In our project, we describe the techniques used to forecast consumer demand in the short-term for a regional electric utility provider, Kansas City Power and Light (KCPL), as well as optimally schedule power plants for the provider using this demand data. Using a double-seasonal Holt-Winters forecasting method (to account for daily and weekly cycles in our data), we were able to use a scaled energy demand dataset to approximate future electricity demand for KCPL in 30-minute increments. With this demand data in addition to KCPL power plant information, we were able to formulate a binary integer-programming model to solve the problem of scheduling power plants. We then modeled both day-ahead and week-ahead scenarios of running the plants with the goal of minimizing costs for the provider while also meeting demand constraints and operational constraints of the plants. Finally, we describe scenarios where we vary certain assumptions of plant operation and analyze the resulting distribution of coal, nuclear, petroleum, and natural gas plants being used to meet demand.