Water quality management with imprecise information

The application of fuzzy set and possibility theories for the representation of imprecise information in water quality management problems is investigated. Imprecision is associated with fuzziness which is non-statistical in nature and refers to the absence of sharp boundaries in the information. Imprecise parameters in water quality decision-making can be represented by possibility distributions defining maximum achievable probabilities. The corresponding possibilistic programming problem is viewed as an alternative to the stochastic one where the parameters are modeled as fuzzy variables instead of random variables. The possibilistic problem can be solved through a succession of classical linear programs. The resulting possibility distributions of the objective value provide a possibilistic assessment of the risk based on possibility levels.