Crop simulation models are efficient tools for evaluating management strategies and predicting crop yields. This study aimed to assess the ability of the AquaCrop model to simulate grain and biological yields of common bean (Phaseolus vulgaris L.) under the interactive effects of irrigation, nitrogen levels, and mulch (wheat residue). A field experiment was conducted over three years (2016–2018) in Shiraz, Fars Province, using a split-split plot arrangement within a randomized complete block design. Data from the first year were used for model calibration, while data from the second and third years were employed for validation. The results showed that AquaCrop performed satisfactorily in simulating bean yield, with good agreement under normal climatic conditions (R² > 0.80) and NRMSE values ranging from 6.7% to 14.2%. However, simulation accuracy declined in years exposed to severe climatic stress, as yields were underestimated in 2017 due to pronounced weather stress. The model also exhibited limitations in responding to nitrogen fertilization, overestimating yield at low nitrogen levels and underestimating it at high levels. The effect of mulch was mainly accounted for in the model through changes in soil surface evaporation. Overall, AquaCrop proved to be an efficient tool for bean irrigation management, although more precise parameter calibration under severe stress conditions and nitrogen management could further improve prediction accuracy.