Dams are civil engineering structures to hinder water flows. Criteria such as purpose, size and construction material are useful to categorise them. The latter is used to classify ‘earth dams’, which tend to have higher risk levels than other types. The failure of a dam leads to significant economic loss and usually to catastrophic impacts. In an effort to comprehensively examine the variables that influence earth dam breaks and describe their interactions, a model has been developed in such a way that it allows to assess risks and to prioritise the allocation of resources for maintenance activities. The research was carried out by systematically reviewing the literature, which led to the choice of Bayesian Networks (BNs) as a tool for assessing risks. Using data from seven case studies in Mexico, a model was built, which helped to rank the dams under study, leading to results comparable with those reported in the literature. While the particular type of BN used and its quantification is presented more extensively in an accompanying paper, the model may be of interest for dam owners, managers, practitioners and academics on their efforts to manage earth dams’ risks.

In this paper, nonlinear static pushover analyses were carried out to assess the capacity and collapse mechanism of two existing transmission towers under wind loading. Different load distribution patterns were adopted from the application of the codified design wind load of the following countries: Mexico, United States, India, Japan and the New Zealand – Australian Code. Three-dimensional inelastic response analyses were performed using open access software. Analyses were employed to define the capacity curves, stress-strain curves for structural elements, the yielding mapping sequences and the collapse mechanisms, and to evaluate the influence of the tower body deformation and to assess the theoretical overstrength and ductility capacities. Results show a damage concentration, which leads to a fragile collapse mechanism with important strength reserves and a non-uniform distribution of yielding within the tower height. Since the collapse mechanism is not compatible with the desired performance inherent to the design philosophy, recommendations for the design stage are proposed, which pretend to ensure that the inelastic behaviour be consistent with the goals implicit in a code-based design to prevent tower collapses.

The time-variant failure probability for critical slopes under strong rainfalls, which cause reductions on the soil shear strength, is calculated and compared to the target failure probability. The soil properties and the rainfall characteristics are considered as random and the correlation between rainfall intensity and duration is included to assess the impact of water infiltration on the slope failure probability. The failure probability, defined as the probability that the safety factor is less than 1, is calculated through a Monte Carlo simulation process. The paper emphasizes the importance of the time during the water infiltration into the soil, as the rainfall sequence occurs. The target failure probability, derived from the minimum expected life-cycle cost, is compared to the slope failure probability to decide if the slope requires mitigation measures. If mitigation is required, the slope model is modified and a new annual failure probability is calculated. The slope annual failure probability, for three sites under strong rainfall hazard, is found to be around 0.78. However, by introducing mitigation actions, the failure probability reduces to 0.037. The slope failure probability allows for the risk-based prioritization of the attention for many slopes at several sites by stablishing basis for optimal resources allocation.