Sistemas de inferência fuzzy para apoio na tomada de decisão em licenciamento ambiental de empreendimentos rodoviários
Carregando...
Data
2019-06-27
Autores
Cavalca, Lucirene Vitória Góes França
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A Avaliação do Impacto Ambiental tornou-se obrigatória para atividades consideradas efetivas ou potencialmente poluidoras. A experiência e o conhecimento de agentes técnicos para direcionar estudos e medidas para minimizar e / ou mitigar os impactos causados ao meio ambiente é fundamental em um processo que envolve muitas variáveis, como o licenciamento ambiental. O processo de licenciamento ambiental recebe muitas críticas devido a burocracia e morosidade. Estudos e pesquisas para melhorar as etapas do processo vêm sendo apresentados ao longo dos anos, mas ainda há muito a ser investigado e compartilhado, sendo pouco o que é encontrado na literatura que enfoque o desenvolvimento de um procedimento lógico estruturado para apoiar a definição do escopo do estudo mais apropriado caso a caso. Aqui, propõe-se uma metodologia baseada em modelagem fuzzy para apoiar a fase de triagem com intuito de minimizar o tempo e a burocracia existente dentro do processo. Contou-se com a colaboração de especialistas do setor de licenciamento para a construção do modelo e para sua avaliação. Foi utilizado um conjunto de dados com trinta e dois projetos rodoviários, disponibilizados pelo setor de protocolos da Companhia Ambiental do Estado de São Paulo. O sistema foi aplicado em processos de licenciamento de empreendimentos rodoviários (construção, ampliação e manutenção de rodovias, além de dispositivos, como trevos e alças de acesso) e os resultados obtidos pelo modelo foram comparados com análise prévia realizada por especialistas do setor. Os resultados mostram que embora o formulário prévio contenha falhas que podem ser aprimoradas, a sistematização é imprescindível para a otimização do processo. A modelagem fuzzy pode ser considerada uma abordagem promissora, e pode ser implementada no intuito de promover maior rapidez e padronização no processo de licenciamento.
Environmental impact assessment has become mandatory for activities either considered to be effectively or potentially polluting. The experience and knowledge of technical agents to direct studies and measures to minimize and / or mitigate the impacts caused to the environment is fundamental in a process that involves many variables, such as environmental licensing. The environmental licensing process receives many criticisms due to bureaucracy and morosity. Studies and researches to improve stages of the process have been presented over the years, but there is still much to be investigated and shared, with few studies reported in the literature focusing on developing a structured logical procedure to support the definition of type of study more appropriated case-by-case. Here, a methodology based on fuzzy modeling is proposed to support the screening phase in order to minimize the time and bureaucracy within the process. It was counted with collaboration of specialists in the licensing sector for the construction of the model and for its evaluation. A data set was used with 32 road projects, made available by the protocols sector of the Environmental Company of the State of São Paulo. The system was applied in the licensing processes of road projects (construction, expansion and maintenance of highways, as well as devices such as clovers and access handles) and the results obtained by the model were compared with previous analysis performed by specialists in the sector. The results show that although the previous form contains flaws that can be improved, systematization is indispensable for the optimization of the process. Fuzzy modeling can be considered a promising approach, and can be implemented to promote faster and more standardized licensing.
Environmental impact assessment has become mandatory for activities either considered to be effectively or potentially polluting. The experience and knowledge of technical agents to direct studies and measures to minimize and / or mitigate the impacts caused to the environment is fundamental in a process that involves many variables, such as environmental licensing. The environmental licensing process receives many criticisms due to bureaucracy and morosity. Studies and researches to improve stages of the process have been presented over the years, but there is still much to be investigated and shared, with few studies reported in the literature focusing on developing a structured logical procedure to support the definition of type of study more appropriated case-by-case. Here, a methodology based on fuzzy modeling is proposed to support the screening phase in order to minimize the time and bureaucracy within the process. It was counted with collaboration of specialists in the licensing sector for the construction of the model and for its evaluation. A data set was used with 32 road projects, made available by the protocols sector of the Environmental Company of the State of São Paulo. The system was applied in the licensing processes of road projects (construction, expansion and maintenance of highways, as well as devices such as clovers and access handles) and the results obtained by the model were compared with previous analysis performed by specialists in the sector. The results show that although the previous form contains flaws that can be improved, systematization is indispensable for the optimization of the process. Fuzzy modeling can be considered a promising approach, and can be implemented to promote faster and more standardized licensing.
Descrição
Palavras-chave
Triagem, Lógica, Modelagem matemática, Análise de processos, EIA