The interaction of defects in a defect colony has a higher pipeline failure risk than the case when such interaction is not considered. The goal of this study is to develop probabilistic interaction rule and predictive failure pressure model for pipelines with colony of corrosion defects. The proposed interaction rule is developed based on the logistic regression algorithm by considering pipe properties and colony configurations as the independent variables. A performance comparison with the existing interaction rules shows that the proposed interaction rule has the most accurate predictions. The failure pressure prediction model is developed by adding a correction factor to the Mixed-Type Interaction (MTI) method; and the correction factor is modeled using a multivariate linear regression using the pipe properties and adjacent defects characteristics as the independent variables. It is found that the proposed model provides unbiased and more accurate predictions. The results of the reliability analysis of an example pipeline with colony defects show that the interaction effect on the failure prediction plays a critical role in the structural performance of pipelines.