Bolted joints are among the most common building blocks used across different types of structures, and are often the key component that sews all other structural parts together. Monitoring and assessment of looseness in bolted structures is one of the most attractive topics in mechanical, aerospace, and civil engineering. This invention proposes a new percussion-based non-destructive approach to determine the health condition of bolted joints. Due to the different interfacial properties among the bolts, nuts and the host structure, bolted joints can generate unique sounds when it is excited by impacts, such as from tapping. Characteristics of the sounds can be analyzed by data processing algorithms to obtain unique features and classify recorded tapping data. A simple machine learning model using the decision tree method can be employed to identify the bolt looseness level. The proposed approach can be used to produce a cyberphysical system that can automatically inspect and determine the health of a structure.