Development of the traffic system at a disaster with Multi-Agent Simulation and reliability analysis of travel time on road network

在本研究中，我们开发了一个交通模型，该模型表达了驾驶员在多智能体仿真系统中的路线选择。我们对自己开发的模型进行了调整，我们开发了评估汽车运行时间可靠性的模型，并开发了一个虚拟模型网络和一个真实的道路网络，用于道路网络，当发生大地震时，我们开发了预测多发火灾同时爆发和火灾蔓延情况的系统。此外，我们评估了地震灾害时紧急车辆的进入特征，并通过路网可靠性分析考虑了每个消防部门地点的紧急车辆（消防泵车，救护车）的车辆配置。本文利用可靠性分析系统，开发了对正常时间（正常情况下）路网紧急车辆通行特性较低的区域进行预测的系统。我们预测了道路网络在地震灾害中是如何被破坏的，并开发了这个模型来建立道路网络。但是，对于避难交通行为的预测模型，我们建立了一个多智能体自适应模型。我们开发了一个模型来预测它是如何发生的，以及随着时间的推移如何传播，在地震发生的同时，经常发生火灾。利用上述系统，可以计算出消防车辆必须提前到达的区域。我们利用Petri-net模拟器开发了一个交通微观模拟系统，模拟地震灾害时普通车辆和急救车辆的运行情况。分析了在正常时间和大地震发生后，利用ITS技术进行最合适的课程教学和最合适的信号控制效果。

In this study, we developed the traffic model who expressed a route choice of a driver with a multi-agent simulation system. We adapted ourselves in the model that we developed the model that evaluated reliability for running time of a car and developed for a virtual model network and a real road network for a road network when a big earthquake occurred, we developed the system which predicted outbreak of a simultaneous frequent occurrence fire and the situation of spread of a fire. In addition, we evaluated access characteristics of an emergency vehicle at the time of an earthquake disaster, and having considered the allocation of cars of an urgent vehicle (a firefighting pump car, an ambulance) of each fire department place by reliability analysis of a road network.We developed the system which predicted the area where an access characteristic of an urgent vehicle for a road network of normal time (in normal) was low by using the reliability analysis system. And we predicted how a road network was damaged at the time of an earthquake disaster and developed the model in order to built a road network. But, about the prediction model of a refuge traffic action, we built a system with a multi-agent of adaptation model. We developed the model that predicted how it occurred and how spread with time where a simultaneous frequent occurrence fire at the time of an earthquake occurred.We get possible to calculate the area where a firefighting vehicle must arrive at early by utilizing an above system. We developed a traffic micro-simulation system with Petri-net simulator to run a general vehicle and a first aid vehicle at the time of an earthquake disaster. We analyzed the most suitable course instruction that utilized ITS technology and an effect of the most suitable signal control for normal time and after big earthquake outbreak.