Development of fuzzy logic algorithm for predicting heavy metal content in poultry product

Poultry products are important global protein sources but are vulnerable to contamination by toxic metals such as copper, cadmium, lead, and arsenic. Excessive intake of these metals poses health risks, necessitating reliable yet accessible detection methods. This study developed a fuzzy logic framework using the Mamdani inference system and triangular membership functions in MATLAB R2021b to estimate heavy metal concentrations in poultry products including eggs, meat, and liver. Three fuzzy logic models were constructed, and multiple rule sets of 25, 50, and 81 rules were tested. Results showed that the 50-rule model achieved accurate classifications while minimizing complexity, correctly identifying safe and unsafe products in line with FAO/WHO permissible limits. Validation using published laboratory data confirmed that the model classified samples exceeding lead values above 0.1 ppm and cadmium values above 0.05 ppm as unsafe, whereas those within safe ranges such as egg samples containing 0.243 ppm copper, 0.033 ppm lead, 0.002 ppm cadmium, and 0.003 ppm arsenic were correctly identified. The key novelty of this study lies in the integration of a fuzzy logic-based heavy metal safety prediction model with an accessible graphical user interface (GUI), enabling non-expert users to perform rapid and interpretable food safety assessments without requiring programming or toxicological expertise. The proposed system provides a rapid, cost-effective, and user-friendly alternative to laboratory testing, supporting food safety monitoring and public health protection.