Leveraging Neural Network Models in the New Generation of Precision Agriculture for Oil Palm Plantations

Authors: Muhdan Syarovy | Year: 2025 | DOI: https://doi.org/10.17605/OSF.IO/J46B5

Precision agriculture has entered a new era with the integration of artificial intelligence, particularly neural network models. For the oil palm plantation industry — which has traditionally relied on conventional agronomic approaches — this transformation opens up opportunities that were previously unimaginable. From yield prediction to early pest detection, and from nutrient analysis to water requirement optimization, neural networks are bringing about a paradigm shift in plantation management.

This article reviews various applications of neural networks in oil palm plantations, covering model architectures commonly used (CNN for image-based analysis, RNN/LSTM for time-series data, and multilayer perceptron for mixed data), the data requirements for each application, and practical considerations for implementation. This review serves as an introductory guide for plantation practitioners who want to understand the potential and limitations of this technology.

The study found that neural network applications in oil palm plantations can be grouped into three main categories, each with different levels of maturity and readiness for implementation. Yield prediction using time-series data is among the most developed applications, while seedling quality assessment using computer vision is the fastest-growing area. However, several challenges remain, primarily related to data availability and the human resource capacity needed to develop and maintain these systems in plantation settings.

Want to learn more about the full methodology and detailed findings? Read the full article here.