Over the last decades, the emergence of new technologies has inspired a paradigm shift for the fourth industrial revolution. For example, circular economy, data mining, and artificial intelligence (AI), which are multidisciplinary topics, have recently attracted industrial and academic interests. Sustainable structural health monitoring (SHM) also concerns the continuous structural assessment of civil, mechanical, aerospace, and industrial structures to upgrade conventional SHM systems. A damage detection approach inspired by the principles of data mining with the adoption of circular-economic thinking is proposed in this study. In addition, vibration characteristics of a composite bridge deck structure are employed as inputs of AI algorithms. Likewise, an artificial neural network (ANN) integrated with a genetic algorithm (GA) was also developed for detecting the damage. GA was applied to define the initial weights of the neural network. To aid the aim, a range of damage scenarios was generated and the achieved outcomes confirm the feasibility of the developed method in the fault diagnosis procedure. Several data mining techniques were also employed to compare the performance of the developed model. It is concluded that the ANN integrated with GA presents a relatively fitting capacity in the detection of damage severity.