This study presents a machine learning workflow for predicting the corrosion resistance of a self-healing epoxy coating containing ZIF-8@Ca microfillers. The workflow used orthogonal Latin square methods to investigate the effects of four parameters on the coating's low impedance modulus. A random forest (RF) model was selected for active learning, achieving good prediction accuracy after five cycles. Bayesian optimization identified the best coating formulation. The resulting coating showed excellent self-healing and corrosion resistance, with minimal corrosion and adhesion loss after 60 days of neutral salt spray testing.
