Fuzzy-sets approach to noise impact assessment

Background, Aims and Scope. Noise impacts are rarely assessed in Life Cycle Assessment (LCA), probably due to lack of data, to the difficulty of setting up an appropriate assessment method including relevant uncertainties and vagueness and to their site-dependent nature. The evaluation, as well as for odour, cultural and aesthetic impacts, seems to be closely related to human judgements and perception based. Although fuzzy-sets have been developed for this purpose since the late '60s and their usefulness has been proven by successful applications, noise impact assessment approaches have been essentially crisp so far. The aim of this paper is to present a method for noise impact assessment based on fuzzy sets with an application to a simple example. Methods. The fuzzy noise impact assessment involves: 1) the quality assessment of the site concerned by the noise impact before the occurrence of noise emissions; quality is expressed by a crisp (i.e. non-fuzzy) function depending on variables (the so-called 'primitives'), which are relevant for the evaluation (e.g. the population density, the type of land use,...); 2) the fuzzy representation of the primitives, e.g. their evaluation by means of linguistic variables (such as 'the population density is high') and by fuzzy numbers; 3) the fuzzy representation of the quality, by fuzzifying the crisp function defined in 1) and 4) the fuzzy representation of the noise impact. In the example, the noise impacts of three processes of coal mining and combustion are assessed. Results and Discussion. The application example proved the operationability of the method. Primitives and noise impact assessment results are represented by fuzzy numbers and intervals that are more informative than crisp numbers for the interpretation of results The quality and impact assessment results obtained seem to be coherent with the nature of the processes involved and of the variables characterizing them. Conclusion and Outlook. Fuzzy intervals and numbers could be more informative and closer to human judgements and perceptions than crisp numbers are, thus improving the pertinence and the interpretation of the results. Despite the increase in sophistication and the fact that the representation of the variables involved in calculations should be developed further (e.g. on the basis of consensus gained in an expert panel), the fuzzy approach seems to be promising for the assessment of noise impacts in LCA.