Fuzzy-based computational intelligence to support screening decision in environmental impact assessment: A complementary tool for a case-by-case project appraisal

摘要: Abstract Screening is a key stage in environmental impact assessment (EIA), but the most common approach based on policy delineation are inherently arbitrary. On other hand, case-by-case can be complex, slow, and costly. This paper introduces computational intelligence hybrid fuzzy inference system (h-FIS), combining data-driven expert knowledge, order to assess its capability of supporting screening project appraisal. For empirical research, dataset with appraisal variables projects highway was made available by Brazilian protection agency (EPA). Firstly, using this dataset, multivariate analyses were performed find criteria (xi) capable indicating statistically significant differences among projects, previously screened EPA experts into three types (simplified, preliminary, comprehensive) study (EIS). Then, h-FIS built through machine learning, FRBCS·W algorithm, xi as input predictors type EIS output target. The performances alternative approaches compared cross-validation accuracy tests kappa index, significance level 0.05. As result, achieved 92.6% index 0.88, which represented almost perfect agreement between decision provided one experts. In conclusion, fuzzy-based dealing complexity involved decision. Therefore considered promising complementary tool for EIA. further advances, future research should algorithms, such genetic systems, strengthen proposed make it generally applicable subject