This paper proposes a methodology to predict the sags level caused by faults that may affect users in a distribution network. The methodology determines the trend of the possible fault location, the fault type and failed phases by analyzing historical data. These trends are modeled using probability densities and are simulated using Monte Carlo techniques (Gibbs algorithm). With the simulation results, a statistical analysis is performed to determine the average sag depth in each user as well as the most vulnerable areas against this type of disturbance. Finally, a sensitivity analysis is achieved with the aim of identifying zones where these disturbances cause a great impact on the average sag depth (disturbing areas) in order to implement the required solutions.