Abstract. In recent decades, many of the countries around the world as well as the south-western Morocco (Guelmim region, Assaka watershed), was subject to flood-storm causing huge human and material damages. The current study focuses on the Prediction of flash flood susceptibility using Fuzzy Analytical Hierarchy Process (FAHP) algorithms and Geographic Information System (GIS) technical. Flash floods areas were identified based on seven flash flood conditioning factors (Soil Moisture Index (SMI), Drainage Density, Rainfall, LULC, Altitude, Slope and Soil). Using AHP the weight derived for the factors were SMI 37% Rainfall 24.30%, Drainage Density 15.57%, LULC 9.98% Altitude 6.39% Slope of the river basin 4.06% and Soil type 2.70%. Then, applying a fuzzy inference system to create flash flood vulnerability maps. The resulting maps were classified into three categories: low, moderate and high flash flood susceptibility; indicated that the areas at the outlet of the watershed and which are close of the main affluent wadis (Seyyad and Oum Al-Achar) were very susceptible to flooding. This study will be helping these zones to be prioritized for the conservation and managing of flash floods.