Work

Machine learning optimization of 1-MCP and chitosan/Aloe vera coatings for climate-resilient, sustainable fig postharvest preservation

Year: 2026
Type: conference-paper
Abstract: . Figs (Ficus carica) are among the most popular fruit crops, loaded with an ample supply of essential elements, dietary fiber, and bioactive molecules with high antioxidant activity. Despite this, they are quite perishable in nature, hence, extremely susceptible to postharvest microbial spoilage. To prevent this loss, researchers have been trying different methods, including cold storage, modified atmosphere packaging, chemical treatments, and coatings. However, measuring shelf life based on laboratory analysis is quite challenging, laborious, and involves the use of chemical reagents for quality assessment, indirectly impacting the environment. Therefore, the current study is designed to promote a dry lab approach to predict the shelf life of Fig fruits treated with 1-Methyl cyclopropane (1-MCP), and edible coating based on Aloe vera (AV) and chitosan (CH). After treatment, fruits were kept for storage at 2ºC with 80-85% RH. Quality studies were done every week, and the data produced was used to predict the shelf life using Artificial Neural Networks (ANNs) and Extreme Gradient Boosting (XGBoost) machine learning platforms. The results indicated that the ANN performed worse than a simple mean predictor, whereas XGBoost performed well. Therefore, this approach can help create the best model for shelf-life prediction, as it supports climate-resilient and sustainable postharvest practices.
Author: Navjot KAUR
Language: English
Subfield: Plant Science
Open Access status: bronze
OpenAlex ID: W7164500421
FWCI:
Cites: 24
Cited by:
Related to: 10