Optimizing Crop Planting Schedule Considering Planting Window and Storage Capacity

Area of Science:

• Agricultural Science
• Data Science
• Operations Research

Background:

• Agricultural productivity is enhanced by technology, but field operations and resource use remain challenging.
• Optimal planting strategies for row crops are critical due to weather uncertainties and storage limitations.

Purpose of the Study:

• To develop a two-stage decision support system for optimizing crop planting schedules.
• To address weather uncertainties and farm resource constraints in agricultural planning.

Main Methods:

• A two-stage approach was proposed: 1) a weather prediction model for Growing Degree Units (GDUs) using 1D-CNN, and 2) an optimization model for planting schedules.
• The 1D-CNN model was compared against other prediction models for accuracy.
• The planting schedule optimization model was evaluated under fixed and flexible storage capacity scenarios.

Main Results:

• The 1D-CNN model demonstrated superior performance, achieving 7-8% RRMSE for GDU prediction across two locations.
• The optimization model generated planting schedules that evenly allocated harvested quantities over fewer harvesting weeks.
• The model proved robust across various scenarios, including different storage capacities and geographic locations.

Conclusions:

• The proposed decision support system effectively optimizes planting schedules by integrating weather predictions and resource constraints.
• The system enhances farm resource utilization and provides a robust solution for agricultural planning under uncertainty.