A fuzzy approach for modelling knowledge in environmental systems evaluation

Uncertainty often appears in modelling environmental systems, in particular uncertainty concerning the data and the relations between the system components. A fuzzy set theory approach is used in this study to help solve this problem of subjectivity in the determining the system parameters involved. Fuzzy rule-based models are usable both in “soft” disciplines, such as ecology or biology, and in “hard” disciplines like physics or engineering since they can combine physical laws, expert knowledge, and measurement data. This paper describes a fuzzy rule-based methodology for environmental evaluation. In evaluating environmental systems, experts use linguistic tools and specific knowledge which are their interpretations of the organisation of the systems. Fuzzy rules could be derived from both experts' reasoning and linguistic expressions and from the relationships between the system variables. The problem of uncertainty is faced through the use of linguistic variables which are real pieces of knowledge.