基于决策融合的城市道路交通事件自动检测算法研究

为了提高城市道路交通事件自动检测算法的性能,引入算法性能可靠度概念对基于浮动车数据和感应线圈数据的事件自动检测算法检测结果进行决策融合。决策融合算法包括3个模块：①感应线圈数据算法模块：选择流量、占有率、路段长度、前一个检测周期的检测参数作为输入参数,训练BP网络进行事件判别;②浮动车数据算法模块：使用误差分析理论确定满足数据精度要求的最小浮动车样本量,选择路段行程时间、行程速度作为BP网络输入参数,进行事件判别;③决策融合模块。引入算法性能可靠度概念,计算模块一和模块二判别结果的权重值,使用加权平均法进行决策融合。通过Vissim仿真获得数据,在Matlab中编程实现算法的计算,仿真结果表明决策融合算法的性能优于单数据源事件自动检测算法。     

内河水上危险货物集装箱运输监测系统研究

水上危险货物集装箱监控系统应具有水上危险货物集装箱状态监控信息采集、预报警、信息远程传输、远程监控与应急决策支持等功能,其框架应为4层架构,即信息采集层、现场监控层、信息传输层、远程监控层。文中分析了构建内河水上危险货物集装箱运输监控系统的关键技术,设计了信息采集与传输层的硬件电路,提出了开发远程应急管理系统的支撑技术与决策模型方法。     

基于灰色理论的沪苏浙高速公路江苏段软基处理方案决策

针对公路软基处理方案决策中,影响因素多,存在大量不确定性、随机性、模糊性的特点,提出了基于灰色理论的沪苏浙高速公路江苏段软基处理方案决策模型。从技术可行性、经济合理性、环境友好性的设计理念出发,确定了模型的经济技术指标,并利用灰色关联分析方法,通过计算比较各方案技术经济指标数据列与最优参考数据列的关联度的大小,采用"塑板+砂砾垫层+预压"方案处理公路软基,效果良好。     

基于熵权-ANP模型的自行车租赁点选址问题

为了寻求自行车租赁点的最优选址,从其多目标决策的特点出发,以网络层次分析法(ANP)为基本框架,把换乘时间、吸引距离及土地成本等作为决策指标,同时,创新性地引入熵权,计算和分析各个决策指标间相互影响的程度,最终建立熵权-ANP 模型。通过一个实例,验证了该模型能客观地对决策方案做出定量分析,得到决策结论。     

基于距离信息的鱼雷攻击意图判别方法研究

编队作战是未来海上作战的基本样式,进行舰艇编队水声对抗研究具有十分重要的意义.如果舰艇编队能够通过数据融合等方式获取鱼雷距离和方位的报警信息,并根据这些信息判断出鱼雷对舰艇编队的攻击意图,即鱼雷攻击的具体目标舰艇,则有助于指挥员采取有效、合理的对抗方法,从而提高舰艇编队的生存能力.论文在建立声自导鱼雷攻击模型和舰艇编队运动模型的基础上,对鱼雷从不同区域来袭时的态势划分,提出了攻击意图的判断准则,并且通过仿真验证,分析了影响判断准确度的因素,为编队协同水声对抗指挥决策提供了必要参考.     

基于遗传算法的舰船抗沉决策模型研究

为了评估抗沉方案的优劣,本文提出紧急情况下辅助快速决策的抗沉决策模型,并采用遗传算法来求解,核心思想是采用可用的抗沉舱进行压载或排空操作来调整船舶浮态。改进的遗传算法采用轮盘赌选择,单点交叉,变异进行迭代计算直到满足停止准则,选择倾斜角和横倾角作为停止准则。通过实例计算,证实了提出的优化模型和遗传算法的有效性。     

���ڷֲ�Agent����·Ӧ����Ԯ���߷����о���

铁路应急救援决策是铁路运输安全管理领域研究的热点之一.本文以铁路应急救援为背景,分析多智能体技术在铁路应急决策领域应用的可行性,提出一种基于分层Agent的铁路应急救援指挥控制结构,并分析了各层Agent的基本功能.根据各层Agent的结构特点设计决策推理的模型,采用基于模糊评价的方法实现管理Agent的决策机制,并阐述了决策算法的计算过程.最后通过一个具体的列车救援决策案例进行模型的验证.研究结果表明,将该方法应用在铁路应急救援决策领域具有可行性.Agent技术和模糊评价方法相结合的决策方法为研究铁路应急救援的智能化提供了新的思路.     

ͻ���¼�Ӧ��ѪҺ������������׶ξ��߷���

基于血液的生理特性及应急保障特性,以应急血液最晚运达时间最小、接收时的最低新鲜度最大、运输总费用最小为目标,建立考虑多品种、多式联运的应急血液调剂出救点选择,即运输路线安排问题两阶段决策优化模型,并设计了一种包含局部邻域优化的遗传,即禁忌混合算法进行求解.以“5.12”汶川大地震应急血液保障为背景构建算例,分析验证了两阶段决策方法在最晚抵达时间、新鲜度均值、运输总费用、血液积压量等指标上的优越性.     

Substantial Changes and Their Impact on Mobility: A Typology and an Overview of the Literature

During the past 40 years, mobility patterns have enduringly changed several times as a result of the occurrence of a number “substantial changes”. Examples of such substantial changes are the rapid emergence of affordable air travel, the oil crises, and profound ICT developments. To most researchers and policy-makers in transportation, it seems more than likely that the next 40 years will also witness a number of substantial changes, some of which might even have larger impacts on mobility than the ones described above. However, literature on substantial changes and their impact on mobility are difficult to access as it is spread across different research fields and suffers from ambiguous use of terminology. As a result, overview of the literature on substantial changes and their findings is missing and discussions on the impacts on future mobility of potential substantial changes are hampered. To overcome these problems, this paper (1) proposes a typology of substantial changes and (2) ties together and reviews the scholarly literature that has focused on identifying the impacts of past substantial changes on mobility patterns. In this paper, we show how the proposed typology on substantial changes can be applied to contemplate on substantial changes and on their impacts.     

A decision support system for total airport operations management and planning

The airport planning and decision making process exhibits various trade‐offs and complications due to the large number of stakeholders having different, and sometimes conflicting, objectives regarding the assessment of airport performance. As a result, the airport performance assessment necessitates the use of advanced modelling capabilities and decision support systems or tools in order to capture the multifaceted aspects, interests and measures of airport performance like capacity, delays, safety, security, noise and cost‐effectiveness. Presently, airport decision makers lack decision support tools able to provide an integrated view of total airport (both airside and landside) operations and analyse at a reasonable effort and decision‐oriented manner the various trade‐offs involved among different airport performance measures. The objective of this paper is twofold: (i) to describe the decision‐oriented modelling framework and development process of a decision support system for total airport operations management and planning, and (ii) to demonstrate the decision support capabilities and basic modelling functionalities of the proposed system. Copyright © 2010 John Wiley & Sons, Ltd.