Infrastructure systems — such as tunnels, bridges, dams, and underground facilities — often emit subtle signals before failure, much like a tree branch creaks before it breaks. These signals are not random; they follow recognizable organizational patterns that hold valuable information about stress, damage, and fracture progression. This project seeks to uncover how such signals — known as energy releases — form and evolve into structured networks during material failure. By reimagining these signals not as isolated events but as part of a dynamic, interconnected system, the work will provide new insight into how cracks initiate, propagate, and coalesce in both engineered and natural materials. The findings will strengthen public safety by enabling earlier, more accurate detection of failure in critical infrastructure. The project will also support national prosperity by training students in advanced sensing, data science, and failure prediction. Partnerships with engineering firms and engagement with policymakers will help ensure that these findings are translated into practice, building resilience into the built environment. This research project will conduct controlled laboratory experiments in rock and concrete to analyze energy releases associated with fracture under varying stress conditions. These energy releases will be modeled as evolving networks based on their spatial and temporal relationships. The research will quantify how stress direction, fracture history,