Abstract
This article presents methods and hardware for multi-criteria non-destructive determination of chicken egg quality parameters, implemented using a multifunctional rotary system. Unlike traditional single-criteria sorting, which relies primarily on weight, the proposed approach utilizes a combination of physical and geometric parameters, including weight, linear dimensions, cross-sectional area and perimeter, volume, density, and shape. The experimental framework for the study was formed by measuring the parameters of 750 chicken eggs, covering the entire range of product categories and morphological variations. Geometric parameters were determined using machine vision methods, weight was determined using a strain gauge, and derived parameters were calculated using formalized models. A multi-criteria evaluation algorithm based on fuzzy set theory was used to make the classification decision, accounting for overlapping feature ranges and regulatory differences between EU and EAEU standards. The results of statistical and correlation analysis showed that egg density is an independent diagnostic parameter, weakly correlated with weight and geometric characteristics, justifying its inclusion in the quality model. A comparison of manual and automatic classification revealed systematic biases in boundary categories during single-criteria sorting and demonstrated the advantages of a multi-criteria approach. The obtained results confirm the effectiveness of the developed methods and hardware for automated egg quality control in industrial poultry farming.