Crop growth in greenhouses is basically determined by the climate variables in the environment and by the amounts of water and fertilizers supplied by irrigation. The management of these factors depends on the expertise of agricultural technicians and farmers, usually assisted by control systems installed within the greenhouse. In this context, decision support features enable us to incorporate invaluable human experience so that we can take quick and effective decisions to ensure efficient crop growth. This work describes a real-time decision support system for greenhouse tomatoes that supports decisions at three stages – the supervision stage identifies climate sensor faults, the control stage maintains climate variables at setpoints, and the strategic stage identifies diseases affecting the crop and changes climate variables accordingly to minimize damage. The DSS was implemented by integrating a real-time rule-based tool into the control system. Experimental results show that the system increases climate control effectiveness, while providing support in preventing diseases which are difficult to eradicate. The system was tested by simulating the appearance of the disease and observing the real system response. The main contribution has been to demonstrate that production rules, which are mature and well-known in the artificial intelligence domain, can act as a shared technology for the whole system. This means that fault detection, temperature control and disease monitoring features are not dealt with in isolation.
Cañadas, J., Sánchez-Molina, J. A., Rodríguez, F., & del Águila, I. M. (2017). Improving automatic climate control with decision support techniques to minimize disease effects in greenhouse tomatoes. Information Processing in Agriculture, 4(1), 50–63. https://doi.org/10.1016/j.inpa.2016.12.002