An AI-Based Prediction Model for Climate Change Effects on Crop production using IoT

Climate change influences agriculture’s sustainability over different aspects. This paper proposes a prediction model grounded on Artificial intelligence techniques to predict the climate change effects on the Crop production. The proposed model is known as the Prediction Model for Climate Change Effects on Crop production (PMCCE). PMCCE utilizes the Artificial intelligence techniques along with the IoT technology to predict the crop production taking in consideration the climate change factors as: Annual Precipitation, Climate impacts per Celsius $\circ {C}$, Co2 concentrations, and temperature. The proposed model provides decision-makers with a comprehensive overview of the suitability of number of the most needed crops (namely, maize, wheat, and rice) for both short-term and long-term cultivation. An optimized Convolutional Neural Network(CNN) is proposed in PMCCE to predict the most appropriate crop based on climate change data and to predict the most significant factors affecting crop production. The experimental results indicate that the proposed model efficiently predicts the most suitable crop that enhances yield production and is better suited to cultivation under specific climate change scenarios. Furthermore, PMCCE outperforms other prediction models that used different methods, as the accuracy of the PMCCE model for the three targeted crops, achieved 98.2%, 98.69%, and 98.1% for Wheat, Maize, and Rice, respectively. Additionally, the PMCCE model identified the most significant factors in climate change data that affect crop production: CO2 and the average temperature were the most significant factors that affects the crops production, then climate impacts, climate scenario and adaptation type.