The aim of this study is to create an association-rule prediction model for acute radiation dermatitis (ARD) in patients who underwent radiotherapy after radical breast-cancer surgery and to evaluate the clinical application value of the model. Data were obtained from 800 patients who underwent radiotherapy after breast-cancer surgery at the First Affiliated Hospital of Hebei North University between June 2014 and June 2024. The patients were randomly segregated into a model group and a validation group at a 6∶4 ratio. The FP-growth algorithm used in association-rule analysis was used to process the baseline dataset of the model group. An ARD prediction model was created based on effective strong-association rules. Internal validation was performed based on the model group. The C-index and a calibration curve were used to evaluate the consistency of the ARD association-rule prediction model. External validation was performed based on the model and validation groups. Receiver operating characteristic (ROC) curves for predicting ARD in the two groups were constructed. The difference in the area under the curve (AUC) between the two groups was evaluated. Based on the ARD association-rule prediction model, the incidence of ARD in patients with specific baseline-data combinations of“body mass index ≥ 28 kg/m²,”“serum albumin level < 45 g/L,”“chemotherapy history (yes),”and“average irradiation field area ≥ 15 cm²”ranged from 42% to 77%. The ARD association-rule prediction model with the highest incidence rate was validated. Based on internal validation, the C-index of the model is 0.883 (95% CI: 0.703-0.919). The calibration curve shows good consistency between the predicted and actual values. Based on external validation, the AUC predicted by the ARD model for the abovementioned two groups are 0.856 (95% CI: 0.701-0.892) and 0.839 (95% CI: 0.715-0.922), respectively. The ROC curve fitting is relatively ideal (χ² = 3.224，p = 0.216), and the difference in the AUC is not statistically significant (p = 0.157). The ARD association-rule model yields accurate prediction results and demonstrates high main-prediction efficiency; thus, it can contribute significantly to the prevention and treatment of ARD.