Machine Learning Application in Horticulture and Prospects for Predicting Fresh Produce Losses and Waste: A Review

Area of Science:

• Agricultural Science
• Computer Science
• Data Science

Background:

• Fresh produce is vital for nutrition and food security.
• Horticultural production requires efficient and accurate operations.
• Postharvest losses and waste are significant global challenges.

Purpose of the Study:

• To review machine learning (ML) applications in horticulture.
• To assess ML's potential in predicting and reducing fresh produce losses and waste.
• To identify future research directions for ML in postharvest management.

Main Methods:

• Literature review of ML applications in preharvest and postharvest horticulture.
• Analysis of ML algorithms for classification and prediction tasks.
• Evaluation of ML's role in quantifying postharvest losses and waste.

Main Results:

• ML algorithms show satisfactory performance in classification and prediction for horticultural tasks.
• ML has demonstrated potential in various preharvest and postharvest applications.
• Existing ML models offer a foundation for predicting and mitigating produce waste.

Conclusions:

• Further investigation into advanced ML models or combinations is needed for enhanced prediction accuracy.
• ML holds significant promise for reducing postharvest losses and waste.
• Future research should focus on optimizing ML for practical application in horticultural supply chains.