Investing in Generation Capacity: A Multi-Stage Linear-Decision-Rule Approach

Investment planning in generation facilities is necessarily a multi-stage exercise, especially in the near future, since most of the power to be built is expected to be from renewable units and their investment costs are still subject to high uncertainty. For this reason, in this paper we propose a multi-stage generation investment model where the demand growth and the investment costs of generating units are uncertain. To tackle this problem, we use a linear decision rule (LDR) approach, which allows us to consider several decision stages while achieving appropriate computing times. Using the proposed LDR methodology, we analyze the outcomes of a number of planning strategies, including achieving a 90% renewable generation mix, a non-carbon mix, and a non-nuclear mix. Additionally, we compare the results with those obtained using a two-stage stochastic programming model within a rolling window framework.