An enhanced bootstrap method to detect possible fraudulent behavior in testing facilities

The testing process determines the specified characteristics of an object. Testing is commonly used to verify the quality, safety, reliability, efficiency, and environmental sustainability of products and services. In Brazil, the National Institute of Metrology, Quality and Technology (Inmetro) is responsible for accrediting the facilities that undertake testing and calibration activities. Typical accredited testing services include: water quality, clinical tests, emissions tests, and vehicle safety tests. Undertaking fraudulent behavior when testing processes can result in products that are dangerous to the ordinary citizen and to the environment. Further, fraudulent behavior produces unfair competition between concurrent organizations. This paper presents a method to detect possible fraudulent behavior in accredited testing facilities. The proposed technique combines the bootstrap method and the Dempster-Shafer theory of evidence, to investigate several fraudulent aspects at once. To validate the proposed method, we use a Brazilian real-world dataset with real cases of fraud. The proposed technique successfully detects the accredited testing facilities with fraudulent behavior.