基于分层logit模型的旅游交通方式选择研究

通过对旅客进行旅游交通方式选择的量化研究,得出各交通方式的选择概率,有利于各旅游地进行规划.首先通过分析过去与现在人们旅游交通工具的选择的不同,得出影响交通工具选择的影响因素.然后根据各个影响因素建立分层logit预测模型,对水平1和水平2进行标定,得出各个影响因素的参数值,从而得到各交通方式的选择概率.再利用该建模对近、中、远3种距离模式进行交通工具选择的预测,最后得到的预测结果与实际情况较为符合,相差的概率在0.05以内.      

A nested logit model of mode choice for inland movement of export shipments: A case study of containerised export cargo from India

Predominance of containerised mode of transport for export of general cargo has influenced the mode choices on the inland segment. Prior to containerisation the general cargo for export moved to ports invariably on road in break bulk form. Growth of containerisation led to development of inland terminals enabling an exporter to obtain export clearance at the inland points. This process facilitated an exporter to move cargo either in break bulk or in containerised form, and as inland terminals facilitated aggregation an exporter could move cargo either by road or rail. Thus containerisation increased the inland transport options for an exporter from one to three. The paper develops discrete choice models in the Indian context for inland movement of containerised export cargo. Primary data was collected from 124 export firms through a structured questionnaire. This data is supplemented with data from transport firms and terminal operators. Using the above data, a nested logit model with rail container and road truck in one branch was found to predict the mode choice behaviour best. Amongst the transport attributes, total cost and total transit time were found to be influencing the mode choices significantly. The study however did not find the effect of reliability and loss and damage significant. In addition, the model also predicts that non transport attributes - like the percentage of letters of credit that materialise with inland way bills and the value of export benefits that materialise after export - are important in the mode choice decision.     

A study of factors influencing the severity of road crashes involving drunk drivers and non drunk drivers

In this study, an attempt has been made to develop Multinomial Logit (MNL) model by analysing the drunken and non drunken drivers involved in road crashes on Indian highways. Multinomial Logit model has been deployed to assess the influence of various parameters like vehicular, environment and geometric factors on the set of drivers who were found to be drunk at the time of getting involved in the road crash and those who were not under the influence of alcohol at the time of meeting with the road crash. The total economic cost of road crashes in the case of non-drunk driver road crash is Rs. 1046.27 million whereas in the case of drunk driver road crashes it is estimated to be Rs. 204.50 million. Further, it can be observed that economic cost of drunk driver road crashes is varying from 13 to 19 % across different types of road crashes.     

A multinomial logit car use model for a megacity of the developing world: Istanbul

The increase in motor vehicle use is one of the important factors that cause traffic congestion, especially in megacities. Thus, the reasons behind this increase require serious attention. This paper offers an analysis of this kind, for a megacity from the developing world, Istanbul. A stratified multinomial logit model accounting for the availability of a second vehicle in the household is estimated for a sample drawn from a questionnaire to gather information of actual car use in Istanbul. This estimation is only possible through a unique data generation process that converts actual preferences into a choice study setting. In addition, a simulation study, generally utilized in the analyses of discrimination between certain layers of society, and a scenario analysis related to changes in income are also included in the paper for a better understanding of the nature of the topic. The results show that the behavior of households with a second vehicle available and not available varies significantly due to household, individual and professional-related characteristics.     

Rear‐end crash potential estimation in the work zone merging areas

This paper proposed a methodology to estimate rear‐end crash potential of the merging vehicles traveling in the merge lane, on the basis of the traffic data extracting from the available videotapes. First, we developed a binary logit model to identify drivers' merging behavior in the work zone merging area. Subsequently, the occurrence potential of rear‐end crash based on time‐to‐collision was computed between the merging vehicle and its neighboring vehicles. The overall crash potential of the merging vehicle was finally determined. It was found that the crash potential decreases with the remaining distance to work zone. Moreover, there will be a rear‐end crash potential of 4.0% if the merging vehicle fails to complete merging at the end of work zone merging area. If the merging vehicle takes an early merge, there will be a lower rear‐end crash potential (1.2%). These findings suggest that we should encourage merging vehicles to take early merges for improving the traffic safety in the work zone merging areas. Copyright © 2012 John Wiley & Sons, Ltd.     

基于静态博弈论的物流配送中心选址优化研究

在叙述物流配送中心选址相关知识的基础上,站在投资人的立场,把物流配送中心的选址问题归结为收益最大问题,建立物流配送中心布局和多维logit选址优化模型,运用静态博弈论的相关理论,采用C++编程作为技术支持,通过模型求解物流配送中心的最佳布局和选址位置,并进行算例分析。     

基于混合Logit的居住地和就业地选择行为模型

为了剖析城市居民在选择居住地和就业地时的决策机理,利用混合Logit模型建立了城市居住地选择模型和就业地选择模型.针对北京28个住宅小区进行了问卷调查,共发放2 520份问卷,回收1 994份有效问卷,利用调查数据对模型参数进行了估计.结果表明,＂出行距离＂是影响居民在进行居住地与就业地选址时的首要因素,除此之外,＂轨道站点数量＂是影响居住地选择的主要因素,＂出行成本＂则是影响就业地选择的主要因素.     

Leave the expressway or not？ Impact of dynamic information

This study investigates drivers＇ diversion decision behavior under expressway variable message signs that provide travel time of both an expressway route and a local street route. Both a conventional cross-sectional logit model and a mixed logit model are developed to model drivers＇ response to travel time information. It is based on the data collected from a stated preference survey in Shanghai, China. The mixed logit model captures the heterogeneity in the value of ＂travel time＂ and ＂number of traffic lights＂ and accounts for correlations among repeated choices of the same respondent. Results show that travel time saving and driving experience serve as positive factors, while the number of traffic lights on the arterial road, expressway use frequency, being a middle-aged driver, and being a driver of an employer-provided car serve as negative factors in diversion. The mixed logit model obviously outperforms the cross-sectional model in dealing with repeated choices and capturing heterogeneity regarding the goodness-of-fit criterion. The significance of standard deviations of random coefficients for travel time and number of traffic lights evidences the existence of hetero- geneity in the driver population. The findings of this study have implications for future efforts in driver behaviormodeling and advanced traveler information system assessment.     

考虑共享停车的用户出行选择研究

为探讨共享停车的出现对用户出行选择的影响,将出行者划分为共享停车和普通停车两种用户类型,分别构建停车阻抗函数,并基于分层Logit(NL)建立停车和路径联合选择结构模型.进一步提出带有两种停车方式的随机用户均衡模型,并改进相继平均算法对模型进行求解,通过算例分析发现预约费用从4元变化到12元时,共享停车需求量从378辆...     

基于极值分布的离散选择模型尺度参数

为了准确把握尺度参数在模型建模、分析以及应用过程中的作用,以基于极值分布的Multinomial Logit模型和Nested Logit模型为研究对象,从建模的基础--误差项的分布形式入手,通过模型形式和误差项性质两方面研究了尺度参数在两种模型中的特性,并对其进行了对比分析.分析结果表明:尺度参数不仅限制了模型效用函数的尺度,而且也反映了误差项的方差水平,但对于以上两种不同的模型,其在尺度限制、误差项方差水平及独立性、自身取值范围等方面均表现出不同的特性,Multinomial Logit模型的尺度参数反映了有关整个效用确定项的误差项方差水平,而Nested Logit模型的尺度参数只是反映了与某一选择枝有关的部分效用确定项的误差项方差水平.