董治
随着城镇化的快速发展，城市群不断发展成熟，成为拉动区域社会经济发展的主导力量。城市群经济发展与城市群商务公务活动发生密切相关，随着城市群商务公务联系日益繁荣，需要进一步优化城市群商务公务出行的交通环境，有效的促进城市群商务公务出行者的移动。城市群交通发展应转变发展思路，注重促进交通系统服务水平的提高。深入了解出行者对于城市群商务公务出行的服务要求，才能探索符合其内在需求的交通发展模式，制定有效的相关政策措施。深入研究城市群商务公务出行行为，有助于挖掘城市群商务公务出行者出行偏好以及需求动机，把握出行特征与活动规律。
本研究在全面梳理国内外相关文献与界定研究关键要素的基础上，主要从行为调查、行为特征、行为机理与行为改变四个方面对城市群商务公务出行行为进行了研究。
首先，本研究梳理了出行行为调查设计的基本原则与一般流程，分析了主要调查方法的优缺点。重点讨论了长三角城市群出行行为调查，介绍了该调查问卷设计内容与方法，调查抽样与实施、数据清理等工作。通过调查工作的开展，总结我国城市群出行行为调查中存在的问题，为进一步改进调查设计提供了参考。
其次，基于长三角城市群出行行为调查数据，采用数理统计的方法，与其他目的出行特征对比的方式，从出行频率、出行时间以及出行方式三个方面对城市群商务公务出行行为特征进行了描述性分析。基于出行态度与生活方式因子进一步对城市群商务公务出行者进行群体细分，将城市群商务公务出行者分为轻松型、适应型以及压力型，分析并比较三类人群在社会经济属性、出行特征及出行者需求方面的差异。
再次，基于长三角城市群出行行为调查数据，从最近一次商务公务方式选择以及年方式使用率两个层面对城市群商务公务出行方式选择行为进行了研究，深入分析其选择机理。基于多项logit（MNL）模型，构建最近一次城市群商务公务方式选择初步总体模型分析主要影响因素。进一步按照商务公务出行决策主体的差异将样本分成有自主选择权的出行者及无自主选择权的出行者，分别进行分类模型构建，比较影响因素的差异。基于分类模型建立联合模型，将其与初步总体模型进行比较，说明了联合模型在城市群商务公务出行方式选择模型构建中的重要性和有效性。此外，将比例logit（FMNL）模型运用在对年方式使用率分布的分析中。使用弹性分析方法，对显著公共变量“便捷性”与“可靠性”进行模拟分析，为政策发展、措施制定提供支持。结合城市群商务公务出行者需求特征，从基础设施、运行管理、人文关怀以及信息技术四个方面分析提升方式服务水平策略，提出信息技术是未来城市群交通服务发展的重要方向。
最后，基于美国加州北部城市群美铁出行调查数据，研究了移动互联信息服务环境下城市群商务公务出行者出行的新特征，通过构建Ordered logit模型分析城市群商务公务出行者旅途中对Wi-Fi服务使用需求。基于加权的多元线性回归模型，研究移动互联信息服务对城市群商务公务出行者使用美铁出行的影响，提出分析影响程度的方法，计算并比较了Wi-Fi服务对不同目的出行者使用美铁出行频率的影响程度。研究表明，美铁上提供Wi-Fi服务后，城市群商务公务出行者使用美铁的频率增长最大，说明该项服务对于城市群商务公务出行者有较大的吸引力。
关键词：城市群，商务公务活动，出行行为，离线选择模型，信息通信技术
With the rapid development of urbanization, the urban agglomerations have been formed and become the dominant force in stimulating regional socio-economic development. Business activities in the urban agglomeration are related to the economic development closely. With more and more business communications, it is necessary to improve the business travel environment in the urban agglomeration and accelerate the effective move of business travelers. The development of transportation in the urban agglomeration should focus on promoting the level of service of transport system. Understanding the demand of business travelers is the base to explore the transportation pattern and formulate relevant policy measures. Travel behavior analysis helps to investigate the individual preference and motivation of the travelers, and demonstrates the travel characteristics and activity patterns in the urban agglomeration. Therefore, it is necessary to study the business travel behavior in the urban agglomeration further.
Based on the comprehensive literature review and the definition of key elements, this study analyzed the business travel behavior in the urban agglomeration from the four aspects, which are behavior survey, travel characteristics, behavior mechanism and behavior change.
Firstly, this study introduces basic principles and general process in the travel behavior survey design, and analyzes the advantages and disadvantages of different main survey methods. It focuses on the discussion of travel behavior survey in the Yangtze River Delta urban agglomeration. The design of the questionnaires is introduced and survey sampling, survey implementation and data cleaning are elaborated as well. It also concludes the problems in conducting the travel survey in the urban agglomeration and provides support for improving the survey design in the future.
Secondly, based on data of the travel survey in the Yangtze River Delta urban agglomeration, using the method of statistical analysis and comparison, the study describes the business travel characteristics from three aspects, which are travel frequency, travel time and travel mode. Then, through the analysis of evaluation of importance, it shows that the demand of business travelers in the urban agglomeration is mainly focused on accessibility, fast and convenient. At the basis of attitude and lifestyle factors, the business travelers were divided into easy type, adaptable type and pressure type. There are differences in the socio-economic attributes, travel characteristics and demand of travelers among the three groups.
Thirdly, based on data of the travel survey in the Yangtze River Delta urban agglomeration, this study analyzed the mode choice behavior of business travelers in the agglomeration from the aspects of the recent business travel and annual business trips, and investigated the mechanism of choice. A multinomial logit model based on the pooled data was estimated initially. Then we built two segmented models according to whether the travelers made the choice themselves or not and obtained the best specification for each segment. We found that some significant variables in the two segments are different. Finally, we combined the two specifications into a single model by creating segment-specific versions of each variable and it was proved superior to the pooled model, showing its necessity and importance. Besides, a fractional multinomial logit model were adopted to estimate the allocation of business trips across three main intercity modes in the past year. Elastic analysis method is used to simulate the significant variables convenience and reliability. Combining the demand characteristics of business travelers in the urban agglomeration and the simulating results, this study concludes four important aspects related to promoting the level of service, which are infrastructure, operation and management, humanities and ICT（Information and Communication Technologies）, and proposes that ICT is an important direction for development of urban agglomeration transport services in the future.
Finally, based on data of the travel survey on Amtrak in the North California urban agglomeration, this study analyzed the ew travel characteristics of business travelers in the urban agglomeration under the environment of ICT are discussed. The usage of Wi-Fi by business travelers while traveling is analyzed by estimating an ordered logit model and the results show that there was great demand of using Wi-Fi in the business journey. Moreover, a weighted multiple linear regression model is built to analyzed the impact of Wi-Fi on passengers’ trip frequency of using Amtrak. Using the estimated parameters from the model, the specific impacts are calculated for different groups divided by trip purpose. The results show that the impact of Wi-Fi on business passengers’ trip frequency of using Amtrak is the highest, implying the importance of ICT service on appealing business travelers in the urban agglomeration.
Key Words: urban agglomeration, business activities, travel behavior, discrete choice model, ICT
