This paper presents a method for studying the occurrence of loss of component scenarios in a power system using stochastic point process modeling. It can manage stationary and non stationary failure and repair rates, common mode failures of any order, components with any number of operating states and diverse repair logistics. The failure and repair sequences in each maintenance zone of the power system are generated using Monte Carlo simulation. A list of possible loss of component scenarios is therefore not required as in other methods. Loss of component scenarios are studied in terms of probability of occurrence, expected frequency and mean duration. By means of two examples, it is shown how the proposed method is a more flexible modeling tool than those based on Markov chains.