Advances in nondestructive techniques for quality and safety assessment of olive products using AI and big data analytics: A systematic review

Abstract Background Ensuring quality and safety of olive products is critical for consumer health, market value, and cultural heritage. Traditional destructive techniques are limited by their time-consuming and sample damage. Nondestructive techniques (NDT) allow real-time and non-invasive evaluation of authenticity, contamination, ripeness, and chemical composition. Artificial intelligence (AI) and big data analytics (BDA) integration enhances data interpretation, decision-making, and predictive modeling fostering transformative impact on olive industry. Aims This systematic review aimed to comprehensively evaluate advances in NDT for quality and safety assessment of olive products, emphasizing the role of AI and BDA and highlighting future prospects for industrial applications. Methods Following PRISMA guidelines, a systematic search was conducted across PubMed, Scopus, Web of Science, and Google Scholar, from 2010 to 2026 involving spectroscopic, imaging, sensor-based methods, integrated with AI and BDA. Results NDT like near-infrared and Raman spectroscopy, ultrasound, hyperspectral and X-ray imaging, and electronic nose/tongue enable rapid and accurate assessment of chemical composition, contamination, and ripeness. When coupled with AI algorithms, they significantly improve detection sensitivity, classification accuracy, and real-time monitoring. BDA facilitate comprehensive data management, predictive modeling, and supply chain traceability. Computational pharmacology contributes to understanding complex chemical interactions and bioactive compound behavior in olive products, enhancing safety and quality assessments. Challenges include data heterogeneity, lack of standardization, high instrumentation costs, interpretability of AI models, and regulatory hurdles. Conclusion Integrating NDT with AI, BDA, and computational pharmacology offers a sustainable, efficient industrial pathway, with future research needed for standardization, portable device development, multi-sensor data fusion, and regulatory validation.