Deciding whether and how to improve statewide travel demand models based on transportation planning application needs

Many states in the USA have developed statewide travel demand models for transportation planning at the state level and along intercity corridors. Travel demand models at mega-region and provincial levels are also widely used in Europe and Asia. With modern transportation planning applications requiring enhanced model capabilities, many states are considering improving their four-step statewide demand models. This paper synthesizes representative statewide models developed with traditional four-step, advanced four-step, and integrated micro-simulation methods. The focus of this synthesis study is as much on model applications and data requirements as on modeling methods. An incremental model improvement approach toward advanced statewide models is recommended. Review findings also suggest model improvement activities should be justified by planning application needs. For statewide model improvement plans to be successful and financially sustainable, the return on model improvement investment needs to be demonstrated by timely applications that rely on improved model capabilities.     

A variational formulation for higher order macroscopic traffic flow models of the GSOM family

The GSOM (Generic second order modelling) family of traffic flow models combines the LWR model with dynamics of driver-specific attributes and can be expressed as a system of conservation laws. The object of the paper is to show that a proper Lagrangian formulation of the GSOM model can be recast as a Hamilton–Jacobi equation, the solution of which can be expressed as the value function of an optimal control problem. This value function is interpreted as the position of vehicles, and the optimal trajectories of the optimal control formulation can be identified with the characteristics. Further the paper analyzes the initial and boundary conditions, proposes a generalization of the inf-morphism and the Lax–Hopf formulas to the GSOM model, and considers numerical aspects.     

论航海高等职业教育的改革问题

航海高等职业教育应转变观念,按照现代职业教育理论和成果改革航海高等职业教育教学环节,在质量体系和教学计划制定程序中引进现代职业教育的理念和成果;构建航海高等职业教育课程体系;改革教学形式和内容;改进学员的考核和评价形式。同时,改革需要注意和STCW公约马尼拉修正案履约相结合;注意"行动导向"教学的形散、神不散问题;注意与体制、机制改革相结合。     

基于多层隐类贝叶斯网络的舰船生存能力评估模型研究

生存能力是舰船设计时需要考虑的一个重要性能指标,然而其涉及的因素较多,因素之间的关系错综复杂,不确定信息充斥其间.在系统分析舰船生存能力评估要素的基础上,针对评估过程中的不确定性信息难以量化处理的特点,引入基于贝叶斯网络的多层隐类模型算法对舰船生存能力进行评估,给出了模型评分原理.最后以实例说明了建模方法与评估过程,并结合专家意见分析了模型的优劣,说明该多层隐类模型的算法符合实际情况.     

钢混组合桥面板的发展概况及设计维护要点

结构优化的少主梁新型组合桥梁体系,对桥面板的跨径,耐久性等提出了新的要求。组合桥面板从历史上单纯以钢板作为模板,发展为合理的利用了钢混两种材料各自的优点,提高了桥面板跨径,减少了重量,加快了施工速度,提高了施工安全性。通过一系列动载和静载试验证明,组合桥面板具有和预应力混凝土桥面板同等级的承载能力和耐久性。组合桥面板制造施工费用与现浇预应力桥面板相近,但工期可缩短30%以上,并且其维护成本低、替换拓宽方便,全寿命费用合理。因此,组合桥面板是值得我国借鉴并在今后进行发展研究的一种结构形式。     

集贤路跨派河桥钢桁拱桥施工关键技术

集贤路跨派河桥主桥为53.9 m+132 m+53.9 m三跨连续桁箱组合形式,主桥钢梁结构复杂、栓接结构精度控制要求高、整体线型控制难度大等技术难题,厂内制作针对不同部位的构件分别制定了不同的制作工艺方法,工地安装采用支架法从两端向跨中架设,支座段牛腿、桥面系及钢桁拱圈各部分,分别采用不同的架设方法.最后在主跨跨中合龙。     

An assessment of models to estimate pedestrian demand based on the level of activity

This paper presents the development and assessment of models to estimate pedestrian demand based on the level of pedestrian activity (high and low). As activity varies by the time of the day, temporal variations were evaluated by considering different time periods. Data collected at 128 low and 48 high pedestrian activity signalized intersections (a total of 176 signalized intersections) in the City of Charlotte, North Carolina were used to develop and assess the models using stepwise regression analysis through backward elimination of independent variables (includes demographic, land use, and network characteristics). The use of different buffer widths (proximal area) to extract these characteristics was also evaluated. Results, in general, show that pedestrian demand varied by the level of activity, explanatory variables extracted by buffer width, and time of the day. The estimates from the models could be used in transportation planning (identify required pedestrian facilities, resource allocation), safety, and operational analyses. Copyright © 2010 John Wiley & Sons, Ltd.     

基于支持向量机的舰船维修费用组合预测研究

在费用预测中,利用单一模型往往存在着信息不足的缺陷。为了提高舰船维修费用的预测精度和稳定性,采用支持向量机（SVM）回归算法,把几种单一预测模型结果作为输入,实际值作为输出,然后用足够多的预测案例训练学习机器,在各组合的模型预测结果与实际之间得到一种非线性映射关系,从而建立了非线性组合预测模型。最后,以某型舰船维修费用为例,对指数平滑法、灰色预测和参数法3种方法的预测结果进行仿真,结果表明此法较传统的单一模型预测法具有更高的预测精度。     

Maximizing bus discharge flows from multi-berth stops by regulating exit maneuvers

Upon having loaded and unloaded their passengers, buses are often free to exit a multi-berth bus stop without delay. A bus need not wait to perform this exit maneuver, even if it requires circumventing one or more other buses that are still dwelling in the stop’s downstream berths. Yet, many jurisdictions impose restrictions on bus entry maneuvers into a stop to limit disruptions to cars and other buses. Buses are typically prohibited from entering a stop whenever this would require maneuvering around other buses still dwelling in upstream berths. An entering bus is instead required to wait in queue until the upstream berths are vacated.Analytical models are formulated to predict how bus discharge flows from busy, multi-berth stops are affected by allowing buses to freely exit, but not freely enter berths. These models apply when: a bus queue is always present at the stop’s entrance; buses depart the entry queue and enter the stop as per the restriction described above; and the stop is isolated from the effects of nearby traffic signals and other bus stops. We find that for these restricted-entry stops, bus-carrying capacities can often be improved by regulating the exit maneuvers as well. This turns out to be particularly true when the stop’s number of berths is large. Simulations show that these findings still hold when a stop is only moderately busy with entry queues that persist for much, but not all of the time. The simulations also indicate that removing any restrictions on bus exit maneuvers is almost always productive when stops are not busy, such that short entry queues form only on occasion, and only for short periods. We argue why certain simple policies for regulating exit maneuvers would likely enhance bus-stop discharge flows.     

考虑合乘的共享自动驾驶汽车选择行为实证分析

共享自动驾驶汽车(Shared Autonomous Vehicles,SAV)是自动驾驶汽车和共享经济相结合的产物,为人们提供了一种新型的出行方式. 为探究出行者在考虑合乘的SAV与私家车或公共交通之间的选择偏好,实施了SAV选择意愿调查,并分析了考虑合乘的SAV的潜在用户特征. 基于问卷调查所得有效数据,采用K-Means 聚类法划分了历史出行模式,利用因子分析对性格态度特征进行了分类. 分别对有无私家车人群建立解释变量的参数服从不同分布的混合Logit 模型,并对参数标定结果进行对比分析. 研究结果表明,出行方式特性非常显著地影响出行者方式选择行为,性格态度特征是影响出行者选择考虑合乘的SAV出行方式的显著因素,且其显著性明显高于性别、年龄等社会经济属性.