Electric power systems can be disrupted due to a variety of stress circumstances.  These can range from constraints on the physical infrastructure as a result of under-capacity problems to deliberate attacks in times of conflict.  Stress conditions can manifest themselves in a variety of ways and determining both the nature of the stress and the best mathematical representation of that stress in the model is complicated.   This talk details a quantitative comparison of the reliability of an electricity system based on distributed natural-gas fired units to a traditional system based on large centralized plants using a Monte Carlo simulation reliability under stress.  The model systematically varies the availability of generating units and transmission and distribution line components (both electric and natural gas), representing a stress that is system-wide and persistent. By incorporating stress into model and by applying the model to different system topologies, the paper shows the usefulness of such techniques both in developing new models and in comparing systems according to their ability to perform under stress conditions.  The results show that the distributed system can be significantly more reliable under stress.  Hisham Zerriffi is a Ph.D. Candidate in the Department of Engineering and Public Policy at Carnegie Mellon University as well as a graduate researcher in the Carnegie Mellon Electricity Industry Center. He has a B.A. in Physics (with minors in Political Science and Religion) from Oberlin College, Oberlin OH and a Masters of Applied Science in Chemistry from. McGill University, Montreal, Quebec, Canada. Before joining CMU he was a Senior Scientist at the Institute for Energy and Environmental Research where his work focused on the future of nuclear weapons R&D programs and nuclear waste management. His current research focuses on the technical and institutional challenges of developing energy infrastructure in areas of high stress.