The interface zone between precast and cast-in-place concrete exhibits distinct physical discontinuities, making it a vulnerable region for durability degradation in concrete structures subjected to chloride ion ingress. Accurately predicting chloride transport and concentration distribution within this interface zone—while properly accounting for factors such as loading, material disparities, and multi-source uncertainties—is crucial for ensuring the durability and service safety of precast-cast-in-place concrete structures. Existing prediction methods primarily target monolithic cast-in-place concrete and struggle to effectively capture the heterogeneous transport characteristics of the precast-cast-in-place interface. Furthermore, traditional empirical models rely heavily on destructive techniques (such as surface roughening) to obtain interface parameters, while data-driven models often lack interpretability; consequently, neither approach fully meets the practical needs for durability prediction and assessment in complex service environments. Addressing these key challenges, this book systematically investigates chloride concentration prediction, data-driven probabilistic prediction, and hybrid-driven prediction at the precast-cast-in-place concrete interface under the influence of multiple factors. It integrates theories of chloride transport at the interface with empirical models, machine/deep learning models, uncertainty analysis, and data-physics hybrid-driven approaches. Additionally, the book establishes a method for assessing the probability of rebar corrosion based on the predicted chloride concentrations. The book covers the following main topics: an introduction; prediction models for chloride concentration at the interface under multiple factors; data-driven prediction models; prediction methods incorporating data augmentation and hyperparameter optimization; data-physics hybrid-driven prediction of chloride concentration and rebar corrosion probability; and conclusions and future outlook.