基于卷积神经网络的列车检测算法研究

列车检测作为列车自动驾驶的核心技术,可以有效地降低列车追尾等事故造成的人身危险和财产损失。为实现精准的列车检测,选用改进的卷积神经网络(PVANET)对输入图像进行特征提取,在此基础上,采用候选区域网络,从生成的特征图里滑动搜索,判断出图像中可能为列车的区域位置,并进一步采用快速区域卷积神经网络对每个候选区域进行分类,计算出其所属类别的置信度,同时精确定位列车。经验证,该方法适应范围广、鲁棒性高,可以有效地检测不同环境光强及不同朝向的列车,保障列车安全,为列车自动驾驶及辅助驾驶提供安全保障。     

Directional Land Value Gradients

The standard urban model supports the concept of a constant land price gradient throughout the urban area. It is a reasonable conjecture that the land price gradient would vary with direction from the CBD. The variation in the gradient could be caused by a number of factors, but the idea that the land price gradient is flatter along radial transportation routes than in other directions is widely recognized even though there is little rigorous empirical work supporting this belief. This paper will examine the structure of urban land prices with a focus on the land price gradient as a function of the direction around the center of the city using a piecewise linear function. The added flexibility in the gradient estimate gained by this approach reveals a dramatically varying directional land price gradient.

安全活动对物流企业绩效作用的实证分析

本文以物流企业为调查对象进行问卷调查,通过Bootstrapping和结构方程模型的路径分析,分析了物流安全活动对成本控制,运输效率和企业绩效之间的影响关系.同时,以路径分析结果为基础,提出了推广物流安全实践活动和提高物流企业绩效的意义.结果表明,安全运输实践及安全运输预防管理对成本控制和运输效率都有着显著的积极影响,运输手段安全管理对成本控制有着显著的积极影响,而运输手段安全管理对运输效率产生的影响力在5％的显著性水平下不显著,成本控制对运输效率也产生了显著的积极影响关系,成本控制和运输效率对企业绩效也都有着显著的积极影响.     

铁路行包运量预测模型研究

铁路行包运量预测是以运输需求和内部供给为导向,综合考虑各种影响因素,对行包运量现状和发展的正确把握.探讨利用人工神经网络结合主成分分析的方法,建立铁路行包运量预测模型,解释并预测行包专列开行后铁路行包运量的增长趋势.实例分析的仿真结果表明,采用主成分分析法的广义回归神经网络模型结构简洁、预测精度高、收敛速度快,对相关铁路部门和企业的决策具有参考意义.     

交通运输对区域经济影响效应的研究进展及展望

交通运输是国民经济发展的命脉,也是区域经济持续快速发展重要的支撑、服务和保障体系的组成部分。本文从区位条件、区域产业结构、区域空间结构、区域市场、区域发展战略、区域合作等方面分析了目前我国交通运输对区域经济影响效应研究的进展,指出交通运输对区域经济影响效应研究中存在的问题,并在此基础上,探讨了我国该领域未来的研究趋向和发展态势。     

Can highway investment policies influence regional growth?

If a highway investment policy instrument is to alter growth in the manner desired, two criteria must be satisfied: (1) the policy instrument must be implemented, and (2) the implementation must achieve the desired policy goal. A look at three decades of highway policy goals (slowing net population growth, reducing arterial congestion, and fostering city-specific population increases) and policy instruments (zoning and sewer restrictions, gridded networks, and highway facility construction or non-construction) provided insights regarding the extent to which Virginia's growth-oriented highway investment decisions satisfied these two criteria.Crucial to the first criterion is that a policy instrument will not be implemented if it lacks a supporting mechanism. For example, the policy instrument of gridded networks in this study lacked any ordinance, financial incentive, or training to facilitate its implementation. Crucial to the second criterion is that highway investments are not effective for achieving a growth-related policy goal. Development will continue in a region despite efforts to slow it by deleting new roads from master plans or will occur elsewhere despite efforts to target it in a specific area.     

AN APPLICATION OF AN INTEGRATED TRANSPORT NETWORK – MULTIREGIONAL CGE MODEL TO THE CALIBRATION OF SYNERGY EFFECTS OF HIGHWAY INVESTMENTS

A transportation network-multiregional CGE model is applied to estimate the synergy effects of a set of highway projects on value added by region and industrial sector. This synergy effect is defined as a difference between the summation of the net GDP increase from the development of each highway sub-link without spatial linkage and the change in GDP resulting from the concurrent development of all links with spatial linkages. Among nine east–west highways in Korea, the East–West 9 highway increases the GDP by 0.3% over the 30-year time period horizon, with 0.016% of the GDP due to the synergy effect. The East–West 9 highway has the largest synergy effect of US$0.164 billion per year on the manufacturing sector of Kwangju Metropolitan Area, resulting in a gain in a regional GRP per capita of US$15.88 per year. Since most synergy effects are generated in less developed regions, highway development can contribute to the reduction in regional disparities.     

The hierarchy of roads, the locality of traffic, and governance

This study investigates the use of road networks both within and outside of home jurisdictions (city (or town) and county of residence) by analyzing GPS data collected in the Minneapolis—Saint Paul metropolitan area. The study tracked volunteers' travel behavior to determine which roads (and thus which class of roads) users chose to accommodate their travel needs. More than half of the travel on county roads and city streets occur outside of one's home city, but most travel is within one's home county. The average share of travel distance in the home county is more than 70% for both county and city streets. The high share, which does not account for non-residents destined for the county to work or shop, e.g., implies that the free rider problem on city and county streets at the county level is minimal. Of particular concern is travel on city roads in cities other than one's own. To the extent that this is to go to a destination in that city, that travel is also local. However, because city and county roads are typically funded by those jurisdictions from land-based sources such as property taxes, through trips with neither end in the city through which they are traveling are in a very real sense “free riders”, and pose a problem. With growing trip lengths and emerging economies of scale in road management, it may be appropriate to consider moving more roads from township, town, or city level to the county level of government.     

A supply chain network design considering transportation cost discounts

This study addresses an integrated facility location and inventory allocation problem considering transportation cost discounts. Specifically, this article considers two types of transportation discounts simultaneously: quantity discounts for inbound transportation cost and distance discounts for outbound transportation cost. This study uses an approximation procedure to simplify DC distance calculation details, and develops an algorithm to solve the aforementioned supply chain management (SCM) problems using nonlinear optimization techniques. Numerical studies illustrate the solution procedures and the effects of the model parameters on the SCM decisions and total costs. Results of this study serve as a reference for business managers and administrators.     

Human Resource Management Practices and Trust in Public Organizations

Abstract

Trust has been acknowledged as a valuable managerial resource within organizations. Working as a lubricant of organizational functioning, trust reduces opportunistic behaviours while it increases voluntary compliance to organizational norms and rules as well as enhancing individual and organizational performance. Considering the importance of trust, it is worthwhile to explore what factors may help build trust within organizations. This research investigates whether perceptions of several human resource management (HRM) practices are associated with trust in government organizations. According to social exchange theory, HRM practices signal management's commitment to employees which in turn leads to greater trust in the organization. Using data from an employee survey conducted for the Georgia Department of Transportation in 2007, this research tests how employee perceptions of HRM practices are related to trust in three distinct levels of management in a large department of state government: trust in department leadership, trust in one's leadership team, and trust in one's supervisor. Binary logit analyses suggest that perceptions of HRM practices focusing on autonomy, compensation, communication, performance appraisal, and career development are associated with trust in public organizations. According to the result, those practices present variation in their leverage on trust in leadership at different levels.