The rapid classification of building damage states or placards after an earthquake is vital for enabling an efficient emergency response and informed decision-making for rehabilitation and recovery purposes. Traditional methods rely heavily on inspector-led on-site surveys, which are often time-consuming, resource-intensive, and susceptible to human error. This study introduces a machine learning-supported surrogate model designed to streamline the assessment of building damage, focusing on the automated assignment of damage placards within the context of New Zealand's post-earthquake evaluation frameworks. The study evaluates two key safety evaluation protocols—Rapid Building Assessment (RBA) and Detailed Damage Evaluation (DDE)—and integrates corresponding databases derived from the 2010–2011 Canterbury Earthquake Sequence (CES) in Christchurch. Six ML classifiers—Multilayer Perceptron (MLP), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbours (KNN), Gradient Boosting Classifier (GBC), and Gradient Bagging (GBag)—were rigorously tested across both databases. The results indicate that the RF-based surrogate model outperforms the other classifiers across both RBA and DDE protocols. Two distinct sets of critical predictors have been further identified for each protocol, allowing for the rapid retrieval of essential data for future on-site surveys, while retaining the RF model's predictive accuracy. The developed surrogate model provides a pragmatic tool for practising engineers to rapidly assign placards to damaged structures and for policymakers and building owners to make informed recovery decisions for earthquake-affected buildings
High demolition rates were observed in New Zealand after the 2010-2011 Canterbury Earthquake Sequence despite the success of modern seismic design standards to achieve required performance objectives such as life safety and collapse prevention. Approximately 60% of the multi-storey reinforced concrete (RC) buildings in the Christchurch Central Business District were demolished after these earthquakes, even when only minor structural damage was present. Several factors influenced the decision of demolition instead of repair, one of them being the uncertainty of the seismic capacity of a damaged structure. To provide more insight into this topic, the investigation conducted in this thesis evaluated the residual capacity of moderately damaged RC walls and the effectiveness of repair techniques to restore the seismic performance of heavily damaged RC walls. The research outcome provided insights for developing guidelines for post-earthquake assessment of earthquake-damaged RC structures. The methodology used to conduct the investigation was through an experimental program divided into two phases. During the first phase, two walls were subjected to different types of pre-cyclic loading to represent the damaged condition from a prior earthquake, and a third wall represented a repair scenario with the damaged wall being repaired using epoxy injection and repair mortar after the pre-cyclic loading. Comparisons of these test walls to a control undamaged wall identified significant reductions in the stiffness of the damaged walls and a partial recovery in the wall stiffness achieved following epoxy injection. Visual damage that included distributed horizontal and diagonal cracks and spalling of the cover concrete did not affect the residual strength or displacement capacity of the walls. However, evidence of buckling of the longitudinal reinforcement during the pre-cyclic loading resulted in a slight reduction in strength recovery and a significant reduction in the displacement capacity of the damaged walls. Additional experimental programs from the literature were used to provide recommendations for modelling the response of moderately damaged RC walls and to identify a threshold that represented a potential reduction in the residual strength and displacement capacity of damaged RC walls in future earthquakes. The second phase of the experimental program conducted in this thesis addressed the replacement of concrete and reinforcing steel as repair techniques for heavily damaged RC walls. Two walls were repaired by replacing the damaged concrete and using welded connections to connect new reinforcing bars with existing bars. Different locations of the welded connections were investigated in the repaired walls to study the impact of these discontinuities at the critical section. No significant changes were observed in the stiffness, strength, and displacement capacity of the repaired walls compared to the benchmark undamaged wall. Differences in the local behaviour at the critical section were observed in one of the walls but did not impact the global response. The results of these two repaired walls were combined with other experimental programs found in the literature to assemble a database of repaired RC walls. Qualitative and quantitative analyses identified trends across various parameters, including wall types, damage before repair, and repair techniques implemented. The primary outcome of the database analysis was recommendations for concrete and reinforcing steel replacement to restore the strength and displacement capacity of heavily damaged RC walls
Road networks are highly exposed to natural hazard events, which can lead to significant economic and social consequences. In New Zealand, events such as the 2011 Christchurch earthquake, the 2016 Kaikōura earthquake, and the Cyclone Gabrielle in 2023 have demonstrated the severe consequences of road network disruptions. Traditional post event economic assessments often focus solely on clean-up and repair costs, neglecting the broader and more enduring impacts these events can have. Furthermore, business cases for resilience investments usually fail when quantifying the economic benefits of mitigation strategies, due to the underestimation of road disruption consequences. Importantly, not all road link disruptions contribute equally to these consequences, making the identification of critical road links a key step in resilience focused investment prioritization. Furthermore, traditional transportation asset management typically evaluates the life cycle of roads under normal conditions, such as traffic loads and standard environmental factors, while neglecting the influence of natural hazards. However, these events can significantly alter road deterioration and increase maintenance costs, emphasizing the need for integrating risk and resilience into transportation asset management approaches. This thesis presents a methodology to evaluate road criticality by assessing the economic consequences of road disruptions in combination with a hazard model in a prioritization index. Initially, the consequences are quantified through increased travel time, higher vehicle operating costs, and increased gas emissions. Thereafter, a new consequence model is introduced to estimate the increase in maintenance costs on alternative routes that absorb diverted traffic following a disruption. These consequence models are initially applied in a 'full-scan' analysis approach, where each road link is removed in turn to quantify its potential impact and, therefore, its criticality. Subsequently, a hazard model is integrated to develop a road prioritization index that combines the expected impacts of road disruptions, the individual road link criticality, and the probability of occurrence of natural hazard events. This index is designed to help road agencies in prioritizing mitigation strategies. Furthermore, the proposed methodology can also be applied to quantify the indirect economic impacts of natural hazard events. The methodology is demonstrated using New Zealand’s South Island inter-urban network as a case study, incorporating an earthquake-induced landslide model, with Python based simulations, providing road agencies a valuable tool to quantify the economic benefits of resilience investments
