Abstract: To solve the problem about the nonlinear model of the prediction of woven fabric permeability,the total tightness, thickness, and weight per square meter and average float were selected as prediction factors of woven fabric permeability,and the projection pursuit regression (PPR) model for prediction of air permeability of woven fabrics was established.the fitted values of tested samples and the predicted values of trained samples were analyzed by the means and standard deviations of relative error as the indicators and were compared with the results of BP neural network and multiple linear regression model. The results showed that the PPR model fitting and prediction accuracy was better than that of BP neural network and multiple linear regression model. In the case of less trained samples, PPR model still had relatively high prediction accuracy and good generalization ability, and could provide a novel approach to predict woven fabric permeability.