达干扰系统目标分配算法

针对多波束干扰系统同时干扰多个目标的资源分配问题,通过分析目标分配算法的一般流程及涉及到的关键问题和技术难题,提出了基于实战化和有限条件的针对多波束干扰系统的非线性0-1整数规划数学模型。针对该模型采取开源软件SCIP进行求解,最后给出数值仿真来说明模型和算法的有效性。     

基于Pareto排序法的战时装备维修任务多目标调度方法

针对战时武器装备维修保障问题,建立了包括最大完工时间、延迟时间和维修单元负荷在内的多目标优化调度模型。为提高解的多样性和收敛性,构建了一种基于Pareto排序法和小生境技术相结合的遗传算法用于模型求解,引入Pareto排序和拥挤距离进行适应度计算,通过混沌系统随机生成权重系数,并使用小生境技术改进选择方式。通过实例验证表明,该方法能够有效地解决装备维修多目标调度问题。