Robust portfolio modeling in environmental decision making

Abstract

The importance of environmental decision making is growing. Private companies and public organizations are facing decisions involving multiple objectives. In particular, focusing solely on financial objectives is no longer enough but taking into account the environmental, social and political objectives is needed.

The methods used to solve these environmental problems have been based on heuristic approaches. However, these methods lack the capability to provide optimal solutions as most of the environmental decisions are portfolio selection problems. Robust Portfolio Modeling (RPM) is a decision analysis method that combines mathematical optimization in portfolio selection to incomplete preference information. This incomplete information is common in environmental decision making which includes multiple stakeholders with conflicting views. However, RPM has not been applied before to real-life environmental cases.

This thesis will first explore the characteristics of environmental decision making, secondly go through different methods used in environmental decision making and finally apply RPM methodology into peatland selection case. The results of RPM are then compared to the results of the heuristic YODA method previously used in the same peatland selection case.

Results indicate that RPM and YODA select highly different type of peatlands. RPM takes better into account the cumulative effects related to portfolio selection than YODA. Therefore, it is argued that RPM might be suitable for environmental decision making.