Sequential Decision-making in Interdependent Sectors with Multiobjective Inoperability Decision Trees: Application to Biofuel Subsidy Analysis

Decision-making involving large-scale systems often involves considerations for temporal changes, interdependencies in organizational structures, multiple competing objectives, and risk and uncertainty, among others. In this paper we develop a risk-based methodology, the Multiobjective Inoperability Decision Tree (MOIDT). It integrates several dimensions of decision-making processes associated with interconnected systems in terms of: (i) evaluation of sequential policies; (ii) analysis of interdependencies; (iii) treatment of multiple objectives and their tradeoffs; and (iv) characterization of uncertainties. To demonstrate the integration of these four components, we present a case study to analyze the impact of government policies towards mass-scale biofuel production. Using a multi-period decision framework, the analysis utilizes economic input-output data to model the probabilistic demand adjustments for sectors that will likely be affected by biofuel policies.