Palková et al. (2025) Data-Driven Optimisation of Irrigation Dose Using Machine-Learning Ensembles for Sustainable European Agriculture

Identification

• Journal: Agris on-line Papers in Economics and Informatics
• Year: 2025
• Date: 2025-12-29
• Authors: Zuzana Palková, Miroslav Žitňák, Ján Valíček, Marta Harničárová, Miroslav Holý, Daniel Levak, Hakan Tozan, Karol Görči
• DOI: 10.7160/aol.2025.170407

Research Groups

CODECS project (Nitra region, Slovakia)

Short Summary

This study developed and evaluated machine learning ensemble models to optimize irrigation doses for sustainable agriculture in the semi-arid Nitra region of Slovakia, aiming to reduce water usage and enhance crop productivity.

Objective

• To predict optimal irrigation doses using digital technologies and statistical modelling to enhance water resource management in agriculture.
• To evaluate the effectiveness of various irrigation systems and develop predictive models for optimal irrigation doses.

Study Configuration

• Spatial Scale: Semi-arid Nitra region, Slovakia
• Temporal Scale: Not explicitly defined, but implies recent data collection for model development.

Methodology and Data

• Models used: Machine learning ensembles, statistical modelling, regression models.
• Data sources: Environmental sensor data (IoT sensors), agronomic models, Valley and Irriga control software. Key factors analyzed include precipitation, temperature, soil moisture, and energy consumption. A unified methodology for heterogeneous data collection, validation, and analysis was developed.

Main Results

• Predictive models for optimal irrigation doses were developed using machine learning ensembles.
• Key factors affecting irrigation efficiency, such as precipitation, temperature, soil moisture, and energy consumption, were identified.
• The findings aim to inform sustainable irrigation strategies that reduce water usage, enhance crop productivity, and safeguard soil resources under changing climatic conditions.

Contributions

• Development of a data-driven approach using machine learning ensembles for optimizing irrigation doses in European agriculture.
• Creation of a unified methodology for collecting, validating, and analyzing heterogeneous data from various sources (IoT sensors, control software).
• Application and evaluation of advanced digital technologies and statistical modelling for water resource management in the specific context of the semi-arid Nitra region of Slovakia.

Funding

• CODECS project

Citation

@article{Palková2025DataDriven,
  author = {Palková, Zuzana and Žitňák, Miroslav and Valíček, Ján and Harničárová, Marta and Holý, Miroslav and Levak, Daniel and Tozan, Hakan and Görči, Karol},
  title = {Data-Driven Optimisation of Irrigation Dose Using Machine-Learning Ensembles for Sustainable European Agriculture},
  journal = {Agris on-line Papers in Economics and Informatics},
  year = {2025},
  doi = {10.7160/aol.2025.170407},
  url = {https://doi.org/10.7160/aol.2025.170407}
}