Search

In this paper, we present an approach employing multiobjective optimization to support decision making in forest management planning under risk. The primary objectives are biodiversity and timber cash flow, evaluated from two perspectives: the expected value and the value-at-risk (VaR). In addition, the risk level for both the timber cash flow and biodiversity values are included as objectives. With our approach, we highlight the trade-off between the expected value and the VaR, as well as between the VaRs of the two objectives of interest. We employ an interactive method in which a decision maker iteratively provides preference information to find the most preferred management plan and learns about the interdependencies of the objectives at the same time. The method is illustrated with a case study in which biodiversity is assessed through an index calculated from the characteristics of the forest. Uncertainty is included both through modifying the input data according to the accuracy of current inventory methods and through growth model errors. This uncertainty is described using a set of 25 scenarios. Involving multiple components of risk is a highly relevant approach in multiobjective forestry; however, estimation of the uncertainty of biodiversity needs further attention. [ABSTRACT FROM AUTHOR]

Developing a forest management plan in a multicriteria perspective is traditionally accomplished utilizing simulation and optimization tools as a means to predict and optimize a variety of criteria under consideration. Goal programming is a useful tool to balance conflicting aspects of the competing criteria. When information regarding an aspect of uncertainty is available, stochastic programming should be utilized to efficiently integrate this additional information. Research has been conducted into determining the accuracy of forest inventory methods; however, the measurement error is typically ignored when generating forest management plans. Through integrating the uncertainty in a systematic fashion, the forest management plan can be improved by describing the potential uncertainty in the plan and by managing the influences of this uncertainty. This paper develops three stochastic goal programming formulations and highlights the usefulness of the approach on a small forest holding. [ABSTRACT FROM AUTHOR]