In the present study, the flow around a rectangular obstacle near a flat plate is investigated to calculate the average friction coefficient of the plate. Examining the pervious results of researchers shows that with the presence of a rectangular obstacle near a flat plate, the friction coefficient is affected by factors such as the dimensional ratio of the obstacle, the distance between the obstacle and the plate, the distance between the obstacle and the initial edge of the plate, and the speed of free flow. The past investigations have only studied the effect of one or two factors on the friction coefficient. This paper proposes an effective model for the average friction coefficient by using four methods of artificial neural network (ANN), fuzzy logic (FL), adaptive neuro-fuzzy inference system (ANFIS), and multiple linear regression (MLR). Also, the proposed model is optimized using the grey wolf optimization (GWO) algorithm. The results of the research show that the proposed FL model can predict the friction coefficient with appropriate accuracy with a correlation coefficient greater than 0.99 and has an acceptable agreement with an error of less than 1% compared to the experimental data. The optimal results obtained by the GWO method show a reduction in the friction coefficient of about 3, 35, and 65%, compared to the lowest, average, and highest values of the laboratory results, respectively.