The study of sustainability challenges requires the consideration of multiple coupled systems that are often complex and deeply uncertain. As a result, traditional analytical methods offer limited insights with respect to how best address such challenges in the log-term. This talk will present recent methodological work for developing adaptive, dynamic decision trees that directly connect systems vulnerabilities to the design of adaptive policies. The approach combines optimization and machine learning algorithms to estimate dynamic strategies based on statistical evidence inferred from exploratory modeling ensembles. Through optimization, short-term options lying on the Pareto Frontier of cost and reliability regret are identified. Then through successive optimization and the applications of decision-rule classifier algorithms, adaptive strategies are estimated for each of the options lying along the Pareto Frontier. The application of this analytical framework to water policy and climate change mitigation cases shows that although all estimated dynamic strategies significantly reduce vulnerabilities by adapting to unfolding future conditions, these strategies differ greatly in terms of their cost, the factors that trigger new investments and their capacity to absorb stressors, which are important attributes to be considered in the design of robust, adaptive strategies.

Bio:
Edmundo Molina is an assistant research professor at the School of Government and Public Transformation of Tecnologico de Monterrey and is affiliated adjunct policy researcher at the Rand Corporation. His work focuses on the development of new computational methods for studying technological change, and the use of data science methods for decision analysis under conditions of deep uncertainty. He currently leads applied research work on Mexico’s water and energy sectors, developing new simulation models and assisting stakeholders in decision-making processes. He also leads the Decision Making Center of the School of Government and Public Transformation and co-leads applied policy research on climate change in Costa Rica, Argentina and Chile. He teaches courses on systems modeling, advanced simulation techniques, econometrics and microeconomics.