Monte Carlo Simulation is a standard technique for project risk analysis. However, some assumptions of this technique are not reasonable in real world. When a project falls behind the schedule, managers take actions to improve the performance. Nevertheless, most of the simulation models as Monte Carlo omit these reactions in their analysis, which results in unreasonable wide distributions or even very optimistic ones. More ever, it is not adequate to simply formulate direct results of these control actions because they have secondary and territory side effects. System dynamics models of projects can capture these reactions and their direct and indirect impacts through feedback loops and casual relationships. In this paper, a system dynamics model has been developed to perform Monte Carlo simulation. Vensim professional software is used to develop the SD model and formulating project operational structures. Comparison of results of system dynamics model and original Monte Carlo analysis is presented by an example. Simulation results shows that SD and Monte Carlo both achieve an almost same average project completion time, however the shape of distribution for project completion time is different in two methods. In fact, modeling of dynamic complexities of projects in SD results in a more reasonable distribution for project completion time.