Malaysia is a tropical country and it is subjected to flooding in both the urban and rural areas. Flood
modelling can help to reduce the impacts of flood hazard by taking extra precautions. HEC-RAS model was used to predict the flood levels at selected reach of the Langat River with a total length of 34.4 km. The Langat River is located in the state of Selangor, Malaysia and it is subjected to regular flooding. The selected reach of the Langat River has insufficient data and a methodology was proposed to overcome this particular problem. Since complete floodplain data for the area are not available, the modelling therefore assumed vertical walls at the left and right banks of the Langat River and all the predicted flood levels above the banks were based on this assumption. The HECRAS model was calibrated and the values of Manning’s coefficients of roughness for the Langat River were found to range from 0.04 to 0.10. The discharge values were calculated for 5, 10, 25, 50, and 100 year return periods and the maximum predicted flood depth ranged from 2.1m to 7.8m. Meanwhile, the model output was verified using the historical record and the error between the recorded and predicted water levels was found to range from 3% to 15%.