基于人群出行行为的疏散总需求预测

发布时间：2018-03-29 01:20

本文选题：紧急疏散　切入点：有序多分类Logistic回归　出处：《兰州交通大学》2014年硕士论文

【摘要】：近年来,随着国内外城市化进程日益加快,许多城市发生过多次多种类型的突发公共事件，特别是自然灾害和事故灾害。这些突发公共事件会导致了大规模人群疏散，为应对意外发生的突发公共事件，必须做好突发公共事件的应急交通疏散研究工作。在突发公共事件时，紧急疏散会涉及大批人群、车辆的集体性出行，其中作为疏散对象的人在疏散过程中处于中心地位，其出行行为直接决定着疏散期间的交通需求。因此，需要深入研究大规模疏散中人群出行行为的机理和规律，并在此基础上科学、准确、定量地预测疏散总需求量，从而确定所需要的救援物资以及疏散过程中所需要的交通工具，保证以最快的速度将受灾人员和救援物资运送到避难场所，以达到在最短时间内完成应急响应任务，减少人员伤亡和财产损失的最终目的。本文首先通过阅读大量文献，对于发生突发公共事件时，紧急疏散人群的出行行为进行充分地了解，分析疏散人群的决策行为表现，设计了面向人群出行行为的调查问卷，然后组织人员对于调查问卷进行填写，回收后剔除无效问卷，将有效的问卷数据录入SPSS，进行初步的统计分析。其次，基于录入SPSS的问卷数据，进行相关性分析。在外部条件给定情况下，采用SPSS中的卡方检验，来确定与家庭疏散决策相关的因素，并结合它们的相关次数及相关度大小，得出在外部条件给定情况下，对于家庭疏散决策的影响程度大小排序。最后，基于家庭疏散决策相关的因素的相关性分析，在不同外部环境条件下，建立家庭疏散决策回归模型，并根据回归模型预测的响应概率，确定家庭决定疏散的概率阈值（累积概率），，并通过预测每一个家庭疏散决策的累加概率，将其与阈值进行比较，得出该家庭是否参与疏散，累加参与疏散的家庭，从而求出疏散总需求。
[Abstract]:In recent years, with the accelerating process of urbanization at home and abroad, many cities have occurred many kinds of public emergencies, especially natural disasters and accidents. These sudden public events will lead to large-scale evacuation of people. In order to deal with unexpected public emergencies, it is necessary to do a good job in the study of emergency traffic evacuation of public emergencies. When public emergencies occur, emergency evacuation will involve a large number of people and the collective travel of vehicles. The person who is the object of evacuation plays a central role in the evacuation process, and its travel behavior directly determines the traffic demand during the evacuation. Therefore, it is necessary to deeply study the mechanism and law of the crowd travel behavior in the mass evacuation. And on this basis, scientific, accurate, quantitative prediction of the total evacuation demand, so as to determine the need for relief supplies and vehicles needed in the evacuation process, To ensure that disaster victims and relief supplies are transported to places of refuge as soon as possible in order to achieve the ultimate goal of completing emergency response tasks and reducing casualties and property losses in the shortest possible time. First of all, through reading a large number of documents, this paper fully understand the travel behavior of emergency evacuation crowd, analyze the behavior of decision behavior of evacuation crowd, and design the questionnaire of crowd oriented travel behavior. Then the organization staff fill out the questionnaire, withdraw the invalid questionnaire, input the valid questionnaire data into SPSSs, and carry on the preliminary statistical analysis. Secondly, based on the questionnaire data of SPSS, the correlation analysis is carried out. Under the given external conditions, the chi-square test in SPSS is used to determine the factors related to the family evacuation decision, and combined with their correlation times and correlation degree. The order of the degree of influence on the family evacuation decision is obtained under the given external conditions. Finally, based on the correlation analysis of the factors related to the family evacuation decision, the regression model of the family evacuation decision is established under different external conditions, and the response probability is predicted according to the regression model. Determine the probability threshold (cumulative probability) of the family decision to evacuate, and by predicting the cumulative probability of each family evacuation decision, compare it with the threshold value, and obtain whether the family is involved in the evacuation, and the cumulative family participating in the evacuation. Thus, the total evacuation requirements are calculated.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U491

【相似文献】