The program of rules for terminating the simulation run when sufficiently accurate or stable results have been achieved is proposed and coded in FORTRAN. An approach to the problem of estimating the precision of simulation results uses time series analysis, basing upon a single long run approach (or method of batch means) with the initial bias removed. When this program is incorporated in a computer simulation model, preferably a stochastic model, statistical analysis of simulation experiments containing autocorrelated time series will be analyzed after the termination of the run. Unless desired accuracy has been obtained, the model is restarted with the latest variables and parameters setting and again analyzed.