Optimizing Multi-Crop AgriculturalProduction under Uncertainty: A Two-Stage Stochastic Approach

Abstract

Efficient crop planning is very important in agriculture because farmers need to allocate limited resources such as land, labour, and budget while facing uncertainty in crop yield, demand, and prices. In this research, deterministic and stochastic programming models are developed to determine the optimal land allocation and crop production plan that maximizes farmers’ profit while meeting food scarcity. All parameters in the deterministic models are certain. To address the uncertain parameters, a 2-stage stochastic programming model is developed where the land allocation, demand, selling, and purchasing decisions are considered as uncertain. Primary data are collected from direct interviews with the farmers to validate the model. Secondary data collected from the Bangladesh Bureau of Statistics (BBS) Yearbook 2023 are also used. Uncertainty is analysed through a stochastic model with 4 scenarios. Both models are implemented and solved using the AMPL programming language. A comparative analysis indicates that although the deterministic model yields higher profit under fixed conditions, the stochastic model provides more robust and reliable decisions by incorporating uncertainty. Therefore, the proposed stochastic crop planning model can serve as an effective decision-support tool for farmers to manage agricultural risks and improve overall farm profitability.