Sustainable Digital Transformation for Industry 5.0 Using Interpretable Predictive Diagnosis and Cyber-Physical Architectures Based on Sensing and Data Intelligence

• Grant number: 25/09204-5
• Support Opportunities: Research Grants - Initial Project
• Start date: March 01, 2026
• End date: February 28, 2031
• Field of knowledge: Engineering - Electrical Engineering
• Principal Investigator: Pedro de Oliveira Conceição Junior
• Grantee: Pedro de Oliveira Conceição Junior
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• City of the host institution: São Carlos
• Associated researchers: Alessandro Roger Rodrigues ; Alessandro Simeone ; Dennis Brandão ; Eduardo Paciência Godoy ; Paulo Roberto de Aguiar

Abstract

The increasing digitalization of industry has expanded the use of smart sensors and artificial intelligence algorithms but still faces challenges related to structural reliability, energy efficiency, and the interpretability of data-driven models. This project proposes the development of an integrated and sustainable platform for predictive diagnostics of structures and optimization of industrial processes, aligned with the principles of Industry 5.0. The approach is based on the fusion of structural signals, multimodal sensing, embedded preprocessing (edge computing), and interpretable artificial intelligence models aimed at anticipating failures and degradation in manufacturing systems and materials, such as tool wear and delamination in composite machining. The solution will be validated through experimentation and functional tests using data from machining tools and composite materials under different failure regimes. The methodology will involve structural sensors (acoustic emission, vibration, and electric current) for continuous data acquisition, feature extraction, and predictive modeling using explainable algorithms. The modules will be integrated into an embedded cyber-physical platform with connectivity and validation in a laboratory environment. The proposal also includes interdisciplinary training of undergraduate and graduate students (IC, Master's, and PhD), scientific outreach initiatives, international collaboration with two Italian institutions (UNIBS and POLITO), and social, technological, and educational impact, contributing to the advancement of applied solutions for sustainable industrial digitalization in Brazil. The results will be evaluated based on the percentage reduction of unforeseen failures, the accuracy of the predictive models, the training of human resources and associated scientific output (number of students and publications), and the strengthening of international collaborations. In addition, sustainability impacts will be assessed by reducing rework and extending the service life of industrial components. (AU)