Explainable artificial intelligence to estimate the Sri Lankan (Ceylon) Tea crop yield
Accurate cash crop yield prediction is important in the context of changing climate to ensure sustainable development. Artificial intelligence (AI) has played a significant role in prediction in nonlinear systems including cash crop prediction. However, despite its global recognition, Sri Lanka'...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | Smart Agricultural Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772375525002321 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Accurate cash crop yield prediction is important in the context of changing climate to ensure sustainable development. Artificial intelligence (AI) has played a significant role in prediction in nonlinear systems including cash crop prediction. However, despite its global recognition, Sri Lanka's renowned Ceylon Tea has never been modeled for yield prediction based on meteorological, soil, and fertilizer parameters. Sri Lanka is one of the most impacted countries due to climate change while its agricultural production is significant in economic contribution. Therefore, this study presents the first-ever research to predict the tea crop yield with respect to meteorological, soil, and fertilizer features. State-of-the-art machine learning models were tested with the aid of Dampahala Tea Company production. Results showcased that the CatBoost algorithm produces the best prediction model with a coefficient of determination, RCatBoost2= of 0.90374 (other models tested, RXGBoost2=0.87385;RLightGBM2=0.79772;RANN2=0.57346;RRidge2=0.42033;RLASSO2=0.42512;RElasticNet2=0.42168). In addition, explainability aspects using SHAP showcased that the morning relative humidity, evaporation, and T0 200 fertilizer have a significant impact on the prediction model. The developed AI model can be disseminated across the country to enhance understanding of predictions under future climatic scenarios. |
|---|---|
| ISSN: | 2772-3755 |