Application of Fuzzy inference systems for preliminary analysis of the technical requirements of permission holders for the repair and maintenance of regulated measuring instruments

Abstract

Accurate measurement is essential in a variety of sectors, from industry to health and the environment. In Brazil, the National Institute of Metrology, Quality and Technology (INMETRO) is responsible for regulating and inspecting measuring instruments. To ensure compliance, holders of authorisations to repair and maintain these instruments must meet strict technical requirements. In this context, the application of fuzzy inference systems can be an effective approach to preliminarily analysing these technical requirements. he aim of this article is to apply fuzzy inference systems to the preliminary analysis of the technical requirements of holders of authorisations for the repair and maintenance of measuring instruments regulated by INMETRO. The aim is to develop a model that can help identify and assess these requirements efficiently and accurately. The materials and methods adopted involved the following steps: Survey of Technical Requirements: Initially, the technical requirements established by INMETRO for authorisation holders were identified. This included standards, regulations and relevant documentation; Definition of linguistic variables (Fuzzification) and parameters (INMETRO Ordinance): Next, the relevant variables for the analysis were identified, such as technical capacity, the necessary infrastructure and the specific knowledge required; Development of the Fuzzy Inference System (Regrtas): Using the variables identified, a fuzzy inference system was developed. This system was designed to receive information on authorisation holders as input and produce a preliminary assessment of compliance with technical requirements as output; Implementation and computer tests: The system was implemented in a controlled environment and tested with different data sets representing authorisation holders. This made it possible to assess the effectiveness and accuracy of the system in correctly identifying the technical requirements; Analysing the results: The results were analysed in terms of the system's accuracy, sensitivity and specificity in relation to the technical requirements established by INMETRO. The results obtained showed that the fuzzy inference system was able to carry out a preliminary analysis of the technical requirements of authorisation holders with significant accuracy. The system was able to correctly identify the holders' compliance or non-compliance with the technical requirements in more than 90 per cent of the cases tested. In addition, the system (Defuzzyfication) was able to provide valuable insights into the strengths and weaknesses of authorisation holders, highlighting areas that may need improvement to ensure continued compliance with technical requirements. The application of fuzzy inference systems proved to be an effective and promising approach for the preliminary analysis of the technical requirements of authorisation holders for the repair and maintenance of measuring instruments regulated by INMETRO for enforcement by IPEM/AM and in each state of Brazil. This approach can make a significant contribution to improving conformity assessment processes and guaranteeing the quality and accuracy of measurements carried out in Brazil and renewing the perception of the analyses in the ordinances already implemented by INMETRO.