Management of agricultural non-point source pollution: A fuzzy optimization approach

Control of agricultural non-point source pollution usually requires some adjustments of farm management decisions such as the cropping selections. A more realistic analysis must include an evaluation of the impact of pollution control on the entire farm enterprise. A farm is a complex system of interacting activities that compete for labour, land and capital resources, and it can be difficult to isolate activities such as fertilizer and pesticide management from the rest of the farm operations. The imposition of environmental controls can affect farming components not directly related to the source of pollution. Because of the complexities of the farm systems, one of the few feasible means of making an assessment is to construct mathematical models of farms. The models usually used in practice are generally income-maximizing linear programming models of farms and they are manipulated by imposing selective constraints on farm income. These models need, however, a precise definition of constraint fulfilment and objective function, as well as precise data of coefficients. This is not always possible in practice. In the paper we show an application of fuzzy linear programming modelling to overcome these difficulties. Lack of practical knowledge on, first, the constraints fulfilment and objective function maximization and, second, on coefficients and right hand sides in balance constraints is accounted for in terms of fuzzy sets. General conclusions and suggestions about fuzzy optimization approach on management of agricultural non-point source pollution and its practical application are finally drawn.