A fuzzy approach to natural resource management from a regional perspective

This paper describes an approach to regional natural resource management that explicitly incorporates both the uncertainty and the multi-objective nature of such large-scale systems. A fuzzy linear programming algorithm is used to analyze sets of land management alternatives designed for individual forests within a region. Fuzziness is introduced in recognition that regional goal aspiration levels, interactions among resources, and cost estimates are imprecise and uncertain. An illustrative example, based on the 1989 RPA National Assessment Resource Interactions Model of the USDA Forest Service, is applied to the California Region and is used to test various fuzzy modelling approaches. Results show that: (a) regional optimization produces significant cost savings when compared with the solution obtained by aggregating individually optimized forest plans for the region, and (b) fuzzy-based regional solutions may offer decision makers a wider range of preferable solutions than those based solely on linear programming.