An automated identification and traceability system for reverse logistics operation of waste electrical and electronic equipment

• Grant number: 22/09178-6
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: November 01, 2022
• End date: April 30, 2023
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Raul Julian Revelo Tobar
• Grantee: Raul Julian Revelo Tobar
• Company: Thalita Braga
• CNAE: Coleta de resíduos não-perigosos ; Coleta de resíduos perigosos ; Atividades profissionais, científicas e técnicas não especificadas anteriormente
• City: São Carlos
• Pesquisadores principais: Diogo Puppim de Oliveira
• Associated scholarship(s): 23/00466-1 - Integration of a Blockchain-based traceability system, integrated with the processes of reverse logistics of electronic waste., BP.TT; 22/14537-5 - Image processing and development of a deep learning model for classifying and identifying electronic waste in a reverse logistics system., BP.TT; 22/13858-2 - An automated identification and traceability system for reverse logistics operation of waste electrical and electronic equipment, BP.PIPE

Abstract

The growing demand for electrical and electronic devices during the COVID-19 pandemic has aggravated the problem in the generation and destination of waste electrical and electronic equipment (WEEE) compared to other types of waste. In 2019, the world generated 53.5 million tons of electronic waste, and 83% of this waste has an uncertain or undocumented destination. Brazil generated more than 2.1 million tons in the same year, according to the global e-waste report. Meanwhile, the recycling rates of these materials are still meager, not even reaching 2% of the total produced in the country, which increases negative impacts on the environment and human health due to improper disposal. In this sense, in 2010, Brazil established the obligation to structure and implement reverse logistics (RL) systems to collect and return WEEE to the productive sector. However, actions for structuring, implementing, and operationalizing the RL system (through the sectoral agreement signed in 2019) and the institution of the National Reverse Logistics Program are recent efforts in the country. Implementing reverse logistics systems is an essential step toward a circular economy. Nevertheless, a comprehensive solution must address more significant challenges, such as the shortage of components and the deficit of raw materials. In this scenario, it is evident that there is a need for a system that allows tracking the flow of WEEE, from manufacturing to final disposal, optimizing information management, and the circularity of components and materials. Therefore, in this project, we want to evaluate the technical feasibility of proposing the development of an identification and traceability system based on deep learning through computer vision and blockchain that enables the automation and agility of the recognition of WEEE and its subsequent monitoring until the final destination. The stage of autonomously identifying consists of two steps. The first is feature extraction from the objects where each device will have a vector with a signature. Furthermore, this first module will be able to label the device type. The second is the use of this feature extractor together with a previously trained CNN (Convolutional Neural Network) model to identify the devices in the final destination. With this feature extraction from the devices, we have the possibility of individually tracking each device throughout the LR flow and recording its guarantee in a digital ledger. The main advantage of implementing a traceability system on the blockchain is the ease of proving the veracity, authenticity, and uniqueness of the records of the RL system model. The development of this identification and traceability system would complement GAIA GreenTech's work in intelligent management and encourage people's responsible disposal. (AU)