This paper shows an example of a stochastic approach to study the impact of distributed generation (DG) on the network constraints (congestions) in power systems. We assume the DG units to be customer-owned, so that they can be connected to or disconnected from the power system by their owners at random. Therefore, the DG units generate power in a stochastic way. The load in the system shows a random behavior too, and the probability distribution of the aggregated generated power and load demand of the distribution system are calculated using Monte Carlo Simulation (MCS).The network constraints (congestions) are evaluated based on the probability distributions of the power flows that result from the simulations. The method that is applied in this paper, shows that looking at the network constraints with a stochastic approach gives a more complete picture of the network than applying a deterministic method, especially when non-dispatchable DG units play a dominant role.