Pavement distress detection is of great significance to pavement maintenance and management. Roadway problems, including cracks, disintegration that, impair the comfort of road users, damage vehicles, increase emissions, etc. Through paintaking manual surveys, such as visual inspections of the pavement taken by the inspection officer, road deteriorations can be found. Researchers and industry have made quiet progress toward creating and implementing automated road surface monitoring systems to cut costs associated with manual inspections. The Road Failures Detection project is developed by using deep learning-based algorithm YOLO (You Only Look Once) for the detection of pavement defects which starts off with data pre-processing, then training the YOLO model, and then testing the designed model and deploying the model using Streamlit app. Streamlit app allows us to host machine learning applications on the web using only Python code. The proposed approach assists in lowering the time needed for road maintenance and offers detection results of up to eight different types of pavement failures that are reassuringly accurate. The dataset used by the system consists of images, annotations, and image sets.