Agriculture, a crucial contributor to Australia’s GDP, exports, and economy, involves inherent risk and uncertainty. Amidst these challenges, Australian farmers must craft optimal crop and livestock strategies to minimize risk whilst meeting their objectives. Traditional mathematical programming methods have aided resource allocation but fail to adequately address the uncertainty surrounding future market conditions and input parameters. This thesis explores two-stage stochastic optimization to enhance Australian small farm performance under uncertainty. We model uncertain events impacting farm operations as probability distributions, aiming for improved resource allocation and risk management. The stochastic program maximizes mean profit, worst-case profit, and optimizes the superquantile. Compared to deterministic approaches, our model increases the mean profit by 4.3%, raises the lowest 10% profits by 20.5% via the superquantile objective, and elevates the minimum profit by 140.8% when maximizing the worst-case profit. Our approach facilitates strategic planning and risk management within Australia's farming sector.

Approved for public release. Distribution is unlimited.

Major, Australian Army