Multiobjective Optimization Framework for Irrigation Scheduling under Uncertainty

The increasing water demand, driven by population growth and industrial expansion, is intensifying the global water scarcity crisis due to excessive consumption and competition among urban, industrial, and agricultural sectors. To address this pressing challenge, an urgent need is to optimize irrigation scheduling strategies to enhance water usage efficiency in cultivated lands. This study investigates the arranged method, a flexible irrigation scheduling approach, to optimize irrigation scheduling as a multiobjective problem. Employing stochastic and robust optimization (RO) techniques to address uncertainties in irrigation systems, three distinct mathematical models are developed: deterministic, stochastic, and robust models. The augmented ε-constraint method is utilized to tackle the multiobjective problem. To validate the proposed mathematical models, an irrigation system in the southwest of Iran serves as the case study. The General Algebraic Modeling System (GAMS) software is employed to implement the multiobjective problem. Comparative analysis of the computational results reveals that the stochastic models yield more reliable solutions, making them suitable for risk-neutral decision-makers. Calculating the expected value of perfect information (EVPI) reveals that prioritizing the acquisition of more precise data regarding uncertain parameters could potentially enhance system revenue by up to 27%. Conversely, the value of the stochastic solution (VSS) highlights that neglecting the inherent uncertainty associated with these parameters poses a substantial risk, potentially incurring revenue losses of up to 50%.