Simulation-Based Optimization for Decision Making Under Uncertain Production Environments

Agricultural decision-making is inherently challenging, as producers must choose among alternative management strategies while outcomes depend on uncertain climatic, production, and market conditions, making it difficult to identify practices that balance expected profitability and risk exposure. This study proposes a simulation-based optimization framework that integrates a production function with stochastic economic scenarios to evaluate management strategies across multiple plausible conditions. The methodology is illustrated through an empirical application to nitrogen fertilization decisions in wheat using long-term experimental data from Oklahoma, USA. Preliminary results show that the simulationoptimization model recommends a higher nitrogen application rate and achieves higher average and median returns relative to benchmark and ensemble approaches. In addition, the framework generated the best economic outcome in 66% of evaluated scenarios, compared with 29% for the ensemble model and 5% for the benchmark model. However, these improvements are accompanied by greater return variability, suggesting a potential trade-off between economic performance and outcome stability.