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

The purpose of this paper is to address the “what,” “why,” and “how” questions posed by engineers who are not familiar with geotechnical reliability and have not kept abreast of recent rapid developments in this field. Geotechnical reliability can be broadly defined as a methodology that enhances decision making at different life-cycle stages covering design, construction, operation and maintenance, retrofit, and decommission/reuse by exploiting the richer characterization of data using probabilistic models. Besides engineered systems, it also covers the risk assessment and management of geohazards such as earthquakes and landslides. Various application areas related to the design and construction stages of engineered systems can be put in context in the form of an uncertainty-informed Burland Triangle. Among these application areas, the estimation of a characteristic value, load and resistance factor design (LRFD), resistance factor calibration for simplified reliability-based design (RBD), and first-order second-moment (FOSM) reliability analysis do not need in-depth knowledge/expertise in reliability and a significant amount of information exists to support these applications in practice. This paper argues for their adoption because they will nudge a mindset shift to be more responsive to data. Data infrastructure is now considered to be as important as physical infrastructure. The concern that there are insufficient data for probabilistic analysis has also been largely and comprehensively resolved by recent advances in Bayesian machine learning methods that can deal with MUSIC-3X (Multivariate, Uncertain and Unique, Sparse, Incomplete, and potentially Corrupted with “3X” denoting 3D spatial variability) site data directly. Sparsity (insufficient data) is only one out of six attributes in a real-world MUSIC-3X data set. Geotechnical reliability is now pictured as one important step toward digital transformation and engaging complex new challenges posed by climate change and resilience engineering. This paper urges the geotechnical engineering profession to lay aside its questions on quantity, quality, and/or other “ugly” attributes and offer our data an opportunity to speak for itself. There is prima facie evidence to warrant a thorough exploration of data-centric geotechnics.

Prioritization is considered a core concept in cost overrun analysis. This study first aimed at proposing a new approach for the calculation of total risk magnitude (TRM) and, second, introduced the application of Pareto analysis in the prioritization of cost overrun causes. A case study is delineated to explain the applicability of the aforementioned approaches. Thirty-eight unique causes of cost overrun in urban-related construction projects were identified and a survey was conducted among construction professionals in Iran. The TRM for each cost overrun cause is calculated using single-attribute (SA), double-attribute (DA), and multiple-attribute (MA) approaches. Henceforth, based on calculated TRM, the causes are prioritized using the relative importance index (RII) and Pareto analysis. A comparative analysis revealed that not only could Pareto-based prioritization be effectively employed for prioritization, but also, the proposed MA approach for TRM calculation optimizes the efficiency of the most influential cause identification.

Smart streetlamps, which can automatically adjust luminance by sensing information about passing vehicles, are gradually becoming an integral part of smart cities. However, when choosing a matching scheme between intersection lighting and vehicle speed, the matching performance and traffic safety can be compromised if the traffic risks caused by various conflicting events are not fully considered. Because it is difficult for drivers to spot pedestrians, cyclists, and motorcyclists at night, their accident mortality is relatively high. We designed conflicting events among pedestrians, cyclists, motorcyclists, and vehicles, and addressed this problem in three stages. First, the reserved reaction time (RRT), corresponding to the moment when a driver applies brakes in response to an emergency, was used as a quantitative index of the nighttime risk at intersections. Second, we selected the most dangerous conflict event using RRT. Finally, under the most dangerous conflicting conditions, the optimal lighting and vehicle speed matching scheme was designed with minimum illumination as the optimization objective and considering an RRT greater than or equal to zero as the constraint. As the research object, we considered a two-way/two-lane and two-way/four-lane major–minor intersection. We then constructed an indoor simulation test platform based on the road and lighting parameters of the actual intersection, and validated the matching optimization method of intersection lighting and vehicle speed in the simulation environment. The results indicate that when the average road surface illuminance (ARSI) is less than 3.3 lx, a cyclist approaching an intersection from the right side is in the most hazardous situation. If the ARSI is greater than 3.3 lx, the most dangerous situation is a pedestrian approaching from the right side of an intersection.

Building industrialization has great significance for improving the efficiency of construction production, achieving the goal of energy saving and emission reduction, and promoting sustainable development. On the other hand, numerous thick wall structures with bulky dimensions and complex connection constructions have been applied into special engineering such as islands, tunnels, and deep sea projects. The precast structure is an open and complex system with a certain level of systematic risk for the whole life cycle. Moreover, structural mechanics and reliability, construction uncertainty, and resilience of precast thick wall structures still need to be revealed. For promoting the application of building industrialization to thick wall structures, nonlinear finite-element analyses on two types of large-thickness precast shear walls have been carried out based on original structural design specifications. The reinforcement connection configurations and the shear performance in terms of the load–displacement curve, lateral stiffness, ductility, energy dissipation capacity of shear walls with cast-in-place (CIP) belts, or grouting sleeves have been analyzed in detail. Numerical results show that the shear performance of two large-thickness precast shear walls is comparable to that of CIP walls. The relative error of the peak load is less than 10% for three specimens in flexural failure mode. The yield load of shear walls is relatively large and the stage between yield and failure is satisfactory. The shear performance of shear walls decreases slowly after reaching its peak value. Meanwhile, we have conducted the qualitative analysis of structural reliability for large-thickness precast shear walls in component production, quality inspection, wet work, and shear performance.

Site characterization is one of the most crucial steps for decision making in geotechnical engineering and to the fullest extent possible should be conducted based on objective data. The current reliance on engineering judgment to interpret data directly cannot exploit the rapid growth of data, machine learning, and other digital technologies. Data-driven site characterization (DDSC) has received much attention in an emerging field called data-centric geotechnics, because a knowledge of the ground is fundamental to geotechnical engineering. As a result, many DDSC methods have been developed recently. Differences and similarities between DDSC methods, however, have not been well studied in terms of methodological and application aspects. This paper proposes a comparison between three emerging DDSC methods from these methodological and application perspectives: (1) geotechnical lasso (Glasso), (2) geotechnical lasso with basis-functions (Glasso-BFs), and (3) Gaussian process regression (GPR). From a methodological perspective, this paper presents a unified Bayesian framework to derive these DDSC methods, in order to shed light on the methodological similarities and differences. From the application perspective, the prediction accuracy for the validation dataset and runtime cost of these three DDSC methods were compared through benchmarking. The differences in performance can be better understood within the unified framework. This paper further proposes a new benchmark involving complex intermixing of soil types, to test the three methods under more realistic and challenging field conditions, although the training and validation datasets remain synthetic.

This study aimed to solve the problem of setting and optimizing the driving parameters for shield tunneling in sand–clay composite strata. First, the strata composite ratio parameter ($\xi$) and the unit thrust ($\overline{F}$) and unit torque ($\overline{T}$) of the shield machine driving parameters were defined. The quantitative correlations of $\overline{F}$ and $\overline{T}$ with varying $\xi$ were obtained. The results can help set the driving parameters of the shield machine. Second, the data point distribution space of $\overline{F}-\overline{T}$ was established and analyzed by mathematical statistics. The distribution area of the data recorded was classified as an abnormal Zone II when the cutterhead of the shield machine is obstructed by mud cakes, ground settlement occurs, and so on. The distribution area of the data recorded was classified as abnormal Zone III when the cutterhead of the shield machine encounters boulders, excessive jacking force leads to ground uplift, and so on. The remaining area was defined as normal Zone I. The developed zoning criterion can make a quality assessment for the shield driving status. Finally, the clustering theory was used to divide Zone I into six subregions. A composite strata identification method for shield tunneling was established. Combined with the research results, a systematic evaluation method of shield machine driving state was summarized. Through field verification, the evaluation method has guidance for the setting of shield machine driving parameters and the evaluation of driving state.

Data-driven site characterization (DDSC) aids geotechnical engineering by inferring and mapping soil parameters of the subsurface domain. In practice, the limited availability of site investigation data may hinder the performance of traditional machine learning methods and implies significant uncertainty in the predictions, which is typically not quantified. In this study, a framework for Bayesian site characterization (BaySiC) is applied on a benchmark example. Adopting Bayesian statistics enables the framework to deal with small training data sets and allows for coherent quantification of uncertainty, which is valuable to engineering practice for assessing the reliability and the determining characteristic values. BaySiC uses site investigation data to infer statistical estimators of cone penetration test (CPT) parameters and their dependence, as well as to learn spatial correlations. Consecutively, it generates a three-dimensional (3D) map of the subsurface by predicting the CPT parameter values and classifying the material type over the soil domain. For the benchmark example, the study formulated two models within the BaySiC framework and demonstrated their conduct in several cases of varying complexity. Eventually, the performance of the models was evaluated and compared in both deterministic and probabilistic terms. One of the models proved highly effective in predicting the material type at new locations of the subsurface domain, whereas the other provided accurate mapping of the CPT parameters even in complex stratigraphic cases. Also, investigating and comparing the results of the models led to insights regarding the effectiveness of their formulation. Moreover, the paper used hypothesis testing as a means of assessing the predictive power of the model independently from the validation data set. Stemming from the benchmark example, the paper draws conclusions that are meaningful to geotechnical engineering and decision-making.

A fair allocation of risk is essential for successful projects involving public–private partnerships (PPPs). The participants to the risk allocation determine whether to take risks and how much of the share of the risks to assume according to their preference for and ability to manage risk. Owing to the different preferences of the public sector and private sector in the context of risk allocation, the perspective of local governance demands that the participants of risk allocation express their cognitive demands and interests through negotiation and dialog to arrive at a consensus on the choice of the scheme to use for risk allocation. To examine how risk-taking behaviors evolve in PPPs, this study develops an evolutionary game model in which the private and public sectors initiate the negotiation. This model is used to analyze the evolutionary stability of each player’s strategy, discuss the influence of various factors on the choices of strategy of the two parties, and to examine the stability of the point of equilibrium of the game system for the participants of risk allocation. The results provide decision-making support in negotiating appropriate concessions in PPP projects to reduce the incidences of renegotiation and incentives for either party to seek to renegotiate.

With the rapid development of computing and digital technologies recently, three-dimensional (3D) subsurface models for accurate site characterization have received increasing attention, for example, with various data-driven methods developed for 3D subsurface modeling. This leads to a need for validating the 3D modeling results obtained from each method and comparing the performance of different methods in a fair and consistent manner. To address this need, a benchmarking study, which is often used in machine learning (ML), is presented in this study to compare the performance of different 3D subsurface modeling methods in four aspects, including accuracy, uncertainty, robustness, and computational efficiency. A suite of performance metrics is proposed for the four aspects above. Multiple sets of real cone penetration test (CPT) data are compiled in the benchmarking study for quantifying performance of 3D modeling methods using sparse measurements as input, a typical scenario in geotechnical practice. The benchmarking study is illustrated using an in-house software package called Analytics of Sparse Spatial Data based on Bayesian compressive sampling/sensing (ASSD-BCS), which can directly generate high-resolution 3D random field samples (RFSs) from sparse measurements. The evaluation results show that ASSD-BCS provides accurate estimates with quantified uncertainty from sparse measurements. In addition, ASSD-BCS exhibits remarkably high computational efficiency and performs robustly under different benchmarking cases.

This paper proposes a probabilistic machine learning methodology to predict travel time and its variability for trips between locations in New York City. First, a hierarchical Bayesian generalized linear regression model was trained to estimate predictive distribution of the trip’s unit-distance travel time conditional on trip features. This intermediate step isolates the effect of trip distance to capture the impact of traffic conditions and other covariates on trip travel time. Then, trip travel time and its variability were obtained given the predictive distribution of normalized travel time and the trip’s fastest path distance. Specifically, a Bayesian regression model with varying coefficients was fitted to capture the effect of temporal variations of traffic conditions and spatial effect of geographic regions on model parameters and in turn on trip travel time. The model was trained using New York City ambulance and taxi trip data and its performance was verified with benchmarks methods. Although it shows promising performance, the proposed methodological framework estimates the posterior distributions of the model’s interpretable parameters and eventually estimates the predictive distributions for trip travel times. The estimated distributions are required for uncertainty quantification to make more reliable decisions in traffic management and traffic control strategies and achieve proactive enhancement in the transportation systems. The proposed approach emphasizes the potential use of origin-destination trip data for travel time prediction, quantifying its spatiotemporal variations, and capturing the predictive uncertainties, which are central to operational transportation systems’ design and performance evaluation.

A mechanics-based methodology called the inverse unit load method (IULM) is proposed for the distributed bending stiffness estimation of girder bridges utilizing in situ deflection responses. This inverse problem has important implications for bridge health monitoring and rapid bridge detection procedures. The IULM formulation is derived from a regularized least-squares function that employs the unit load method and Betti’s law as its underlying design theory. An anomaly index is defined to optimize the IULM discretization, which guarantees the stability and accuracy of the solution results. The present estimation model only requires real bridge deflection influence lines (DILs) as input; thus, it is generally suitable to estimate the bending stiffness of girder bridges with different boundary conditions under healthy or damaged conditions. Numerical validation cases for a three-span continuous bridge with different types of health states have been performed. The effects of the discretization strategy and measurement noise have been assessed with respect to the accuracy of the IULM solution. The numerical results demonstrate the excellent predictive capability, practical utility, and robustness of the IULM methodology. The present approach has promising potential in full-field bending stiffness estimation of girder bridges.

The probability distribution of typhoon key parameters is commonly incorporated into typhoon models to estimate the typhoon-induced wind speeds associated with certain return periods in typhoon-prone regions. In most studies that focus on the typhoon wind hazards of the southeast coastline of China, the typhoon key parameters are assumed to be independent. This paper develops a copula-based joint probability distribution for the typhoon key parameters to investigate its potential influence on the typhoon wind hazard on the southeast coastline of China. To this end, the best track typhoon data from the China meteorological administration are used to extract the key parameters of the typhoon. The analyses show that the observed correlation coefficients among the parameters could be larger than 0.4 at some locations on the considered coastline. The C-vine copula is then employed to establish the joint probabilistic model of these key parameters. Comparison between the observed and modeled joint probability distributions suggests the adequacy of the copula method–based probability distribution model. Then, a local track model and a typhoon wind field model are assembled to simulate the history of the typhoon-induced surface wind given the typhoon key parameters. Finally, Monte Carlo simulation is adopted to estimate the wind speed associated with 50- and 100-year return periods. Results show that neglecting the correlation among the typhoon key parameters could cause a relative difference of up to 7% at some locations on the coastline.

Analyzing the importance grade of each metro station is the premise and foundation to ensure the safe and efficient operation of the metro network. This paper explores the identification method of critical stations in the metro network on the basis of considering the influence of passenger flow. Eight indicators were selected from both the static and dynamic aspects to construct the critical station identification index system considering the impact of passenger flow and cascading failure. Then, the triangular fuzzy analytic hierarchy process (TFAHP) method and critical importance through intercriteria correlation (CRITIC) method were used to calculate subjective and objective weight of all indicators, and the combined weighting method based on game theory was applied to calculate the comprehensive weight. The improved technique for order of preference by similarity to ideal solution (TOPSIS) method was developed to calculate the station importance and rank all stations according to their importance. Finally, the Xi’an metro network by mid-2021 was taken as the case study. The results show that there are 21 critical stations in the Xi’an metro network in three time periods (i.e., morning rush hours, evening rush hours, and nonrush hours) on weekdays and weekends, respectively, and those critical stations are divided into three levels. High centrality, high passenger flow intensity, and strong destructive ability are three important characteristics of critical stations in the metro network, and different critical stations have different importance characteristics. The Xi’an metro operation management authority needs to conduct a classification management scheme for those critical stations. The conclusions can provide the basis for the metro operation management authority to formulate accurate station management scheme, so as to increase the robustness of the metro network.

Evaluating the system reliability of layered soil slopes is a challenging issue because multiple failure modes may be included along the slip surfaces, which makes the overall failure probability greater than any individual slip surface. In this paper, an efficient system reliability analysis concerning the layered soil slopes is conducted based on the sequential compounding method (SCM) that has the ability to compound multiple failure events into an equivalent event sequentially. First, the first order reliability method (FORM) is employed to quantify initial reliability indices and correlation coefficients among these failure modes. Subsequently, the SCM is used to calculate the equivalent reliability indices and correlation coefficients until the multiple failure events are reduced to a compound event, and then the system reliability of the slope is obtained accordingly. The application of the approach to probabilistic evaluation of layered slopes is illustrated by two typical examples, and the correctness is verified by a Monte Carlo simulation (MCS). The results show that the SCM can deliver accurate system failure probability and greatly improve the computational efficiency compared with the MCS, which is an advantageous and promising strategy in evaluating the system reliability of layered soil slopes.

This paper investigates the earthquake-induced failure of high-rise steel moment-resistant frame buildings subjected to sequential long-duration ground motions. A fiber-element model was developed with considering the strength and stiffness deterioration caused by local buckling and low-cycle fatigue fracture. The modeling method taking account of strength and stiffness deterioration was validated by component test results of both column bases and beam-column joints. At the structural level, simulation results including both global and local behavior were compared with the shaking table test results of a high-rise steel building in Japan, which showed a reasonable prediction of its seismic failure up to collapse.

Probabilistic seismic hazard analysis (PSHA) is a tool to quantify the annual rate of exceedance of seismic intensity measure at a location of interest. The present study investigates the uncertainty embedded in three PSHA parameters: magnitude of earthquake, source-to-site distance, and corresponding ground shaking. Although previous researchers have worked to quantify the uncertainty in these hazard parameters, the source-to-site distance uncertainty in PSHA has not been explicitly addressed to acquire more accurate and reliable seismic hazard estimates. This study develops an uncertainty quantification framework for PSHA considering source-to-site distance uncertainty, including interdependency between hazard parameters. To capture this uncertainty, an approach of considering grids in the area seismic source zone is adopted using a geographic information system (GIS) and implementing Latin hypercube sampling (LHS) to quantify the uncertainty of the occurrence of an earthquake at any distance from the site. The interaction of magnitude and source-to-site distance has been investigated using copula theory. As a part of practical implementation, PSHA of Silchar city in Northeast India is taken into consideration due to its tectonic features and high seismicity. This work applies two processes to quantify uncertainty: (1) considering the center point of a grid, and (2) stratified sampling of points within a grid by LHS. The results reveal that on applying LHS (with uncorrelated random events), the hazard increases due to discretization of the area sources, and robust consideration of uncertainty. While, in case of correlated random events, the hazard levels are relatively low. The results are being subjected to verification and are compared with the previous studies to ensure fulfilling specifications and its intended purpose. Structural designers and decision makers can utilize the developed PSHA framework to plan and build new projects and to evaluate the seismic risk of existing infrastructure in a city.

Enterprise standards for business processes such as ISO 9001:2015 are used worldwide in the energy sector. Use of these standards improves business processes and improves organizational quality management. There are however trade-offs to implementation in the form of organizational disruption, startup costs, and institutional resistance to change. Increasing quality management enables improved response to disruptions. Accelerating the implementation of these standards requires employee engagement to identify the benefits to agility and barriers so they can be overcome. Given the scale and regularity of disruptions to the energy sector in past decades, it is urgent that these practices be applied in the industry at a supply chain scale, which will make it an appropriate example to survey employees within the petroleum industry. This example is focused on evaluating the risks and opportunities for the implementation of ISO standards in the oil and gas industries. An employee survey is used as the basis for these evaluations. Based on hypothesis testing, results indicate that lack of motivation among employees to follow the standards and lack of adequate training contribute the most to the risks. Additionally, several opportunities seem to contribute to the reduction of these risks, per the survey takers’ opinions. In a first step, the various factors are compared and studied from the perspective of employees. A survey of employees is prepared from the literature and a pilot study. The results in part illustrate how particular oil and gas companies are pursuing ISO certifications. The survey helps in identifying risks and preparing a roadmap of the processes that can be used by industries and organizations to improve quality practices. This paper should be of wide interest to industries and organizations that are implementing ISO 9001:2015 and other enterprise standards in the energy sector.