GithubHelp home page GithubHelp logo

dtalite_software_release's Introduction

DTALite/NeXTA Software Introduction

####Contact: Dr. Xuesong Zhou at Arizona State University, U.S. Email: [email protected],[email protected]

###NeXTA: Network EXplorer for Traffic Analysis

Latest Software Release 09-15-2016:

#####NOTE: NeXTA is only compatible with a Microsoft Operating System.

Instruction for the use of NeXTA and DTALite:

Step 1: make sure that you have installed the Microsoft Visual C++ 2015 Redistributable Package (x86) for parallel computing in DTALite (https://www.microsoft.com/en-us/download/details.aspx?id=3387)

Step 2: make sure that you have installed Gnuplot Software for some visualization functions in NeXTA (http://www.gnuplot.info/)

Step 3: Download and unzip the NeXTA/DTALite software package.

If GIS data importing/exporting are needed for you, please download and unzip the NeXTA GIS software package and obey the related instructions.

#[A: NEXTA-DTALite Software Release]

###A1:Data structure and workflow of DTALite ###A2:NEXTA user guide ###A3:ABM+DTA Integration user guide ###[A4:OD demand estimation user guide] (https://docs.google.com/document/d/1UqFXVRbf0eOuq-liPXCDF6PVuSNjglWggLPbwT_y3TI)

#B: NEXTA-for-GIS Software Release. ###B1: User guide

###DTALite: Light-weight Dynamic Traffic Assignment Engine

White Paper: DTALite: A queue-based mesoscopic traffic simulator for fast model evaluation and calibration. Xuesong Zhou, Jeffrey Taylor. Cogent Engineering, Vol. 1, Iss. 1, 2014

Network EXplorer for Traffic Analysis (NEXTA) (Version 3) is an open-source GUI that aims to facilitate the preparation, post-processing and analysis of transportation assignment, simulation and scheduling datasets. NeXTA Version 3 uses DTALite, a fast dynamic traffic assignment engine, for transportation network analysis.

  1. Create, import, edit, store, export and visualize transportation network data. nexta nexta
  2. The NEXTA now support importing the following data
  • GIS shape files

  • CSV file based network data

  • TMC-based sensor data (e.g. Inrix or traffic.com)

NEXTA also can export to the following formatNEXTA

  • Google Earth KML

  • GIS shape file: node/link/zone, or convert user-defined CSV to shape

  • Subarea analysis to provide path flow pattern from Vissim simulation.

  1. NEXTA provides an excellent multi-project management interface with the following features.
  • Synchronized display

  • Click the same location across different networks

  • Compare link moe across different networks

  • Find link moe difference between baseline and alternative networks

  • Vehicle path analysis across different simulation results

  • Simulation vs. sensor data

  • Link based comparison

  • Path travel time comparison

  • Validation results, diagonal line display to identify outliers

  1. Import multi-day traffic measurement data and provide multi-criteria path finding results (mobility, reliability and emissions) nexta

  2. Simulation and visuailiing dynamic outputs

nexta nexta nexta

DTALite uses a computationally simple but theoretically rigorous traffic queuing model in its lightweight mesoscopic simulation engine. Its built-in parallel computing capability dramatically speeds-up the analysis process by using widely available multi-core CPU hardware. It takes about 1 hour to compute agent-based dynamic traffic equilibrium for a large-scale network with 1 million vehicles for 20 iterations.

The scope includes:

  • dynamic traffic assignment of large-scale network

    • typical network: 2000 traffic zones, 200000 links, 2-10 million vehiches
  • Network capacity planning

    • add/remove link, prepare basic data for optimizing signal timing (Synchro), and Micro simulation (Vissim)

    • export traffic capacity analising package (HCM, travel time reliability)

  • Operation of work zone area, application of tolling road/link: based on dynamic tolling strategy, value of time, Agent model

  • Emission analysis, traffic safety analysis

The DTALite package provides the following unique features using the AMS data hub format through NEXTA.

  • Unlimited number of link types

  • Unlimited number of demand types/demand files, 24 hour loading period: Flexible network conversion and linkage with GIS Shapefile (importing, script for mapping planning data to our data hub): save time, allow flexible number of link types and node types)

  • Flexible demand data format: 3-column (o,d, value), multiple columns (o,d, SOV, HOV, Truck, subtotal), matrix, with 15-min departure time interval. Agent file

  • Typical vehicle types: mapping from trip types to vehicle types, vehicle emission rates for different vehicle types, different ages Semi-continuous Value Of Time distribution

  • Common types of sensor data, link count, lane count, at user-defined interval, speed data, density data, route travel time data,

  • Unlimited number of safety prediction models, based on link volume, length, link type, # of intersections/drive ways per miles Movement-specific parameters (based on HCM/QEM methodology)

dta

DTALite/NEXTA package provides a wide range of data output.

  • Agent based trajectory

  • Link-based MOEs, band-width display, user defined offset

  • Safety and emission statistics based on dynamic assignment results.

  • Time-dependent path travel times for user-defined path

  • OD based MOEs

  • Select link analysis

  • Select path analysis

  • Subarea analysis

  • Summary plot based on a wide range of categories and MOEs

DTALite/NeXTA applications in The United States

maps

中文简介

NeXTA是一个开源的图形用户界面,主要功能包括建立路网及道路属性信息的输入,交通分配结果的处理和分析,仿真运行和数据输出。主要功能包括:

一、建网、导入、编辑、存储、导出以及可视化交通网

二、 NEXTA 支持以下交通网络数据类型的导入:

  • GIS shape files -- CSV based network files

NEXTA 支持以下交通网络数据类型的导出:

  • Google Earth KML
  • Subarea analysis to provide path flow pattern from Vissim simulation.

三、 NEXTA 提供高效的多项目管理界面,如:

  • 多窗口同步演示

  • 在不同网络场景下进行同一点的选择

  • 在不同网络场景下对同一条道路进行MOE比较

  • 比较基本网络和可选网络的道路MOE

  • 在不同仿真结果进行车辆轨迹分析

  • 仿真结果和实际采集数据比较

  • 以Link 为基础进行仿真结果分析

  • 以路径旅行时间为基础进行仿真结果分析

  • 仿真结果校核,用回归分析确定异常值

四、 导入多日的交通分析数据并且提供多条件路径查找功能

nexta

五、 行仿真和可视化动态交通仿真结果

nexta nexta nexta

DTALite 是轻量级的中观交通仿真引擎,它的特点在于计算机运行上简单但计算理论严格的交通仿真模型;其并行计算功能利用了目前普及的多核CPU配置大大加快仿真进程;例如,在一个大范围路网内进行以车辆为对象的动态交通分配,对一百万辆车进行20个循环迭代,只需要一个小时。 适用范围包括:

  • 大规模网络动态交通分配

    • 典型网路举例:2000交通小区,20000个路段,200万-1000万个车辆
  • 网络能力规划

    • 添加/移除车道,为信号配时优化( synchro )或微观仿真( VISSIM )准备基础数据

    • 导出交通能力分析包(HCM高速能力、旅行时间可靠性)

  • 施工区的运用,道路价格的运用: 基于动态费用、时间异构值的Agent模型

  • 排放研究交通安全研究

DTALite 与NeXTA AMS 数据库格式的优势:

-	道路类型数量没有限制Unlimited number of link types

-	交通需求的类型和文件数量没有限制,可以进行24小时的数据加载,灵活的路网/shape file转换 

-	交通需求数据类型可以是多种格式:3列OD数据、多列数据(如:SOV, HOV, Truck, subtotal)、矩阵、15分钟间隔的交通需求矩阵、以车辆为单位的交通需求数据

-	典型车辆类型:将trip类型转换为车辆类型,不同车辆类型的排放率,车辆的使用年限等

-	Value Of Time 分布 

-	检测数据的类型:路段检测数据,车道检测数据,用户自定义的检测时间间隔车速检测数据,密度检测数据,出行时间数据

-	特定的车辆转向参数设定

dta

DTALite/NEXTA package 提供如下数据分析结果:

 -	车辆轨迹Agent based trajectory 

 -	路段MOEs,路段宽度显示,用户定义的路间offset 

 -	基于动态分配结果的安全性和排放统计结果 

 -	用于可定义路径的动态路径出行时间 

 - 	以OD为对象的MOEs 

 -	路段选取分析 

 - 	路径选取分析 

 -	局部地区分析

 -	MOEs和其它多种类别的图形化结果展示

DTALite/NeXTA 在美国应用项目的分布

maps

dtalite_software_release's People

Contributors

ddiqqi avatar tonyliu2015 avatar xzhou99 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.