Data Archives of Intelligent Transportation Systems Used to Support Traffic Simulation

Abstract

Collecting data for traffic simulation is expensive, particularly for large simulated systems. When traditional methods are used, data are normally collected for only one day or a few days and may not represent variations in traffic demands and conditions throughout the year. Collected data usually are imperfect, and additional efforts are needed to compensate for missing and erroneous data and to resolve data inconsistencies. In recent years, agencies have started archiving data collected by intelligent transportation systems (ITS). The ITS data archives can provide cost-effective, detailed information for the development and calibration of simulation tools if procedures are used to ensure data quality and to allow optimal categorization and use of the archived ITS data. An effort to develop a series of data manipulation procedures for the use of ITS data archives in support of simulation modeling is discussed. These procedures allow the extraction of collected volume data from ITS data archives, automatic identification of temporal patterns in the data, automatic segmentation of daily demands into dynamically captured subperiods to best fit variations in demand, resolution of possible spatial inconsistencies in the data, and estimation of missing volumes. The developed procedures have been implemented as an automated tool for populating simulation models. The procedures and the developed tool can easily be adapted by other traffic agencies to interface with their ITS data archives.

Estimation of Diversion Rate during Incidents on Basis of Main-Line Detector Data

“…The utilized k-means clustering algorithm (18) categorizes the demand data for different days into patterns, on the basis of the similarity of the time series of volume counts on different days. This is an iterative partitioning algorithm that minimizes the sum of time series distances to cluster centroids, summed over all clusters.…”

Interval data‐based k ‐means clustering method for traffic state identification at urban intersections

“…In brief, although many studies have examined intersection traffic state identification, few have considered the impact of traffic flow uncertainty on identification results. Recently, clustering analysis [9,10] has been increasingly popular for identifying traffic states on freeways, because of its reduced calculation complexity and iterative error-correction capabilities. Xia and Chen [11,12] proposed a freeway traffic state identification method based on two clustering algorithms for determining the clustering centres and dispersion distributions of traffic states.…”

Analyzing the Impact of Traffic Congestion Mitigation: From an Explainable Neural Network Learning Framework to Marginal Effect Analyses

“…By mining information from different data sources (e.g., traffic count sensors, mobile phone service records, floating car data, and automatic vehicle identification (AVI) data), planners and decision-makers hope to gain deep insight into human mobility patterns, which will accordingly increase the confidence level and reduce the uncertainty of transportation strategy evaluations [3,4,5,6,7,8]. Over the past several decades, many researchers have highlighted the need for a fully integrated connection from big data sources to pattern recognition and the deployment of final traffic demand/control strategies [9,10,11,12].…”