APPLICATION OF INDUSTRY 4.0 TECHNOLOGIES FOR THE REDUCTION OF DESTRUCTIVE TESTING IN MOTORCYCLE FUEL TANK PAINT QUALITY CONTROL: A PROPOSITIONAL CASE STUDY

Abstract

This article aims to analyze the implementation of Industry 4.0 technologies as a smart solution to reduce or eliminate destructive testing in the quality control of motorcycle fuel tank painting. The research is classified as applied, exploratory, and descriptive, employing a case study strategy with a mixed-methods approach (qualitative and quantitative) at a manufacturer in the Manaus Industrial Pole. The results reveal that the conventional inspection protocol, while technically reliable, creates an operational bottleneck of 350 hours per month and an estimated annual cost of R$ 426,268.80 due to mandatory rework resulting from the invasive nature of adhesion tests. A conceptual model is proposed that integrates Computer Vision, Artificial Intelligence (Deep Learning), and non-invasive sensors for the continuous monitoring of process variables. The analysis demonstrates the technical feasibility of reducing inspection time from 20 minutes to a range of 5 to 15 seconds per part, enabling the inspection of up to 100% of production and the mitigation of physical and chemical occupational risks. It is concluded that the model provides a robust response to the research problem, highlighting that the transition to predictive monitoring optimizes operational efficiency and eliminates financial waste, although it requires future experimental validation in a real industrial environment.