Recent efforts to identify methods of modeling pitting corrosion damage of high-level radioactive-waste containers are described. The need to develop models that can provide information useful to higher level system performance assessment models is emphasized, and examples of i,ow this could be accomplished are described. Work to date has focused upon physically-bas_'t phenomenological stochastic models of pit initiation and growth.These models may provide a way to distill information from mechanistic theories in a way that provides the necessary information to the less detailed performance assessment models. Monte Carlo implementations of the stochastic theory have resulted in simulations that are, at least qualitatively, consistent with a wide variety of experimental data. The effects of environment on pitting corrosion have been included in the model using a set of simple phenomenological equations relating the parameters of the stochastic model to key environmental variables. The results suggest that stochastic models might be useful for extrapolating accelerated test data and for predicting the effects of changes in the environment on pit initiation and growth.Preliminary ideas for integrating pitting models with performance assessment lt 'lt, models are discussed. These ideas include improving the concept of container "failure",• and the use of "rules-of-thumb" to take information from the detailed process models and provide ,,_to the higher level system and subsystem models. Finally, directions for future .... , r-,4-" ',L: '_Yr_