Proposal for an automated system for the efficient classification of Persea americana

Abstract

This study presents the design and implementation of an automated system for the efficient sorting of Persea americana (avocado) in the city of Cajamarca, Cajabamba province, Condebamba district. The objective is to replace the traditional manual process, which is characterized by high variability, errors due to visual fatigue, and high operating costs. The research was conducted in four phases: diagnosis of the manual process, design of the automated prototype based on machine vision and mechanical control, comparative performance evaluation, and economic analysis. The proposed system integrates a conveyor belt, ultrasonic sensors, and pneumatic actuators, allowing the fruit to be sorted according to size, ripeness, color, and surface defects. The results show significant improvements: a 58.7% reduction in processing time, a 79.7% decrease in errors, a 96.2% increase in accuracy, and a 142.4% increase in production capacity. The economic analysis demonstrates a payback period of 1.29 years and a 49.7% reduction in operating costs. It is concluded that automation is technically and economically viable, improving the competitiveness of the agro-industrial sector and aligning with the principles of Agriculture 4.0.