Big Data Analysis

This research project addresses the complexity and inherent challenges in calibrating microscopic vehicular traffic simulators, such as insufficient accurate data, model complexity, and the subjectivity of traditional methods. The main objective is to innovate in this field by applying Big Data and Artificial Intelligence (AI) techniques to create a superior calibration approach. The proposed methodology focuses on collecting and processing large volumes of traffic data to train and validate a machine learning algorithm specifically designed to automatically adjust simulator parameters. The plan includes integrating this algorithm with existing simulation tools, such as SUMO, and rigorously comparing it against conventional methods to empirically demonstrate improvements in efficiency, accuracy, and objectivity. Expected outcomes include the development of a robust algorithm, the publication of indexed scientific articles (Scopus), and providing a more accurate tool for transport planners and managers, ultimately contributing to transport optimization and more sustainable cities.<br/><br/><b>Goal</b>: <br/>To develop a new approach for the calibration of microscopic vehicular traffic simulators that is more efficient, precise, and objective than traditional methods, utilizing Big Data processing and artificial intelligence algorithms.<br/><br/><b>Research lines</b>: <br/>Artificial intelligence and data mining