权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Travel Behaviour Analysis Model with Soft-computing Techniques

基于软计算技术的出行行为分析模型

基本信息

批准号：
10650524
负责人：
AKIYAMA Takamasa
金额：
$ 2.11万
依托单位：
Gifu University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
1998
资助国家：
日本
起止时间：
1998 至 1999
项目状态：
已结题

项目摘要

The travel behaviour analysis model is proposed to describe the multi-stage decision of trip maker in the research. The model is applied to evaluate the transport policy. Firstly, the travel pattern descriptive model is proposed on the basis of activity distribution to demonstrate the key structure of the travel behaviour model. Secondly, the extension of the model description is mentioned. The application to the evaluation of transport policy is discussed as well. The research results are summarized as follows:1. The applicability of soft-computing technique is mentioned to investigate the knowledge based modelling for modal choice and route choice with fuzzy reasoning as well as neural network (NN). The descriptive models with uncertain events in human decision for travel behaviour are summarized.2. The six activity based models in the stages of travel are constructed to describe the individual daily travel patterns. Particularly, fuzzy neural network (FN) model is mainly applied to create the model with the combination between fuzzy reasoning and neural network.3. The hybrid model is proposed to combine the stochastic and soft-computing approaches complementally. The fuzzy logit model (FLM) is constructed with combining between fuzzy reasoning and logit model. The travel behaviour can be easily described with randomness as well as fuzziness.4. The evaluation of transport policy is mentioned to consider the applicability of the proposed model with person trip survey data. It would be confirmed that the model could describe the travel patterns corresponding to the change of transport environments5. The research results in the above stages have been reported in international and domestic conferences as JSCE and so on. The further study topics are summarized as well.

研究中提出了出行行为分析模型来描述出行者的多阶段决策。应用该模型对交通政策进行了评价。首先，提出了基于活动分布的出行模式描述模型，展示了出行行为模型的关键结构。其次，提出了模型描述的外延。并对其在交通政策评价中的应用进行了探讨。1.提出了软计算技术的适用性，研究了基于模糊推理和神经网络的模式选择和路径选择的知识建模方法。总结了人类出行行为决策中存在不确定性事件的描述性模型。构建了六个出行阶段的基于活动的模型来描述个体的日常出行模式。其中，模糊神经网络模型主要用于建立模糊推理与神经网络相结合的模型。混合模型的提出是为了将随机方法和软计算方法互补地结合起来。将模糊推理与Logit模型相结合，构建了模糊Logit模型。出行行为既具有随机性，又具有模糊性。在交通政策的评价中，考虑了所提出的模型在人员出行调查数据中的适用性。可以肯定的是，该模型可以描述与运输环境变化相对应的出行模式。上述阶段的研究成果已在JSCE等国际国内会议上发表。并对进一步的研究课题进行了总结。