Optimizing natural gas flows with linear programming and scenarios

Recently, we developed a linear programming model to aid an electric utility company in evaluating several long-term operating decisions. Variations of this model provide insights into daily operations and valuation of supply contracts. This paper outlines the important characteristics of this class of network models including the usage of a gas storage facility, the use of market forecasts of gas prices, and the accommodation of gas contract structures. In addition, we develop a scenario optimization procedure in which demand forecasts provide scenarios. We also describe ways in which these models were automated to drive a decision support system.