STOCHASTIC ACTIVITY NETWORKS - A 4-PARAMETER APPROACH FOR COMPLETION TIMES

A new approach to find an accurate and realistic estimate of the compl etion time of PERT type network projects is presented. The project com pletion times calculated by the new method are compared with those of the standard textbook PERT method against bench mark values supplied b y a simulation program. Beta distribution with four parameters is assu med for each activity. The four moments of the paths are calculated an d Schmeiser-Deutsch distribution is fitted to each path. The project c ompletion time is computed as the maximum of n stochastic variables re presented by each path length. The project completion times found by t his new technique are more accurate and much closer to the simulated v alues than those of PERT values. The significance is more prominent in cases where there are multiple paths with near equal lengths. The res ults may have practical applications in project management and overcom e the perennial problem of 'late projects'.