Integrating multiple criteria decision analysis in participatory forest planning: Experience from a case study in northern Sweden

Abstract

Forest planning in a participatory context often involves multiple stakeholders with conflicting interests. A promising approach for handling these complex situations is to integrate participatory planning and multiple criteria decision analysis (MCDA). The objective of this paper is to analyze strengths and weaknesses of such an integrated approach, focusing on how the use of MCDA has influenced the participatory process. The paper outlines a model for a participatory MCDA process with five steps: stakeholder analysis, structuring of the decision problem, generation of alternatives, elicitation of preferences, and ranking of alternatives. This model was applied in a case study of a planning process for the urban forest in Lycksele, Sweden. In interviews with stakeholders, criteria for four different social groups were identified. Stakeholders also identified specific areas important to them and explained what activities the areas were used for and the forest management they wished for there. Existing forest data were combined with information from interviews to create a map in which the urban forest was divided into zones of different management classes. Three alternative strategic forest plans were produced based on the zonal map. The stakeholders stated their preferences individually by the Analytic Hierarchy Process in inquiry forms and a ranking of alternatives and consistency ratios were determined for each stakeholder.Rankings of alternatives were aggregated; first, for each social group using the arithmetic mean, and then an overall aggregated ranking was calculated from the group rankings using the weighted arithmetic mean. The participatory MCDA process in Lycksele is assessed against five social goals: incorporating public values into decisions, improving the substantive quality of decisions, resolving conflict among competing interests, building trust in institutions, and educating and informing the public. The results and assessment of the case study support the integration of participatory planning and MCDA as a viable option for handling complex forest-management situations. Key issues related to the MCDA methodology that need to be explored further were identified: 1) The handling of placespecific criteria, 2) development of alternatives, 3) the aggregation of individual preferences into a common preference, and 4) application and evaluation of the integrated approach in real case studies.3

Decision Support for Participatory Forest Planning Using AHP and TOPSIS

“…As the stakeholders used AHP for preference elicitation a broad understanding of how different forest management strategies might affect their objectives was not required. Furthermore, we chose to combine AHP with TOPSIS because if only AHP had been used for evaluating management plans then the stakeholders would have been presented with a relatively small number of alternatives, which limits the likelihood of finding the optimal plan [9]. Further, in this study three different sets of weights were applied for describing the influence of the interest groups in the evaluation.…”

“…In this study, an upper limit for CR was set at 26%, even though Saaty [6] recommends a maximum CR of 10%. Previous participatory forest planning cases have also allowed a CR above 20% (e.g., [9,19]). We allowed a CR of 26% because certain stakeholders did not have specific knowledge of forestry and this introduced more uncertainty and a larger margin of error.…”

“…In this study, the AHP and TOPSIS methods were used together in a participatory MCDA process consisting of five steps, adapted from the works of Nordström et al [9], Kangas & Kangas [5], Yoon & Hwang [12], and Keeney [15]. The steps are as follows:…”

“…None of the stakeholders made any corrections to the hierarchy. Based on the recommendations of Nordström et al [9] and Kangas [19], thematic maps of the Vilhelmina municipality forest were created to help the representatives discuss the importance of various objectives. The forest was classified into four different management zones based on data from the 2006 management plan and from geographic overview maps of the municipality: (1) a zone with no commercial cutting; (2) a zone with prolonged rotation; (3) an unmanaged zone, and (4) a zone with commercial cutting.…”

“…Saaty and Ozdemir [8] have recommended that the number of objectives and alternatives should not exceed 7˘2, respectively, to make the decision process more manageable. Furthermore, Nordström et al [9] report that, based on their experience, including five or more objectives into the AHP will often become problematic. However, limiting the number of alternatives included in the analysis introduces a risk that the most suitable plan will be excluded from the actual decision process.…”