ARTIFICIAL INTELLIGENCE IN SHELF-LIFE PREDICTION AND WASTE REDUCTION OF FRUITS AND VEGETABLES
Al-Shodhana · 30.01.2026 · 14 (1) · 383-392
DOI: 10.70644/as.v14.i1.32 ↗Kurz gefasst
Fachartikel in der Zeitschrift Al-Shodhana, online erschienen am 30.01.2026. Verfasst von Ms. Krithika und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Fruits and vegetables are part of a healthy diet because they furnish a wide range of vitamins, minerals, fibre, and antioxidants, along with natural flavour and freshness. However, their delicate and easily perishable nature makes storage, transportation, and marketing quite challenging. Seasonal changes and poor handling often lead to significant post-harvest losses, lower customer satisfaction, and financial strain for farmers and sellers. Traditional quality checks, such as visual inspection or laboratory testing, are still used, but tend to be time-consuming and labour-intensive. Globally, food waste has been a serious concern, with nearly one-third of food produced each year going to waste. In Indonesia, for instance, much of this waste comes from households due to limited awareness about proper storage practices, misunderstanding of expiry dates, and the lack of eficient ways to determine freshness. Recently, machine learning has emerged as a promising solution to these challenges. It can analyse complex data and reveal hidden patterns that are often missed by traditional methods. By analysing quality-related data on fruits and vegetables, researchers can develop forcasting systems that estimate shelf life, optimise storage conditions, and even anticipate consumer preferences - ultimately helping to reduce waste and improve food quality. This study examines three pre-trained AI models - MobileNetV2, VGG19, and EficientNetV2S - using the Kaggle “Fresh and Rotten Classification” dataset, which includes approximately 30,400 images. The purpose of employing these models is to develop AI applications capable of detecting spoilage, thereby contributing to the reduction of food waste. In the field of precision agriculture, artificial intelligence is widely used to evaluate crop quality during both harvest and postharvest stages. Detecting diseases and damage is particularly important, as postharvest decay not only it reduces yield but can also result in the production of harmful toxins. In this research, CNN-based methods such as U-Net, DeepLab, and Mask R-CNN are applied to identify and predict deteriorated regions in stored apples, effectively distinguishing decayed tissues from healthy ones. The study underscores the significance of postharvest physiology and management practices - including harvesting, handling, packing, storage, and hygiene - in minimizing losses and enhancing overall crop quality.
Zitieren
APA
Krithika, M., & Nishad, M. A. (2026). ARTIFICIAL INTELLIGENCE IN SHELF-LIFE PREDICTION AND WASTE REDUCTION OF FRUITS AND VEGETABLES. Al-Shodhana, 14(1), 383-392. https://doi.org/10.70644/as.v14.i1.32
Vancouver
Krithika M, Nishad MA. ARTIFICIAL INTELLIGENCE IN SHELF-LIFE PREDICTION AND WASTE REDUCTION OF FRUITS AND VEGETABLES. Al-Shodhana. 2026;14(1):383-392. doi:10.70644/as.v14.i1.32
BibTeX
@article{Krithika2026,
author = {Krithika, Ms. and Nishad, Ms. Anfha},
title = {ARTIFICIAL INTELLIGENCE IN SHELF-LIFE PREDICTION AND WASTE REDUCTION OF FRUITS AND VEGETABLES},
journal = {Al-Shodhana},
year = {2026},
volume = {14},
number = {1},
pages = {383--392},
doi = {10.70644/as.v14.i1.32}
}
EUVIMED