%0 Articles
%T Integration of preference elicitation and the development of alternative forest plans: focusing on the requirements of the decision maker
%A Eyvindson, Kyle
%D 2012
%J Dissertationes Forestales
%V 2012
%N 147
%R doi:10.14214/df.147
%U http://dissertationesforestales.fi/article/1930
%X Modern forest management frequently revolves around the concepts of developing strategic, tactical and operational level plans. These plans are developed through the use of simulation and optimization software, based on scientific models and mathematical programming. The optimal management schedule depends upon the decision maker(s) (DM) preferences. When developing forest plans for the DM(s) the method of acquiring preference information should be as value free as possible. To facilitate a DM-orientated approach, a selection of alternatives based on the acquired preferences should be made available to the DM so that a true choice can be made. The development of the forest plans should represent the desires and wishes of the forest owner at the time the plan is created. In order to balance the costs with the quality of the service provided, tools are required which develop client specific forest plans.

The first objective of this thesis is to analyse different preference elicitation methods and study the impacts of information content on the selection of a plan. In papers I and II, plans were selected using an a posteriori method of preference elicitation. For paper III, preference elicitation was done in an interactive fashion, to develop an acceptable forest plan using both a priori and a posteriori preference elicitation methods.

The second objective is to develop techniques for incorporating preference information into optimization methods. In paper IV, a series of goal programming models were used to incorporate the preference information from several DMs to generate a number of potentially desirable forest plans. Paper V develops a goal programming formulation which separates the treatment of different goals into two partitions; one strives to maintain the difference from the target for the goals in balance, the other strives to obtain the most efficient aggregate solution.