Decision analytical methods help in finding compromise solutions and reducing conflicts between different interest groups. Mathematical tools facilitate the evaluation and comparison of diverse economic, social and environmental impacts. The methods used in this thesis also enable public participation in the planning process, by which they will receive information about the the project's impacts and learn to understand their causes and each others' objectives.

In the first stage of the procedure, the decision and the stakeholders' preferences are structured. The test subjects' opinions on the importance of different attributes were analysed by value tree technique, which aims at pointing out the most important issues and the conflicts in the stakeholders' opinions. In the experiment, the highly structured technique was observed to involve several estimates, such as the value functions, that are difficult to make and thus unreliable. The analysis indicates that in the Lake Päijänne case the compromise solutions between the current and the natural regulation policy are likely to gain more support than the extreme alternatives.

In the second stage of the procedure, new decision alternatives are generated with the method of improving directions. These alternatives are potentially efficient for the whole group of stakeholders. As our application of the method is the first one tested in practise, we are happy to notice that experiment proceeded quite fluently and that the subjects could generate consistent results. However, the subjects' preferences were based on the preliminary estimates of the impacts, and no conclusions on the actual results can be made without reliable estimates.

In the third stage, the new alternatives are evaluated in a decision conference, with the aim of selecting one of them. Also some mitigation measures for adverse impacts of the decision are formulated. Our experiment did not include this stage, in which the results highly depend on the group processes and on the decision makers' willingness to make compromises. We suggest group decision support methods to be used in the final decision making.

The main benefit of the procedure presented, compared to the EIA processes generally used, is that the decision analytical methods are interactive and provide the means to measure and present the participants' opinions. The impacts of different alternatives can be visualized with the computer. The applications of the methods used are user-friendly, and the user does not have to know the mathematical details.