ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering

Unlike previous models that do not consider uncertainties, this paper proposes a new nondeterministic method to estimate the high-cycle fatigue (HCF) resistance of welded hollow spherical joints (WHSJs) in long-span structures, including bridges, gymnasiums, and factories. This macro damage mechanics model with failure probability is introduced by accounting for the uncertainties of stress concentration and weld defects. Then artificial neural networks (ANNs) for three kinds of load type are built as a substitute for a finite-element (FE) model to obtain the concentrated stress more efficiently. Subsequently, the probability model of the defect factor is identified as a Gaussian distribution, while the stress concentration factor (SCF) has a Gaussian distribution and three-parameter t distribution. Also, a series of fatigue test on WHSJs are used to validate the proposed model, yielding a reasonable fatigue life prediction. Finally, fatigue failure probability analysis, which includes a joint probability density function (PDF), is conducted using the new nondeterministic method, which could provide a reference for fatigue design and damage quantification of WHSJs in long-span structures. Meanwhile, two Monte Carlo simulations corresponding to both possible distributions of concentrated stress were run to verify the accuracy of the HCF failure probability model of WHSJs. The results guarantee the feasibility of the proposed probability model applied in HCF fatigue design for weld joints of long-span structures.

A novel methodology for the quantification of mixed and nested polymorphic uncertainties has been developed. It is designed to be applied to moderately computationally expensive deterministic models, and it preserves the distinction between aleatory and epistemic uncertainties throughout the entire process. Quasi-Monte Carlo sampling is used to efficiently represent the local and global behavior of the models. By decoupling uncertainty propagation and processing, the methodology achieves efficient reuse of samples and can support multiple outputs. In addition, simultaneous estimation of sensitivity indices is possible to facilitate decisions on where to reduce epistemic uncertainties. It is demonstrated on a structural dynamics example and compared with a fully stochastic approach using the pignistic transform. The proposed methodology has been demonstrated to be significantly more efficient than a naive implementation, but adds computational cost compared with a fully stochastic approach.

The spatial variability of soil is usually modeled as a nonstationary unconditional random field (URF), which utilizes only the statistics of the soil field, and conditional random field (CRF), which utilizes both the statistics and the actual data at “sampled” locations. The simulated undrained shear strength of soil and random bearing capacity of monopiles using URF and CRF are compared, and the results show that CRF can more accurately reflect the real soil strength profile and significantly reduce the coefficient of variation of random bearing capacity of monopiles. However, of course, the priority of CRF over URF is based on the quality of the sampled data. The effects of various site investigation parameters (vertical investigation interval, investigation depth, and horizontal investigation distance) on the bearing capacity of monopiles are investigated. It can be concluded that, to ensure the quality of sampled data for establishing CRF, the vertical investigation interval and horizontal investigation distance should be less than 0.75 times the vertical and horizontal scales of fluctuation, respectively, and the investigation depth should be greater than 1.2 times the embedment depth of monopiles. Because the estimation of the horizontal scale of fluctuation is rarely performed due to the sparse in situ data in the horizontal direction, the worst-case scale of fluctuation for monopiles is also investigated. The worst-case horizontal scale of fluctuation mainly depends on the monopile diameter and is 5–7 times the diameter of monopiles. Therefore, to be on the safe side, it is suggested that the horizontal scale of fluctuation be 5–7 times the diameter of monopiles in the design and evaluation of monopile foundations.

The heteroscedastic Bayesian model updating framework assigned different variances to the modal errors using heteroscedastic parameters modeled by gamma distributions. However, the error density shows significant asymmetry, not captured by the assumed Student’s $t$-distribution. Thereby, the effect of heteroscedasticity is not adequately reflected in the variances of the updated stiffness and the prediction error variances due to hindered error propagation along the Markov chain Monte Carlo (MCMC) chain. This is overcome in the present study by proposing a Gaussian mixture–based autoregressive model in a conditional heteroscedastic framework (which is termed GMARCH). The GMARCH model adjusts the error intermittently at different stages of the MCMC chain and models the unknown error and its variances at any stage with respect to the previous stages. The proposed heteroscedastic error model obtains a direct estimate of the most probable values of the heteroscedastic parameters for the modal observables at different modes. An existing experimental data set derived from a multi-degree-of-freedom spring-mass model is used to illustrate the effectiveness of the model in addition to simulated data from a multistory shear building. The accuracy and computational effectiveness of the proposed approach are compared to those of the existing methods.

Ship trajectory prediction plays an important role in ensuring ship safety; through accurate ship positioning, the future trajectory of ships and their encounter time and location can be obtained, which facilitates the maritime regulatory authorities to assess the risks of ship encounters and implement effective traffic control. Meanwhile, with the rapid development of the shipping industry, the increasingly complex maritime traffic poses potential risks, which may cause serious traffic accidents and huge economic losses. To improve the accuracy of ship navigation risk prediction and ensure the safety of ship navigation, automatic identification system (AIS) data and deep learning models are used to extract the ship trajectory change feature pattern and apply it to ship trajectory prediction. This study builds the improved bidirectional long short-term memory network (Bi-LSTM) model based on rectified adaptive moment estimation (Radam) and lookahead, respectively. The AIS data of the Port of Tianjin area were selected for model training, and the results of comparison experiments show that the improved Bi-LSTM model has a stronger generalization ability, which further improves the trajectory prediction accuracy, and shows excellent predictive performance. The prediction model is feasible for the prediction of ship navigation trajectory.

This research describes a novel method for approximating the survival signature for very large systems. In recent years, the survival signature has emerged as a capable tool for the reliability analysis of critical infrastructure systems. In comparison with traditional approaches, it allows for complex modeling of dependencies, common causes of failures, as well as imprecision. However, while it enables the consideration of these effects, as an inherently combinatorial method, the survival signature suffers greatly from the curse of dimensionality. Critical infrastructures typically involve upward of hundreds of nodes. At this scale analytical computation of the survival signature is impossible using current computing capabilities. Instead of performing the full analytical computation of the survival signature, some studies have focused on approximating it using Monte Carlo simulation. While this reduces the numerical demand and allows for larger systems to be analyzed, these approaches will also quickly reach their limits with growing network size and complexity. Here, instead of approximating the full survival signature, we build a surrogate model based on normalized radial basis functions where the data points required to fit the model are approximated by Monte Carlo simulation. The resulting uncertainty from the simulation is then used to build an interval predictor model (IPM) that estimates intervals where the remaining survival signature values are expected to fall. By applying this imprecise survival signature, we can obtain bounds on the reliability. Because a low number of data points is sufficient to build the IPM, this presents a significant reduction in numerical demand and allows for very large systems to be considered.

Due to the complex modern construction processes and the diversity of building materials, a diversified construction environment has emerged, making it difficult to accurately predict and monitor the risk of fire hazards. Therefore, the development and implementation of a fire hazard detection system are necessary to effectively prevent fires at the source and to adequately protect construction personnel and building assets. On construction sites, early warning for fires often occurs after the fire has occurred, resulting in missing the best time for fire rescue. However, building information modeling (BIM) and wireless sensor networks have been widely discussed in many aspects of building disaster prevention management as methods for fire detection and early warning. In this study, a fire hazard intelligent detection system is constructed using a cloud computing platform and Bluetooth sensors. The system uses embedded Bluetooth wireless sensors to form a topology, and the cloud computing platform and user interface are written based on Python to create a two-way fire hazard detection system that displays real-time and dynamic detection information. The results based on practical case applications show that the cloud computing platform processed can effectively provide monitoring information to support fire safety assessment and planning, early detection and alarm response, guide effective decision-making, and facilitate fire rescue and control efforts to improve overall building safety and disaster response capabilities.

This study proposes a method to select optimal sites for human-life search and rescue operations in inland waters. It was used to bridge the disparity between the distribution of current water search and rescue facilities and the nuanced spatial distribution of risks associated with water traffic accidents. An evaluation model for the demand of search and rescue demand was developed considering regional water traffic flow, wind direction, wind speed, channel depth, and the historical frequency of water traffic accidents. An objective function was formulated to optimize the selection of search and rescue operation sites in inland waters. Finally, the method realized the precise layout planning of search and rescue sites in inland waters using the objective function. The effectiveness of the proposed method was verified by taking the Yangtze River as the research area, which improved 22.92% of the capacity to respond to life-threatening search and rescue emergencies.

Resonance and cancellation phenomena are fundamental scientific topics in the field of train–bridge interactions. Prior studies have explained their underlying mechanisms and consequential effects. This paper re-examined the two phenomena from a random perspective and delineated them as random resonance and random cancellation. Subsequently, a practical framework is proposed to quantify the probability distributions of train speeds that induce the resonance and cancellation in railway bridges considering random uncertainties. The proposed framework incorporates the input parameter design, stochastic modal analysis, modal identification, and sparse polynomial chaos expansion (PCE). In the modal identification, a novel indicator was developed which aids in identifying modal shapes and constructing eigenpairs involving eigenvalues and modal shapes. The framework’s validity and efficiency were confirmed using three bridge examples: a simply supported bridge, a continuous beam bridge, and a long-span cable-stayed bridge. The results show that the critical speeds for the resonance and cancellation exhibit significant variabilities. This work is expected to provide some reference for determining reasonable train operation speeds and optimizing railway bridge design.

Communities can face devastation from earthquakes, including loss of lives, destruction of infrastructure, and damage to buildings. By implementing rigorous standards, communities can ensure that both new and existing structures are designed and retrofitted to withstand seismic forces. While current building codes prioritize life safety during earthquakes, they may not fully consider the postdisaster functionality and usability of buildings. Consequently, this paper introduces a novel simulation-based approach that promotes resilience-informed decision-making related to seismic design requirements for individual buildings within a community. This multidisciplinary approach takes into account the interdependencies between the physical and social systems in the community in terms of population dislocation after earthquakes. Furthermore, the methodology herein produces cost estimates for implementing seismic design upgrades, ensuring meeting seismic new retrofit level requirements for existing buildings in the community. An illustrative example using Salt Lake County (SLC), Utah, US, subject to a hypothetical earthquake scenario, is presented to demonstrate the effectiveness of the simulation-based framework and decision support methodology. SLC encompasses a significant number of buildings, electric power substations, and power generation plants, serving a population of approximately 1 million. The analysis and simulations are conducted using the Interdependent Networked Community Resilience Modeling Environment (IN-CORE), an open-source platform that facilitates the execution of the proposed framework and methodology. Through this comprehensive approach, communities can enhance their preparedness, make informed decisions, and work toward building resilience in the face of seismic events.

The efficient balancing of ship passing operation efficiency and transport capacity is a crucial aspect of passing capacity in inland water systems. However, the current approach to designing ship passing capacity is based on expert experience rather than objective evaluation of results, which can result in a loss of efficiency or cause traffic jams. To address this issue, this study adopts the widely used traffic wave theory in road traffic analysis to investigate ship passing capacity in narrow waterways. A novel model is proposed that considers key elements such as ship traffic density, average speed, and waterway width to evaluate the performance of shipping traffic in the Anhui section of the Yangtze River in China. Using the proposed model, the ship passing capacity of the studied water area is analyzed on a monthly basis. The results indicate that passing capacity is influenced by ship density and waterway width, while the parameters of ship speed are less significant. This work provides a useful model for evaluating waterway passing capacity and further supports ship traffic management in inland waters.

The distance of supporting steel structures in tunnels is determined by experience, and, to our best knowledge, there are few precise guidelines in the current design of underground construction. This paper aims to propose an optimum distance of a steel arch structure taking the uncertainties of geomaterials into consideration. The probability of collapse failure of soils behind steel arch structures in tunnels is assessed by using Chebyshev inequalities. First, the critical distance of a steel arch structure is derived based on the limit equilibrium method. Note that the distribution types of shear strength parameters are based on supposed ones is not necessarily true because the distribution types of parameters are uncertain. Chebyshev inequalities can estimate the failure probability in spite of the distribution types of variables being unknown. Then, the upper bound probability of collapse failure of geomaterials behind a steel arch structure in tunnels is estimated based on the Chebyshev inequalities. Finally, the accuracy and efficiency of the proposed method are validated by the bootstrap method combined with the Akaike information criterion. It can be concluded that the probabilistic assessment of collapse failure of tunnels by Chebyshev inequalities is quick and conservative.

We present solutions for a combined probability density function computed as the weighted sum of individual probability density functions at the terminal branches of a logic tree. Logic trees are commonly utilized to quantify epistemic uncertainty in system analysis and can be computationally expensive when many branches and variables are included. The information provided herein can reduce those computational demands. We first derive closed-form expressions for the first two moments of the combined probability density function and subsequently derive expressions for the probability density function in cases where the mean value of the individuals is either normally or uniformly distributed and the variance is constant.