Browsing by Author "Jaakkola, Laura"

The aim in forest planning is to select treatment schedules for each stand of the forest so that the forest owner’s utility is maximised. Different optimisation methods are used in forest planning to find the optimal treatment schedules in the context of the forest planning problem. The demands for forest planning are increasing. The reasons for this include developments of data analysis methods, increased attention to the importance of conserving biodiversity and the role of forests in mitigating climate change. When designing decision support tools for forest management in this changing operating environment, it is important to understand and strive to meet the needs of the users of these planning tools. The MELA program is a forestry analysis and planning software maintained by the Natural Resources Institute Finland, and a new version of the software is under development. One aim of this thesis is to undertake requirement analysis for the optimisation part of the software focusing especially on the problems of spatial optimisation. Spatial optimisation problems are such that the optimal treatment schedule selection for a single stand (or other calculation unit) of the forest is dependent on the treatment schedule selections of adjacent or nearby stands. Currently the optimization calculation for the MELA program is done with linear programming software JLP. However, forest planning problems with spatial goals and constraints are often difficult or impossible to formulate as linear programming problems. Another aim of this thesis, in addition to researching user needs, is to research different ways to solve the kind of spatial optimization problems that are identified as important for potential users. Forest planning specialists and researchers were interviewed to answer the research questions of this theses. The interviews were conducted as semi-structured thematic interviews on the topic of optimisation calculations in forest planning. Based on the interviews, a use case of spatial optimization was formed, in which the aim is to aggregate harvests in time and location so that the stands included in the same harvest site are close enough to each other, have the same accessibility classification and are accessible from the same road. Research literature addressing these types of problems is reviewed and compared with the example use case. The most effective solution model for the use case cannot be unequivocally chosen, but the summary of the thesis presents a few solution models and identifies some advantages and disadvantages of these solution models compared to each other.