Traffic noise, particularly generated by the tyre-road contact has a substantial influence on citizen well-being. As the EU enforces stricter emission standards, compliance becomes a challenging task. The current homologation process assesses Tyre-Road Noise (TRN) under specific conditions, limiting its relevance to real-time traffic scenarios. This project addresses the inadequacy of existing knowledge for comprehensive mitigation by exploring the potential for alleviating Tyre-Road Noise Emissions (TRNE). The approach involves establishing a comprehensive data foundation for TRNE and its influencing factors in actual road traffic conditions. A statistical and machine learning based prediction model is developed to extrapolate emissions on diverse roads from specific measurements on a single road, enabling the derivation of noise reduction measures. A fleet of vehicles gathers data on TRNE through straightforward measurement techniques, with advanced methods estimating parameters validated against ground-truth data. The noise prediction model, based on statistical and machine learning and considering road influences, aims to provide manufacturers and road authorities with additional noise reduction strategies. The refined prediction of road influences enhances measurement comparability across various homologation routes, leading to a more precise balance in conflicting goals. Ultimately, the predictability of the impact of tyre and road measures on public roads and citizens' traffic noise exposure is significantly improved. Our work package (WP 6: Preparation and development of (AI) methods) deals with the development and implementation of methods for large-scale data analysis. The main objective is to provide a comprehensive overview of the existing estimation methods for tyre-road noise prediction and to improve them by integrating statistical and machine learning techniques. In addition, WP 6 will work closely with the other data collection WPs to provide more accurate and comprehensive estimates.