A Global Forecasting Approach to Large-Scale Crop Production Prediction with Time Series Transformers
Accurate prediction of crop production is essential in effectively managing the food security and economic resilience of agricultural countries. This study evaluates the performance of statistical and machine learning-based methods for large-scale crop production forecasting. We predict the quarterl...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Agriculture |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-0472/13/9/1855 |
| _version_ | 1797581688202592256 |
|---|---|
| author | Sebastian C. Ibañez Christopher P. Monterola |
| author_facet | Sebastian C. Ibañez Christopher P. Monterola |
| author_sort | Sebastian C. Ibañez |
| collection | DOAJ |
| description | Accurate prediction of crop production is essential in effectively managing the food security and economic resilience of agricultural countries. This study evaluates the performance of statistical and machine learning-based methods for large-scale crop production forecasting. We predict the quarterly production of 325 crops (including fruits, vegetables, cereals, non-food, and industrial crops) across 83 provinces in the Philippines. Using a comprehensive dataset of 10,949 time series over 13 years, we demonstrate that a global forecasting approach using a state-of-the-art deep learning architecture, the transformer, significantly outperforms popular tree-based machine learning techniques and traditional local forecasting approaches built on statistical and baseline methods. Our results show a significant 84.93%, 80.69%, and 79.54% improvement in normalized root mean squared error (NRMSE), normalized deviation (ND), and modified symmetric mean absolute percentage error (msMAPE), respectively, over the next-best methods. By leveraging cross-series information, our proposed method is scalable and works well even with time series that are short, sparse, intermittent, or exhibit structural breaks/regime shifts. The results of this study further advance the field of applied forecasting in agricultural production and provide a practical and effective decision-support tool for policymakers that oversee crop production and the agriculture sector on a national scale. |
| first_indexed | 2024-03-10T23:08:08Z |
| format | Article |
| id | doaj.art-0c68c1b772c8444ba3b7c877eabda403 |
| institution | Directory Open Access Journal |
| issn | 2077-0472 |
| language | English |
| last_indexed | 2024-03-10T23:08:08Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agriculture |
| spelling | doaj.art-0c68c1b772c8444ba3b7c877eabda4032023-11-19T09:08:15ZengMDPI AGAgriculture2077-04722023-09-01139185510.3390/agriculture13091855A Global Forecasting Approach to Large-Scale Crop Production Prediction with Time Series TransformersSebastian C. Ibañez0Christopher P. Monterola1Analytics, Computing, and Complex Systems Laboratory, Asian Institute of Management, Makati City 1229, PhilippinesAnalytics, Computing, and Complex Systems Laboratory, Asian Institute of Management, Makati City 1229, PhilippinesAccurate prediction of crop production is essential in effectively managing the food security and economic resilience of agricultural countries. This study evaluates the performance of statistical and machine learning-based methods for large-scale crop production forecasting. We predict the quarterly production of 325 crops (including fruits, vegetables, cereals, non-food, and industrial crops) across 83 provinces in the Philippines. Using a comprehensive dataset of 10,949 time series over 13 years, we demonstrate that a global forecasting approach using a state-of-the-art deep learning architecture, the transformer, significantly outperforms popular tree-based machine learning techniques and traditional local forecasting approaches built on statistical and baseline methods. Our results show a significant 84.93%, 80.69%, and 79.54% improvement in normalized root mean squared error (NRMSE), normalized deviation (ND), and modified symmetric mean absolute percentage error (msMAPE), respectively, over the next-best methods. By leveraging cross-series information, our proposed method is scalable and works well even with time series that are short, sparse, intermittent, or exhibit structural breaks/regime shifts. The results of this study further advance the field of applied forecasting in agricultural production and provide a practical and effective decision-support tool for policymakers that oversee crop production and the agriculture sector on a national scale.https://www.mdpi.com/2077-0472/13/9/1855crop productionagricultural productiontime series forecastingartificial intelligencemachine learningdeep learning |
| spellingShingle | Sebastian C. Ibañez Christopher P. Monterola A Global Forecasting Approach to Large-Scale Crop Production Prediction with Time Series Transformers Agriculture crop production agricultural production time series forecasting artificial intelligence machine learning deep learning |
| title | A Global Forecasting Approach to Large-Scale Crop Production Prediction with Time Series Transformers |
| title_full | A Global Forecasting Approach to Large-Scale Crop Production Prediction with Time Series Transformers |
| title_fullStr | A Global Forecasting Approach to Large-Scale Crop Production Prediction with Time Series Transformers |
| title_full_unstemmed | A Global Forecasting Approach to Large-Scale Crop Production Prediction with Time Series Transformers |
| title_short | A Global Forecasting Approach to Large-Scale Crop Production Prediction with Time Series Transformers |
| title_sort | global forecasting approach to large scale crop production prediction with time series transformers |
| topic | crop production agricultural production time series forecasting artificial intelligence machine learning deep learning |
| url | https://www.mdpi.com/2077-0472/13/9/1855 |
| work_keys_str_mv | AT sebastiancibanez aglobalforecastingapproachtolargescalecropproductionpredictionwithtimeseriestransformers AT christopherpmonterola aglobalforecastingapproachtolargescalecropproductionpredictionwithtimeseriestransformers AT sebastiancibanez globalforecastingapproachtolargescalecropproductionpredictionwithtimeseriestransformers AT christopherpmonterola globalforecastingapproachtolargescalecropproductionpredictionwithtimeseriestransformers |