TraMPa: Transportation Modeling Using Publicly Available Data: An Evolution for Model Input Data

TRAMPA调查使用公开的数据来开发交通模型。拟议的项目建立在各种可用开放数据源可获得的不断增长的数据量的基础上。分析将基于不同级别的开源数据可用性来进行验证、敏感性分析和元模型分析，以调查基于真实世界场景的模型性能。该项目旨在尽可能地使用开放可用数据来进行模型开发、模型校准和模型验证。一个核心发现将是开放可用的数据在多大程度上可以在多大程度上取代运输建模中的传统数据源。本项目将测试的数据来源包括所有可以在线公开、合法和免费获得的与交通相关的数据。这包括可通过应用程序编程接口(API)、网络抓取或直接下载(开放数据)访问的数据。相比之下，其他研究人员可能无法获取传统数据，或者需要花钱。通过探索开放数据的潜力，这项提议旨在建立证据，证明开放数据如何支持更透明和更可重现的运输建模方法。开放式模式将与慕尼黑大都市区、美国马里兰州和南非开普敦实施的传统模式进行比较。通过将两个传统模型与相同研究领域和相同情景下的开放模型(及其混合模型)进行比较，将有可能量化开放数据取代传统数据源的潜力。可以看出，开放模型甚至会比传统模型表现得更好，因为假设用于开放模型的数据比用于传统模型的传统数据具有较少的偏见和较少的错误填充。为了更好地了解开放数据在开放运输模型定义中可能产生的影响，将通过验证、敏感性分析和元分析来测试传统数据和开放数据的不同组合，以审查根据关键绩效指标估计的模型的性能。开放式模型与传统模型(及其混合体)的这种比较将有助于探索数据可用性和渗透率水平对运输模型的影响。如果这项研究表明，开放模型的表现优于传统模型，那么对数据收集、模型设计和所需资金的影响可能是巨大的。

TraMPa investigates the use of openly available data for the development of transportation models. The proposed project builds upon the ever-growing amount of data that can be obtained by a diverse number of open data sources available. The analysis will be based on different levels of open source data availability to perform validation, sensitivity analyses and meta-model analyses for the investigation of the model performance based on real-world scenarios.This project aims at using openly available data as much as possible for model development, model calibration and model validation. A core finding will be how far openly available data may replace traditional data sources in transport modeling. Data sources to be tested in this project include all transport-related data that can be obtained openly online, legally and free of charge. This includes data that can be accessed through an Application Programming Interface (API), web-scraping or direct download (open data). Traditional data, in contrast, may be inaccessible to other researchers or cost money. By exploring the potential of open data, this proposal aims at establishing evidence how open data may support a more transparent and reproducible transport modeling approach. An Open Model will be compared with a Traditional Model that has been implemented for the Munich metropolitan area, the State of Maryland in the USA and Cape Town in South Africa. By comparing two traditional models with open models (and hybrids thereof) for the same study areas and the same scenarios, it will be possible to quantify the potential of openly available data to replace traditional data sources. It is perceivable that the Open Model will even perform better than the Traditional Model, as it is hypothesized that the data used for the Open Model is less biased and less error-filled than conventional data used for the Traditional Model. In order to allow for a better understanding of the effect that open data could have in the definition of the Open Transportation Model, different mixes of traditional data and open data will be tested by validation, sensitivity analyses and meta analyses to examine the performance of the models estimated based on Key Performance Indicators (KPIs). This comparison of the Open Model with the Traditional Model (and hybrids thereof) will help explore the impact that levels of data availability and penetration have on transportation models. Should this research show that the Open Model outperforms the Traditional Model, the implications on data collection, model design and required funding could be substantial.