This research presents an integrated approach for flood risk assessment based on a hydraulic simulation, a fuzzy risk assessment, a statistical analysis, and a Geographic Information System (GIS). Specifically, Computer simulation is conducted for studying different river flow scenarios and obtaining the relationship between flow rate and water surface level, which is used to determine the flow rate under specified risk levels. Statistical analysis is carried out to establish a set of risk criteria under different return periods by considering important factors that affect the peak flow rate based on historical data. A Fuzzy-Set risk model is then developed to assess flood risks using hydraulic simulation and statistical analysis outputs. GIS and digital elevation model (DEM) are used throughout the study for visualizing the study area, constructing conceptual hydraulic model, and spatially presenting different level of flood risks. The developed model is applied to a section of the Red River in the Southern Manitoba, Canada. Reasonable results have been obtained, which are useful to support the related flood management decisions. It indicates that the developed integrated risk assessment method possesses new functions of systematically assessing flood risks with various system uncertainties quantified.