Objective To explore the application effect of constructing a surgical site infection (SSI) prediction model in perioperative infection prevention and control management in a tertiary general hospital. Methods A two-stage research design was adopted. Stage 1 (retrospectively constructing model): 50 021 hospitalized patients who underwent surgery from January 2019 to December 2022 were included to establish SSI risk prediction model. Stage 2 (prospectively evaluating intervention): 49 260 patients who underwent surgery from January to December 2023 were selected as the pre-intervention group, and 56 463 patients who underwent surgery from January to December 2024 were selected as the post-intervention group, effect of perioperative infection control intervention was evaluated. Statistical analysis of SSI relevant indicators before and after intervention was conducted. Results Among 50 021 cases in the modeling group, the incidence of SSI was 0.48% (n=242). Multivariate logistic regression analysis showed that surgical duration ≥60 minutes, preoperative length of hospital stay ≥7 days, incision type (class II, III), ASA grading (class III/IV), and preoperative peripheral white blood cell count>10×10⁹/L were all independent risk factors for SSI (all P<0.05). The area under the receiver operating characteristic curve (AUC) of this prediction model was 0.783 (95%CI: 0.712-0.854), and its predictive performance was significantly better than any single variable. After the intervention, the overall SSI incidence and SSI incidence of patients with selective surgery, class III incision, and aged≥ 60 years were all lower than those before the intervention, differences were all statistically significant (all P<0.05). The qualified rates of preoperative hospitalization days, 0.5-1 hours preoperative antimicrobial administration, preoperative hand-washing, intraoperative heat preservation, sterile operation of postoperative dressing change, as well as object surface cleaning and disinfection after intervention were all higher than those before intervention, and the differences were all statistically significant (all P<0.05). Conclusion The application of prediction model can identify high-risk surgical patients early, improve the implementation rate of core infection control measures, and effectively reduce the incidence of SSI.