Pattern Recognition Using Clustering Algorithm for Scenario Definition in Traffic Simulation-based Decision Support Systems

摘要

The decision and selection of input scenarios to be tested in connection with a particular study are criticalimportant to the decision outcomes of the evaluation process. This paper presents a scenario clusteringapproach intended to mine historical data warehouses to identify appropriate scenarios for simulation as apart of an evaluation of transportation projects or operational measures. As such, it provides a systematicand efficient approach to select and prepare effective input scenarios to a given traffic simulation model.The scenario clustering procedure is discussed in connection with two main simulation-based applications:travel time reliability analysis, and traffic estimation and prediction systems. The ability to systematicallyidentify similarity and dissimilarity among weather scenarios can facilitate the procedure of selectingcritical scenarios for reliability analysis and studies. It can also support real-time weather-responsivetraffic management (WRTM) by quickly classifying a current weather condition into pre-defined weathercategories and suggesting relevant WRTM strategies that can be tested via real-time traffic simulationbefore deployment. A detailed method for clustering weather time series data based on the K-meansclustering algorithm is presented. The clustering method is demonstrated using weather scenariosconstructed from historical data. The study also performs a cluster analysis based on simulation outputsproduced from the given weather scenarios to compare input- and output-based clustering results.