UCAV编队对地攻击智能决策系统总体结构

UCAV编队对地攻击是未来战争制信息权和制空权的关键所在,结合无人作战飞机(UCAV)与有人驾驶飞机各自的特点和优势,提出了UCAV编队对地攻击系统的基本组成和对地攻击流程.根据UCAV编队对地攻击的战术决策类型和决策方式,构造了UCAV编队对地攻击智能决策系统的总体结构并且提出了决策系统应该解决的关键技术,详细分析了所建立的UCAV编队对地攻击决策系统的决策方式和决策流程,并且对该系统所涉及的关键技术进行了详细的研究.     

基于贝叶斯决策方法的项目管理能力评估研究

项目管理能力是评价科研单位是否具有牵头组织能力的重要指标之一。针对现有评价方法在稳定性、科学性方面的不足,以传统的科研项目管理能力评估方法为基础,引入贝叶斯理论,建立了贝叶斯决策评估模型,并利用实际案例验证了其在科研项目管理能力评估中的有效性。结果表明,该评估方法能够充分利用历史资料和专家意见,对初步评估结果进行再度评估,具有较强的客观性、准确性。     

基于前景理论的CGF Agent决策建模研究

前景理论(Prospect Theory,PT)描述了个体在不确定条件下的决策规律,较好地刻画了决策者的风险偏好。分析表明,PT在作战决策建模中具有一定的适用性和可行性。提出了一种基于PT的计算机生成兵力(Computer Generated Forces,CGF)Agent决策建模方法,给出了CGF Agent的建模框架,对CGF Agent决策过程中方案选择的具体步骤和算法进行了详细阐述。实验表明,所提出的方法在一定程度上提高了CGF Agent决策行为的真实性,证明了所提出方法的有效性。     

FMEA与粗糙集相结合的故障预测方法

基于FMEA和粗糙集决策系统基本理论,对飞机中分系统故障预测问题进行了研究。首先定义故障预测所需的征兆并对系统进行FMEA,按照FMEA分析结果分次向系统模型注入引起特定故障的事件,采集含有故障趋势的系统模型的信息用于构建适用于故障预测的粗糙集决策表,对该表进行了处理,最终发现用于故障预测的知识(即决策规则和算法);最后运用所发现的知识进行故障预测。     

One Minute in Haditha: Ethics and Non-Conscious Decision-Making

In November 2005, U.S. Marine Sergeant Frank Wuterich fired on and killed five unarmed Iraqi men standing by a car near the site of an improvised explosive device explosion (IED) in Haditha, Iraq. Moments earlier, the IED had destroyed a vehicle in the convoy that Wuterich led, killing one of the men in it. This article argues that analyzing Wuterich's conduct as a rapid non-conscious response captured by the model of Recognition-Primed Decision-Making (RPD) allows us to reconstruct his implicit decision-making in ways that provide insight into psychological processes that may operate during combat. Emerging research in neuroscience suggests that RPD in situations with moral significance involves a complex neural computational process in which emotion plays a powerful role. This article maintains that incorporating these insights can provide greater understanding of the moral psychology of behavior in combat, and may help guide military-ethics training for asymmetric conflicts that pose difficult significant challenges in distinguishing combatants from civilians.     

Dynamic server assignment policies for assembly‐type queues with flexible servers

We seek dynamic server assignment policies in finite‐capacity queueing systems with flexible and collaborative servers, which involve an assembly and/or a disassembly operation. The objective is to maximize the steady‐state throughput. We completely characterize the optimal policy for a Markovian system with two servers, two feeder stations, and instantaneous assembly and disassembly operations. This optimal policy allocates one server per station unless one of the stations is blocked, in which case both servers work at the unblocked station. For Markovian systems with three stations and instantaneous assembly and/or disassembly operations, we consider similar policies that move a server away from his/her “primary” station only when that station is blocked or starving. We determine the optimal assignment of each server whose primary station is blocked or starving in systems with three stations and zero buffers, by formulating the problem as a Markov decision process. Using this optimal assignment, we develop heuristic policies for systems with three or more stations and positive buffers, and show by means of a numerical study that these policies provide near‐optimal throughput. Furthermore, our numerical study shows that these policies developed for assembly‐type systems also work well in tandem systems. © 2008 Wiley Periodicals, Inc. Naval Research Logistics, 2008     

Performance analysis of queue length monitoring of M/G/1 systems

This study investigates the statistical process control application for monitoring queue length data in M/G/1 systems. Specifically, we studied the average run length (ARL) characteristics of two different control charts for detecting changes in system utilization. First, the nL chart monitors the sums of successive queue length samples by subgrouping individual observations with sample size n. Next is the individual chart with a warning zone whose control scheme is specified by two pairs of parameters, (upper control limit, d_u) and (lower control limit, d_l), as proposed by Bhat and Rao (Oper Res 20 (1972) 955–966). We will present approaches to calculate ARL for the two types of control charts using the Markov chain formulation and also investigate the effects of parameters of the control charts to provide useful design guidelines for better performance. Extensive numerical results are included for illustration. © 2011 Wiley Periodicals, Inc. Naval Research Logistics, 2011     

基于参数稳态判定的两阶段序贯仿真分析策略

两阶段序贯分析方法是仿真(模拟)实验研究中的一种常用方法,它虽然可以通过控制实验给出符合精度要求的结果数据,但以其现有形式有时却无法避免冗余计算的存在和实验过早终止的可能.在回顾了现有终态仿真解算终止策略特点的基础上,分析了所需样本总数的估计值与当前可用样本间的变化关系,提出了一种基于参数稳态判定的改进两阶段序贯方法....     

演化硬件中变异概率对进化算法首次命中时间的影响

在(1+1)EAs中,采用马尔可夫链推移时间分析法,推导出了平均首次命中时间的表达式。从理论上分析了变异概率对平均首次命中时间的影响。结果表明适当的变异概率会缩短平均首次命中时间,加快进化算法的寻优时间。     

基于三角模糊数的登陆点选择模型

针对登陆点选择的特点,对登陆点选择步骤进行分析,确定影响的四个基本因素,利用三角模糊数型多属性决策方法建立登陆点选择模型。与层次分析法相比,此算法时更能突出登陆点选择时,指挥者对登陆点的偏好所影响登陆点选择的基本因素值只能大概给出,基本因素之间的权重信息也不能完全确知的特点,通过计算机编程,此模型能够快速地实现对登陆点的排序选优,为指挥员提供辅助决策。