Resin transfer molding (RTM) is a closed-mold procedure in the composite manufacturing process. The resin enters the mold from the inlet port, and the gradient pressure of the inlet and outlet makes the flow move into the mold. A common disturbance that happened during the process is space between the fiber and the mold wall, where resin can flow easily due to low flow resistance in these channels, giving rise to the race-tracking phenomenon. The burden of experimental tests required to train in an efficient manner such predictive models is so high that favours its substitution with synthetically-generated simulation datasets. The first step is preparing the input training data from OpenFoam simulation and validate with experimental tests to develop the ML method for detection and prediction. The first part of this work shows the ability of simulation in comparison to the experimental tests. The machine learning method is one of the best techniques that detect automatically flow disturbances caused by the presence of dissimilar permeability regions in the liquid molding of composites. Finally, the purpose of this work is the development of supervised machine learning (ML)model to detect flow disturbances caused by the presence of a race-tracking or dissimilar material region in liquid molding manufacturing of composites. The machine learning model was designed to predict the position, size, and relative permeability of these regions through the use of only the signals corresponding to an array of pressure sensors evenly distributed on the mold surface.