基于随机边界函数的出行时间分析

出行时间是居民出行特征的重要分析指标.研究从出行时间预算入手,将随机生产边界函数用于对出行预算和出行时间边界的描述,建立了出行时间边界和出行者社会经济属性、家庭结构、出行者日活动安排及居住地用地特征之间的数量关系模型,给出了出行时间预算和出行时间边界的相互关系.以居民出行调查数据为来源,对通勤者和非通勤者分别建立随机出行时间边界模型,分析了工作活动影响下通勤者出行时间与非通勤者出行时间对各类影响因素的不同响应程度,比较了通勤者与非通勤者出行时间预算和出行时间边界的不同分布规律.研究表明,随机生产边界模型的组合误差结构为研究个体出行时间差异提供了很好的分析手段.

Travel Time Analysis Based on Stochastic Frontier Method

Travel time is one of the most important issues in travel behavior analysis. Based on the concept of travel time budget, the stochastic production frontier model is employed to describe the unobserved travel time frontier, which is defined as the maximum amount of time that an individual is willing to allocate to travel in a day's time. The model estimates the relative importance of influence factors from the traveler's socio-demographic characteristics, family structure, activity pattern, and land use. And the relation between travel time budget and travel time frontier is expressed as a ratio. Then travel time frontier models are established for commuters and non-commuters using travel survey data. Model estimation results showed that work activity made travel time pattern of commuters quite different from that of the non-commuters and that travel time frontiers of the two kinds of travelers had different sensitivity to those influence factors in the model. The notion of travel time frontier provides a powerful framework for understanding induced travel effects from an activity-based time use allocation perspective. And the combination structure of the random effect in the stochastic production frontier model presents a useful tool for the analysis of individual travel time variation.