Presented here is a case study on the condition and performance monitoring (CPM) of a subsea blowout preventer (BOP) pipe ram. The proposed real-time CPM solution uses adaptive physics-based models that process sensor measurements at the point of origin (known as edge analytics). The adapted model coefficients are treated as a vector, the magnitude of which estimates the degree of health degradation and the phase of which identifies its source. The benefits of using an adaptive model-based approach over traditional machine-based learning and artificial-neural-networks solutions include zero algorithm-training times, broad applicability to BOPs, model modularity, and accurate health-degradation estimates. The proposed CPM methodology is validated on a BOP pipe ram using both operational and simulated data. A sensitivity study of the method to system uncertainty is also presented.