Abstarct: To increase and Join production in coal mining panels, predicting and determine the appropriate speed of these devices can greatly help the project implementation time and economics of designs. For this purpose,1260 types of coal, cutting by the coal mining machine were carried out in the E3 extraction panel of the Tabas mechanized mine. First, after collection and recording the shearer speed of each cut, the information about gas flow was performed at three points along the total length of the panel. These three points include emitted methane gases as a percentage on sensor 88 and the tailgate input sensor¬(TG)and the sensor embedded on the Armored face conveor (AFC).Shearer speed prediction was performed with three models of linear and nonlinear multivariate regression¬(exponential and logarithmic).The results show that the multivariate linear regression model with coefficient of determination R2=0.90 has a more accurate prediction than the other two methods using the linear multivariate regression model the amount of shearer speed can be predicted with good accuracy. For this purpose, the genetic algorithm in MATLAB software has been used to optimize the speed of the shearer device. determine the appropriate speed results show that cross diagrams baxsed ¬on¬ coefficient of¬ determination(R2), according to Equations (logarithmic, exponential ¬and linear), linear ¬type ¬has a ¬higher coefficient of determination than other equations. Therefore, the best model is selected for determine the appropriate speed. Using the linear equation in the genetic algorithm, the extraction speed of the shearer machine was estimated to be 4.79 m/min.