Abstract
Technology advancement has contributed significantly to productivity improvement in the agricultural sector. However, field operation and farm resource utilization remain a challenge. For major row crops, designing an optimal crop planting strategy is crucial since the planting dates are contingent upon weather conditions and storage capacity. This manuscript proposes a two-stage decision support system to optimize planting decisions, considering weather uncertainties and resource constraints. The first stage involves creating a weather prediction model for Growing Degree Units (GDUs). In the second stage, the GDUs prediction from the first stage is incorporated to formulate an optimization model for the planting schedule. The efficacy of the proposed model is demonstrated through a case study based on Syngenta Crop Challenge (2021). It has been shown that the 1D-CNN model outperforms other prediction models with an RRMSE of 7 to 8% for two different locations. The decision-making model in the second stage provides an optimal planting schedule such that weekly harvested quantities will be evenly allocated utilizing a minimum number of harvesting weeks. We analyzed the model performance for two scenarios: fixed and flexible storage capacity at multiple geographic locations. Results suggest that the proposed model can provide an optimized planting schedule considering planting window and storage capacity. The model has also demonstrated its robustness under multiple scenarios.
Highlights
Inspired by the performance of TBATS and 1D-convolutional neural networks (CNN) in time series analysis, we applied them for the prediction of Growing Degree Units (GDUs)
We propose a two-stage crop planting model that recommends the planting dates considering plant growth as well as logistical and capacity limitations
Our proposed methodology is divided into two stages: predicting GDUs based on historical data and developing an optimal planting schedule
Summary
With the current world population growth rate, it is anticipated that by the end of 2050, the agriculture system needs to support 10 billion people, and 3 billion increase from the current population (Ranganathan et al, 2018). Inspired by the performance of TBATS and 1D-CNN in time series analysis, we applied them for the prediction of GDUs. In the agricultural sector, various models are proposed to determine optimal yield quantity (Nafziger, 1994; De Bruin and Pedersen, 2008; Hörbe et al, 2013; Waongo et al, 2015). The focal research question in this study is to design an optimal planting schedule that will lead to consistent weekly harvesting quantity within storage capacity in the harvesting phase. This problem statement is based on the INFORMS Syngenta Crop Challenge 2021 (Syngenta Crop Challenge (2021) – the challenge).
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
