Web-Based Smart Irrigation System Using Machine Learning

Abstract

The Smart Irrigation System (SIS) represents an advanced approach for controlling, monitoring, and automating irrigation processes by incorporating artificial intelligence techniques, particularly Machine Learning (ML) methods. Although several ML models have been applied within SIS, there has been limited effort to systematically examine the empirical evidence related to these models. In addition, ML-based SIS solutions often encounter various challenges and raise concerns that require further investigation. This paper provides a systematic literature review of ML-driven SIS, offering a comprehensive and unbiased overview of existing research in this domain. A total of 55 studies published between 2017 and 2023 were analyzed, identifying nine widely adopted ML models. The review is structured around four key analytical dimensions: the type of ML technique used, estimation accuracy, comparative analysis of models, and the context of estimation. The results demonstrate that ML techniques applied in SIS show superior performance when compared to traditional methods. However, despite their advantages, the adoption of ML models in smart irrigation remains limited, indicating the need for further research to achieve more robust and generalizable outcomes. Based on these findings, the study also proposes future directions and guidelines to assist researchers and practitioners in understanding the key contributions, challenges, and advancements in this evolving field.