Integrating the fuzzy analytic hierarchy process with dynamic programming approach for determining the optimal forest management decisions

The paper considers a forest management problem, which consists of decisions on investment, silvicultural and harvesting activities for an existing forestland over a long time horizon, while guaranteeing sustainability, maximizing the expected profit, referring to ecological objectives and the public's acceptance of decisions. In order to solve this problem, several decision support systems and models were developed. Here, we present a fuzzy, dynamic and multi-objective model for optimal forest management, which determines the sequence of decisions that jointly maximizes economic, ecological and social objectives, respects prescribed constraints and imprecision, and takes the forest system from its existing state to the goal state. The idea of fuzzy analytic hierarchy process is introduced within a discrete dynamic programming process to evaluate the conflicting objectives. The cumulative impacts of objectives are formulated on the basis of attributes as a sum over all products of their membership functions and their relative importance (weights). Finally, the problem is presented in a form of a network, and the optimal policy over all time periods is determined using the iteration method for the solution of sequential decision processes of Bellman's type. A case study involving the forest Panovec, Slovenia is used to demonstrate the application of the presented model.