The economics of reliability for electric generation systems

Based on a Ph. D. thesis in the Dept. of Civil EngineeringProviding excess generation capability for reliability purposes costs a utility money. It is also true, that providing higher reliability adds value to electric service. After some point, however, the additional benefits do not warrant the additional cost. This work deals with the questions of how reliable should generation capability be for meeting system loads, what models should be used to measure this reliability and what bases should be used for answering the above two questions. We critique the measures that have been used and suggest that an energy shortage related measure in probably the best one to use. In addition, we specify what load and what benefit measurement models should be used. The resulting procedures are then used to develop techniques for 1) creating long term expansions of electric utility systems at various reliability levels and for 2) analyzing the costs and benefits of plant additions to arbitrary system expansions. From representative data, we conclude that present generation reliability levels for operation are probably too high. We discuss the magnitude of possible savings and we find that although only a few % of total system costs, they may have substantial profit impact