Construcción de un prototipo de software en su primera fase que permita detectar e indicar los desechos de material reciclable doméstico con el uso de inteligencia artificial

Abstract

Circular economies are gaining attention as a potential way to increase society's prosperity while reducing resource demands, raw materials, and minimizing indirect costs. The need for a systemic approach and new collaboration mechanisms is emphasized. Recycling is essential for sustainability, as large amounts of plastic waste are being leaked into the oceans every minute. It is proposed to seek systemic solutions that positively contribute to recycling initiatives. In the business sector, corporate social responsibility initiatives promote recycling, reduction, and reuse. The development of a technological product is proposed to enable the identification and classification of recyclable waste at the source, generating metrics to optimize logistical and transportation processes. Through the use of Deep Learning and computer vision techniques, the goal is to implement a web application that identifies and classifies waste in real time, facilitating its proper disposal from the source. The collected information will enable quantification, notification, and the scheduling of an optimized collection process. The aim is to obtain data to measure and segment the proportion of material destined for recycling, serving as a basis for designing an optimal logistical model that encompasses the collection and final disposal of recyclable materials.