The research explores the scour actions around bridge piers, which are key parameters for structural stability and safety. Two artificial intelligence modelling approaches were adopted for improved prediction accuracy of scour depth. The experimental study was conducted using a newly set-up laboratory experiment. The scour depth model is formulated with different independent geometric and hydraulic parameters such as median size of sediment, aspect ratio, and diameter of pier. The investigation of scour depth AI-based modelling is done using data from past studies. Two machine learning approaches, Support Vector Machine (SVM) and Artificial Neural Network optimized by Particle Swarm Optimization (ANNPSO), were used to analyse the data and produce accurate predictions. The application of PSO to ANN upgrades the model performance and further improves the prediction accuracy. The results confirm that both ANN-PSO and SVM significantly represent the complex interaction of the underlying physics within the data. These models are trained to predict scour depths, enabling them to assess and mitigate scour-related risks under real-life scenarios.