Due to seismically deficient details, existing reinforced concrete structures have low lateral resistance capacities. Since these building structures suffer an increase in axial loads to the main structural element due to the green retrofit (e.g., energy equipment/device, roof garden) for CO2 reduction and vertical extension, building capacities are reduced. This paper proposes a machine-learning-based methodology for allowable ranges of axial loading ratio to reinforced concrete columns using simple structural details. The methodology consists of a two-step procedure: (1) a machine-learning-based failure detection model and (2) column damage limits proposed by previous researchers. To demonstrate this proposed method, the existing building structure built in the 1990s was selected, and the allowable range for the target structure was computed for exterior and interior columns.

In South Korea, over 400,000 Non-building Structures are inadequately managed and exposed to potential risks due to insufficient inspection systems, leading to an increase in accidents and significant losses of life and property. Therefore, it is crucial for users to conduct proactive self-inspections to identify and mitigate potential hazards. This study reclassified Non-building Structures into four main categories by analyzing their structural characteristics and associated risks through statistical analysis. Among these, retaining walls, which account for the largest proportion, were systematically analyzed to identify common damage patterns. Based on this analysis, self-inspection checklists were developed for both non-experts and experts. The proposed process involves an initial visual inspection using a simple non-expert checklist, followed by a more detailed expert-level inspection if any anomalies are detected. The reliability of this process was validated through approximately 120 validation processes.

Existing reinforced concrete (RC) building frames constructed before the seismic design was applied have seismically deficient structural details, and buildings with such structural details show brittle behavior that is destroyed early due to low shear performance. Various reinforcement systems, such as fiber-reinforced polymer (FRP) jacketing systems, are being studied to reinforce the seismically deficient RC frames. Due to the step-by-step modeling and interpretation process, existing seismic performance assessment and reinforcement design of buildings consume an enormous amount of workforce and time. Various machine learning (ML) models were developed using input and output datasets for seismic loads and reinforcement details built through the finite element (FE) model developed in previous studies to overcome these shortcomings. To assess the performance of the seismic performance prediction models developed in this study, the mean squared error (MSE), R-square (R2), and residual of each model were compared. Overall, the applied ML was found to rapidly and effectively predict the seismic performance of buildings according to changes in load and reinforcement details without overfitting. In addition, the best-fit model for each seismic performance class was selected by analyzing the performance by class of the ML models.

In this study, in order to establish a strategy for developing an fire following earthquake risk assessment method that can utilize domestic public databases(building datas, etc.), the method of calculating the ignition and fire-spread among the fire following earthquake risk assessment methodologies proposed by past researchers is investigated After investigating and analyzing the methodology used in the HAZUS-MH earthquake model in the United States and the fire following earthquake risk assessment methodology in Japan, based on this, a database such as a domestic building data utilized to an fire following earthquake risk assessment method suitable for domestic circumstances (planned) was suggested.