地理学报 ›› 2008, Vol. 63 ›› Issue (12): 1246-1256.doi: 10.11821/xb200812002

• 城市地理 • 上一篇    下一篇

基于家庭的城市居民出行需求理论与验证模型

张文佳, 柴彦威   

  1. 北京大学城市与经济地理学系, 北京100871
  • 收稿日期:2008-04-18 修回日期:2008-09-19 出版日期:2008-12-25 发布日期:2010-08-03
  • 通讯作者: 柴彦威(1964-), 教授, 博士生导师。E-mail: chyw@pku.edu.cn E-mail:chyw@pku.edu.cn
  • 作者简介:张文佳(1985-), 男, 硕士研究生, 主要研究方向为行为地理和城市地理。E-mail: zhangwenjia@pku.edu.cn
  • 基金资助:

    国家自然科学基金项目(40671058)

Theories and Confirmed Model of Urban Resident's Travel Demand: Considering Intra-household Interaction

ZHANG Wenjia, CHAI Yanwei   

  1. Department of Urban and Economic Geography, Peking University, Beijing 100871, China
  • Received:2008-04-18 Revised:2008-09-19 Online:2008-12-25 Published:2010-08-03
  • Supported by:

    National Natural Science Foundation of China. No.40671058

摘要:

西方的城市居民出行需求研究已经由基于出行的分析转向基于活动的分析, 而国内的相关研究几乎空白, 特别是基于家庭的分析。基于出行的交通需求理论在交通规划中的不足, 强调活动分析法的理论基础和应用优势, 提出了基于家庭的活动—移动行为的理论并进行模 型验证。采用天津市居民时间利用日志调查的第一手数据, 以家庭为研究单元, 建立结构方 程模型, 在解读天津市民工作日的活动—移动模式的基础上, 验证了基于家庭的活动分析法 的理论。结果表明, 出行来源于相应的活动参与, 且活动分析法很好地解决了四阶段法等基 于出行的模型的一些不足。男女家长之间存在明显的活动—移动联系, 并且在非工作活动上存在联合参与行为。在考虑活动和家庭成员的间接效应时, 社会经济属性对居民出行时间的的总体效应和直接效应显著程度不一, 验证了基于家庭的活动分析法在出行行为研究上比传统基于出行的分析更加全面和深入。

关键词: 行为地理学, 交通地理学, 出行需求, 活动分析法, 结构方程模型, 天津市

Abstract:

From the 1950s to the present, the research perspective of urban resident's travel demand transits from trip-based to activity-based, while the study units divide Traffic Analytical Zone (TAZ) into the individuals and even household. This paper summarizes that there are many problems of trip-based travel demand model in theory and practice. For this, the researches of activity-based and intra-household interaction come into being. Activity-based approach is the theoretical foundation of household-based travel behavior analysis, while its own foundation dated from the seminal work of Chapin (1974) and Hägerstrand (1970). Many literatures have confirmed that activity-based analysis could solve the problems using trip-based models so commendably that its contents extend and deepen from the 1980s to now. Thought most activity-based models still focus on the theories, household-based travel models gradually become the mainstream in transport planning in Western countries. Researches of household-based travel behaviors require to synthetically consider the effect on individual's activity-travel behavior of family attribute, the constraint to individual's behavior from household demand, and the couple constraint between household members. These aspects are briefly reviewed in this article. We discover that trip-based approach is hardly competent for studies of intra-household interaction. By contrast, activity-based approach could employ utility-based model, rule-based model, micro-simulation model and complex multi-variable model, such as structural equation model, to better interpret resident's activity-travel behaviors. Finally, Tianjin's time-use dairy survey data are used to confirm the theories of intra-household activity-based approach and to deeply understand the Tianjin's activity-travel behaviors. We model the socio-demographics as exogenous, and activity duration time and relevant travel time as endogenetic in the context of structural equation model (SEM). The final LIRSEL model fits well. The estimate results confirm that travel is derived of activity participation, and activity-based approach can better solve the problems of trip-based model, such as linking the discrete travels through activities interaction. Moreover, significance of activity-travel interaction between household heads positively exists. In particularly, males dominate the outside work-related activities and females dominate the outside household-related activities, and they jointly participate in non-work activities in Tianjin's households. Furthermore, comparing the total effect and direct effect on household heads' travel time of socio-demographics, which takes both the indirect effect of activities and other household members into consideration, we found the significance is not the same and positively argue that the total effect contains more actual information about activity-travel behavior. In conclusion, theories and confirmed model have proved that activity-based approaches are much better than the trip-based one, especially in terms of the intra-household interaction. In the future research, we should spread this perspective and approach to a better understanding of the behavior and transportation in transiting China.

Key words: behavioral geography, transport geography, travel demand, activity-based approach, intra-household interaction, structural equation model (SEM), Tianjin City