原文传递 Simulation Estimation of Mixed Discrete Choice Models with the Use of Randomized Quasi-Monte Carlo Sequences: A Comparative Study.
题名: Simulation Estimation of Mixed Discrete Choice Models with the Use of Randomized Quasi-Monte Carlo Sequences: A Comparative Study.
作者: Sivakumar-Aruna; Bhat-Chandra-R; Okten-Giray
关键词: Forecasting-; Network-analysis-Planning; Network-equilibrium-model; Road-pricing; Travel-behavior; Travel-demand; Trip-chaining
摘要: The network equilibrium model is a useful tool for long-term transportation planning and is one promising alternative to the traditional four-step travel forecasting model. However, some issues with the model remain to be considered. For example, almost all variations of the model adhere to the traditional trip-based approach, in which trip chains made by users are treated as separate, independent entities in the analysis. This research aims to develop a simple, tractable model to overcome this problem. One proposed model is based on piston-type trip chaining, and another accommodates any other type of trip chaining and includes congestion phenomena. These proposed models have certain key features: they have been successfully formulated as convex minimization problems, so uniqueness and algorithm convergence are easily proved; traveler behavior is based on theoretically sound random utility models, which allows the benefit of transportation projects to be calculated such that it is consistent with travel demand forecasting; and optimal road pricing can be calculated even in large-scale networks. These models are examined with the use of simple network examples, with special attention paid to the effect of trip-chaining behavior at the level of second-best toll. In a simple two-destination network, the second-best toll of the trip-based model is lower than that of trip chain-based model, indicating one of the biases of the trip-based model.
总页数: Transportation Research Record: Journal of the Transportation Research Board. 2005. (1921) pp11-18 (5 Fig., 5 Tab., 19 Ref.)
报告类型: 科技报告
检索历史
应用推荐