<p>Accurate estimation of the suspended sediment loads in the Wadi Ouahrane river basin is important for water resources planning and management. Suspended sediments are determining factor of the service life of various works, like dams, bridge, and capacity of reservoir. Therefore, the present study proposed, integrated Sediment Rating Curves (SRC) and machine learning methodological framework based on hydro-meteorological variables for assessment and simulation of the suspended sediment yield in the Wadi Ouahrane basin covering area of 270 Km² in northwest Algeria. Four machine learning models namely Multi-layer Perceptron (MLP), Local Weighted Linear Regression (LWLR), Random Subspace (RSS) and Random Forest (RF) were developed and validated for monthly sediment yield prediction. The cross validation and evaluation of developed models were performed using various statistical measures and graphical representations. The results revealed that during validation, MLP outperformed other models as recorded the highest Pearson correlation coefficient (R = 0.858) and strong efficiency indices i.e., Kling Gupta Efficiency (KGE) = 0.842, Nash-Sutcliffe Model Efficiency (NSE) = 0.730, and Index of agreement (d) = 0.922. Followed by RF model with R value of 0.810 and NSE (0.638) were slightly lower than MLP, although it achieved the lowest MAE (11.025) that suggesting acceptable prediction ability. The study demonstrates the new approach by integrating traditional SRC with machine learning models based on hydro-meteorological variables for monthly sediment yield prediction. The proposed framework effectively captures the nonlinear relationships between hydrological variables and sediment yield in semi-arid regions and improves the accuracy of suspended sediment prediction. The findings from present study provide a reliable framework for sustainable sediment management and water resource planning in wadi Ouahrane river basin.</p>