Global climate change can create uncertainty and affect local weather patterns, which in turn can increase the frequency and severity of extreme weather events such as flooding. Flooding is considered one of the major causes of bridge collapse and it is important to ensure the safety and functionality of these structures against such hazards. This paper presents a methodology for an artificial neural network that can estimate peak flow discharge by analyzing a projected precipitation database and considering various parameters related to climate change uncertainties. The results of the machine learning model can then be used in a hydraulic model to identify flood-prone areas and estimate hydraulic parameters such as flow velocity and water column height using a geographic information system. In addition, the methodology can evaluate the most significant and recurring impact of flooding on bridges, namely the scouring process. The proposed methodology was tested in a masonry arch bridge case study in Portugal. The results show the effectiveness of the methodology in predicting flood risk and assessing the potential impact on bridge safety and functionality.