In recent years, the electricity industry has seen a drive towards the integration of renewable and environmentally friendly generation resources to power grids. These resources have highly variable availabilities. This work proposes a stochastic programming approach to optimize generation expansion planning (GEP) under generator supply capacity uncertainty. To better capture upside opportunities and reduce exposure to downside risks, flexibility is added to the GEP problem through real options on generator addition, which are to be exercised after uncertainty realizations. In addition, with the end goal of providing decision makers with easy-to-use guidelines, a conditional-go decision rule, akin to an if-then-else statement in programming, is proposed whereby the decision maker is provided with a threshold of excess total generator capacity from the previous time period, below which a predetermined generator addition plan (the option) is exercised. The proposed methodology and its decision rule are implemented in a real-world study of Midwest U.S. Comparisons are made to quantify the value of flexibility and to showcase the usefulness of the proposed approach.
