This paper starts from the dynamic scheduling from the point of view, according to the three specific conditions, to judge the working state of R GV and C NC, through the enumeration method List various situations, schedule the R GV, and find the quantity of ingredients and work efficiency. For the material processing of one process, firstly, the optimal RGV feeding order should be determined and the shortest time is found. Since the CNC on the right side of RGV is less than that of the CNC on the left side of RGV, when the RGV reaches a certain position, feed the CNC on the right first, and then feed the CNC on the left side and go to the next position after completion. Then the first process is solved through the 0-1 variable set the working state of RGV and CNC, establish a CNC status matrix, judge C NC working state, and solve the sum of CNC state, if not working, require loading operation, thus determine the time of RGV consumption and CNC consumption, and then substitute the data into the written code to obtain the finished product quantity and CNC working efficiency. For the material processing of the two processes, assuming that the CNC does not need to be replaced with tools within 8 hours, the distribution of the CNC is first determined. Because the two processes are different, and the number of CNC in each process should be balanced, so the C N C installation ratio should not be too large, and the same position priority rule should be used to determine the installation position. After determining the order of up and down of the two processes, the moving logic and up and down logic are determined respectively according to the relevant formula and code, and finally the system work efficiency of all cases is obtained by cycling. In the case of C NC fault, the fault occurrence node and maintenance status duration of CNC are simulated by random number, and the event-driven policy is adopted, Update the status of the CNC and RGV in real time when a fault occurs, and recalculate the target function value, so as to dynamically schedule the RGV according to the fault interference. Considering the above three cases, the RGV dynamic scheduling model and the corresponding solution algorithm suitable for different working parameters, complex processes, and possible fault interference can be obtained.