Flooding, moisture, and climate combined with traffic conditions are responsible for the failure of pavement and traffic disruption. Flooding may deteriorate the structural integrity of the road foundation resulting in huge rehabilitation and maintenance costs. Estimating flooding extent and flow characteristics is, therefore, key to assessing the impact of inundation on road links and networks in low-lying areas that are frequently exposed to flooding water. Such an assessment is helpful for decision-makers in pavement design, rescue operations, and infrastructure planning. Flooding from heavy precipitation is the primary weather-related hazard to transportation infrastructure in the state of Louisiana. Quantification of flood inundation area and depth is essential to quantify the exposure of infrastructure such as pavement and traffic flow to flooding in the changing climate. Two-dimensional hydraulic models that use the governing equation of flow to simulate flow characteristics are widely used to simulate flood extent and depth. However, simulating hydraulic models at acceptably satisfactory resolution and for large spatial extension may require substantial effort. More recent methods have used satellite remote sensing to characterize the extent of the flood, but satellite data still have limitations in producing reliable flood depth information. Modeling framework based on hydraulics and machine learning provides efficient large-scale flood simulation applications. This study tested the GIS tool, the Flood Depth Estimator (FwDET), to estimate complex inundation patterns in an urban setting. When driven by the LiDAR Digital Elevation Modeling and flood extent map, the tool can provide a rapid assessment of the spatial distribution of depth in the flood-prone region and state highway of Amite and Comite Rivers, Louisiana.