基于AVC最小的交通行为特征研究方法

发布时间：2019-03-17 18:00

【摘要】：为克服现有意愿调查(stated preference,SP)实验设计方法标定结果精度低的缺点,以Logit模型的渐进协方差矩阵行列式最小为目标,对选择枝属性水平组合进行迭代计算,得到相应的选择枝属性水平组合方案,设计了一种新的出行意愿调查实验设计方法.以某城市中心城区到卫星城新增地铁交通方式为例,通过分析拟合优度、参数估计标准差,对比了新方法与正交设计法的模型标定结果精度.算例结果表明,与正交设计法相比,采用本文方法得到的参数估计标准差最小,拟合优度达到0.15,提高了0.05,提高了Logit模型参数标定精度,且具有选择枝属性组合数量弹性选择的优点.
[Abstract]:In order to overcome the disadvantage of low precision of calibration results in the existing stated preference,SP experimental design methods, the iterative calculation of the horizontal combination of selected branch attributes is carried out with the aim of minimizing the determinant of the progressive covariance matrix of the Logit model. In this paper, a new experimental design method of trip intention investigation is designed, and the corresponding combination scheme of branch attributes is obtained. Taking the new subway traffic mode from the center of a city to the satellite city as an example, the accuracy of calibration results between the new method and the orthogonal design method is compared by analyzing the goodness of fit and the standard deviation of parameter estimation. The numerical example shows that compared with the orthogonal design method, the standard deviation of parameter estimation obtained by this method is the smallest, and the goodness of fit is 0.15, which improves 0.05 and improves the calibration accuracy of Logit model parameters. Moreover, it has the advantage of selecting the combination of branch attributes and quantitative elasticity.
【作者单位】：西南交通大学交通运输与物流学院;现代城市交通技术江苏高校协同创新中心;
【基金】：国家自然科学基金资助项目(50908195,51178403) 教育部新世纪优秀人才支持计划资助项目(NCET-13-0977) 中央高校基本科研业务费专项资金资助项目(SWJTU11CX080,2682014CX130) 高等学校博士学科点专项科研基金资助项目(20130184110020)
【分类号】：U491

【参考文献】