The reliability of the reaction force developed in the lubricated revolute joint for a slider-crank system considering realistic joint characteristic,namely,joint with clearance and lubrication in the revolute pair,and the randomness of the system parameters is presented. The continuous approach and the hydrodynamic theory are used to evaluate the contact force for the case of joints modeled as a contact pair with dry contact and the force generated by lubrication action respectively. The system dynamic model was set up based on Newton-Euler method,The prediction accurary of Support Vector Machine Regression is difficult to reach the target accurary because the selection of parameters isn’t accurate. In order to overcome this limitation,the parameters are pretreated through Genetic Algorithm to get the optimum parameters values. Then approximate function relation between random parameters and performance function is derived and the reliability of the reaction force developed in the lubricated revolute joint of the crank-slider mechanism is discussed by Monte-Carlo method. Finally,take an example to verify the feasibility and effectiveness of the proposed method.