D.E.W. Meganingtyas, Dafik Dafik, I.H. Agustin, R.I. Baihaki, Z.R. Ridlo, A.B. Angrenani, R. Nisviasari
Vertical farming is a method of growing crops in which cultivated plants are arranged vertically using soil media within controlled indoor environments known as Controlled Environment Agriculture (CEA). Managing vertical farming systems requires precise regulation of air temperature, humidity, light intensity, and soil moisture levels. In recent years, Artificial Neural Networks (ANN) have emerged as powerful models for forecasting and decision support in precision agriculture. This study integrates ANN with a Resolving Efficient Dominating Set (REDS) graph optimization framework to enhance both data acquisition and forecasting efficiency in soil-moisture prediction. A total of 864 observations of temperature, air humidity, and soil moisture were recorded in four daily phases in an indoor Aloe vera vertical farm using DHT11 and capacitive soil moisture sensors. Four ANN models: Feedforwardnet, Patternnet, Fitnet, and Cascadeforwardnet, and two ANN architectures: ANN-466 and ANN-567, were trained and tested under a 70-30 data split. The results show that the best architecture for Phase 1 was the Feedforwardnet ANN-567 model, with an MSE of 0.3810. The best architecture for Phase 2 was the Patternnet ANN-567 model, with an MSE of 1.13× 10^-9. The best architecture for Phase 3 was the Cascadeforwardnet ANN-567 model, with an MSE of 1.12× 10^-10. Finally, the best architecture for Phase 4 was the Cascadeforwardnet ANN-567 model, with an MSE of 1.07× 10^-17. The integration of REDS and ANN establishes a cohesive and reproducible framework for precision irrigation management in CEA systems, offering spatial efficiency in sensor deployment and temporal accuracy in soil-moisture forecasting. Copyright © 2026 International Academic Press
Department of Mathematics, State University of Jakarta, Indonesia; PUI-PT Combinatorics and Graph, CGANT-University of Jember, Indonesia; Department of Mathematics, University of Jember, Indonesia; Department of Natural Science Education, University of Jember, Indonesia; Department of Mathematics Education, University of Jember, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock