An Optimization Model for Production Costing in Electric Utilities

Estimating the annual production cost is a crucial component in long range capacity planning for electric utilities. Traditional simulation-based methods are quite expensive in terms of computer resources (space and time), while simpler mathematical programming-based models generally assume a linear variable production cost. This paper presents a method that incorporates quadratic production costs for each unit and uses optimization submodels to develop a system production cost versus output curve. The method is tested using data from a large southeastern utility. Production cost estimates within one percent of the estimates from a large simulation model were obtained with a small fraction of the simulation model's computing time. The concluding section discusses how the new method might be incorporated in procedures for capacity planning.