Análise preliminar dos requisitos técnicos dos titulares de autorizações para reparo e manutenção de instrumentos de medição regulados: Aplicação de sistemas de inferência fuzzy

Abstract

Accurate measurement is essential in sectors such as industry, healthcare, and the environment. In Brazil, the National Institute of Metrology, Quality and Technology (INMETRO) regulates and inspects measuring instruments, requiring authorization holders for repair and maintenance to meet stringent technical requirements. This study applied fuzzy inference systems for the preliminary analysis of these requirements, developing a model capable of efficiently and accurately identifying and evaluating compliance. The method involved surveying technical requirements, defining linguistic variables, developing and testing the fuzzy system, and analyzing results. The system was implemented and tested in a controlled environment with different datasets, demonstrating high accuracy in identifying compliance in over 90% of cases. Additionally, it provided valuable insights into authorization holders' strengths and weaknesses, highlighting areas that may need improvement for continued compliance. The application of fuzzy inference systems proved to be a promising approach to improving compliance assessment processes, ensuring the quality and accuracy of measurements in Brazil, and renewing the perception of analyses in ordinances already implemented by INMETRO. This approach can also significantly contribute to decision-making by regulatory agencies, such as IPEM/AM, providing an additional tool to ensure compliance and reliability of measuring instruments used nationwide.