基于开放式创新的供应链企业合作研究

[交通会议论文数据库] 吕璞

摘要:　　传统的创新模式已经不能适应当今激烈的市场竞争，越来越多的企业开始选择与其他利益相关者合作的开放式创新模式。

视觉减速标线控速效果评价

[交通会议论文数据库] 郭延永 刘攀 梁启宇

摘要:　　为分析视觉减速标线的控速效果和影响范围,选取6处视觉减速标线路段进行了实地车速采集.利用断面车速数据构建速度-距离曲线,研究了车速的变化规律和设施控速的影响范围；采用t检验和比例检验分别检验了视觉减速标线上下游控制点及标线末端车速平均值之间的差异显著性和超速比例的差异显著性,验证了视觉减速标线的控速效果,并通过与对照组数据的对比排除了其他因素的干扰.结果表明,视觉减速标线的控速范围为上下游约250 m；车辆平均速度下降程度为5.2～ 7.0 km/h,超速车辆比例下降18 8％～25.0％.

基于城市更新的慢行系统城市适宜性要素分析

[交通会议论文数据库] 刘一杰

摘要:　　通过对当前慢行系统现状的分析梳理，发现国内城市存在盲目建设或忽视慢行系统发展的状况，原因之一缺失慢行系统的城市适宜性研究。基于城市更新背景和理论，分析、归纳并提出基础要素、协同要素两部分组成的慢行系统城市适宜性要素，并论证其内容与慢行系统的相互影响关系，主要包括城市地理、文化、用地、交通模式、规划编制、公共空间以及政府行为等方面。最后，结合哥本哈根和香港典型案例的城市慢行适宜性展开评估验证。

行人微观仿真中碰撞检测和响应的探讨

[交通会议论文数据库] 郑勋 李海鹰 孟令云

摘要:　　本文综述了行人微观仿真中碰撞检测和响应的五大类模型：社会力模型、元胞自动机模型、基于规则的模型、多智能体模型和混合模型.HiDAC模型是一种代表性的混合模型，该模型结合了行人的心理和几何规则，可以较好地解决行人碰撞和检测问题.本文介绍了HiDAC模型的避免力、排斥力、停止规则、排队行为、推挤行为等，通过一个实例对HiDAC模型进行仿真，并针对仿真效果提出了可能存在的问题，有待进一步改进的建议.

公共汽车运行服务特征参数波动分析

[交通会议论文数据库] 陈国俊 杨晓光 张栋 刘好德

摘要:　　公共交通服务质量直接影响着乘客公共交通出行的选择意愿，提升公共交通服务水平是公共交通优先发展战略实施的关键要素。由于公共汽车运行过程受到内外部环境因素的约束和影响，车辆运行过程中实际的服务质量产出低于理论水平。本文通过对运行服务特征参数的统计分析发现：路段行程时间所占比例远远高于站点停靠时间所占比例，路段车头时距偏差是导致运行服务偏差产生的主要原因；因此，导致我国公共汽车服务水平低下的关键因素是路段运行服务的波动。

基于贝叶斯空间技术的城市道路网路事故预测联合模型

[交通会议论文数据库] 曾强 黄合来

摘要:　　为了对城市道路网络中路段和交叉口进行事故预测，本文建立了基于贝叶斯空间技术的联合模型。该模型考虑了相邻路段和交叉口间的空间关联。根据美国佛罗里达州希尔斯伯勒县的一个道路网络三年的事故、道路和交通特征数据，将该联合模型和三个传统的事故预测模型——泊松、负二项、条件自回归模型，进行了对比。结果显示，联合模型和条件自回归模型在模型拟合和预测能力上优于泊松和负二项模型，表明了考虑空间关联的合理性。在本算例中，尽管条件自回归和联合模型的拟合优度及预测能力相当，但是路段和相邻交叉口间的空间相关性比路段和路段、交叉口和交叉口间的空间相关性更显著。这一结果有力地支撑了所构建的联合模型在路网层面安全建模的有效性。为了进一步展示所建模型的优越性，最后 还将其运用到了对城市交通走廊、出行路径和交通分析小区的安全评价。

多约束下的车辆导航算法研究

[交通会议论文数据库] 于尧 杨兆升 莫祥伦

摘要:　　由于现有车辆导航算法未考虑出行者在出行过程中连续访问多点的实际需求,导致出现路径重复诱导,无法得到最优路径.本文提出了一种新的路径导航算法ASSA.ASSA算法以城市热点信息为基础,通过热点近邻区域的时空关联推理,设计了多规则下的热点访问机制,并结合路网动态信息,给出了动态路径调整策略,进而保证用户出行费用最低.对所提出的算法进行了实验测试,结果表明,在不同的约束条件下,ASSA算法不但在计算时间和计算距离上明显优于NS算法,且由于引入了导航终点的约束,避免了非最优路径的产生,从而能够高效、快速的响应用户的导航查询.

基于决策树的驾驶疲劳等级分析与判定

[交通会议论文数据库] 胥川 王雪松 陈小鸿 张慧

摘要:　　疲劳等级判定是车载疲劳预警系统的关键技术，现有研究通常采用单一特征进行.本研究通过驾驶模拟实验采集车辆位置、方向盘操控、眼动等多源数据并计算疲劳特征指标，同时采集驾驶员主观疲劳程度并通过视频回放进行校核，在此基础上建立疲劳等级与特征指标的决策树模型.研究结果表明：对于区别疲劳等级最显著的变量有PERCLOS(Percentage of eye closure)、车道偏移标准差、越线时空面积、方向盘反转率，且上述变量与疲劳等级呈正相关；PERCLOS为最优的疲劳等级划分变量，当PERCLOS小于2.8％时，驾驶员处于严重疲劳状态的路段比例为0％；当PERCLOS大于21.9％时，驾驶员处于未疲劳状态的比例为0％；疲劳分级模型的预测正确率为64.3％，其中对于轻度疲劳的预测正确率最高，达87.5％.

User Equilibrium Model with Stochastic Travel Times: Including Parking Space Searching and Access Times

[交通会议论文数据库] Zhu Chengjuan Jia in Han Linghui

摘要:　　To generalize the existing equilibrium assignment and parking choice approaches to accommodate travelers' reactions to variability in travel times including parking space searching and access times,a variational inequality model is proposed.The travel disutility simultaneously considers the path travel time budget, parking fees and public transport fares.Several illustrative experiments are used to analyze the flow distribution and disutility under different network conditions and confidence levels.The travel flow pattern and disutility can be affected by the travel time of the metro effectively.The market share of parking clusters can be regulated by different traffic management and control measures.

Methodology to Identify the Recommended Values of Advisory Exit Speed Limit on Freeway Exit Ramp in China

[交通会议论文数据库] Hu Sitao Xiang Qiaojun Yang Weiwei

摘要:　　Many crashes occurred around freeway exit ramp areas due to speeding and speed variance,especially in China.Traditionally, a posted ramp speed limit will be installed around physical gore area to manage speed around the area.Unfortunately, the field study showed that speeding and speed variance haven't been improved too much after introduction of posted ramp speed limit.An additional speed limit on deceleration lane may be needed to accommodate the speed changes ahead of physical gore, which could be functioned by Advisory Exit Speed Limit (AESL).However, no systematic research on the application of AESL has been reported yet.Thus, the primary objective of the paper is to develop a general methodology to identify and to validate the recommended values of AESL To achieve this goal, seven Sites with same exit ramp configuration in China were selected and two Scenarios (with AESI/ without AESL) were investigated to quantify the effectiveness.Two hours field data were collected for six Sites without AESL and the other one Site wtth/without AESL,the speed and deceleration rate of approximately 808 vehicles were obtained.The 85th percentile speed of 6 Sections at six Sites (Site 1 to Site 6) without AESL were used to identify the recommended values of AESL while the other one Site (Site 7) with two Scenarios was used for validation.The 85th percentile speeds of Section 6 (the physical gore) at three Sites exceed the posted ramp speed limit (40 km/h) more than 40% while those of other Sites exceed the posted ramp speed limit (60 km/h) only about 10%.So AESL is very necessary for the exit ramps which the posted ramp speed limit is 40 km/h.Based on the recommended values of AESL, an experiment application of AESL was conducted at Site 7.t-test was introduced to verify the differences in speed before and after experimental application of AESL.Results showed that with AESL, the mean speed and the 85th percentile speed reduced significantly at Site 7, but no obvious improvement on speed variance achieved even there were some tendencies on speed decrease on deceleration lane.The paper can provide researchers, managers and engineers with reasonable foundations, when determining the recommended values of AESL The proposed methodology will be especially helpful in China and other developing countries for similar exit ramps.