Journal of Computing in Civil Engineering

Bat algorithm (BA) is one of the well-established metaheuristic algorithms based on the echolocation characteristics of microbats. Literature studies demonstrate that BA provides good results in a wide range of optimization problems due to its simple structures, easy implementation, and effectiveness. However, the standard BA has two general shortcomings when applied to complex optimization problems. The first is the high probability of falling in local optima, and the second is insufficient population diversity, leading to unwanted premature convergence. This paper proposes an improved bat algorithm based on the Doppler effect (IBA-DE) to improve the performance of the standard BA. In the proposed IBA-DE, a new equation with some idealized rules is driven from the Doppler effect and adopted for updating the bat velocities in the algorithm body. The local search part of the standard BA is also improved via searching around the better bat individuals. This part is also equipped with Lévy flight to maintain the population diversity. The performance of IBA-DE is evaluated on three real-world engineering design problems. The results obtained by these problems show that IBA-DE is superior to the basic BA and other considered algorithms. For further investigation, IBA-DE is applied to the optimal design of four special truss structures. Comparing the optimization results found by the proposed algorithm with those of some other state-of-art metaheuristics indicates that the proposed IBA-DE can be an efficient optimizer for the optimal design of real-size structures.

Bridges are designed to last for more than 50 years and consume up to 50% of their life-cycle costs during their operation phase. Several inspections and assessment actions are executed during this period. Bridge and damage information must be gathered, digitized, and exchanged between different stakeholders. Currently, the inspection and assessment practices rely on paper-based data collection and exchange, which is time-consuming and error-prone, and leads to loss of information. Storing and exchanging damage and building information in a digital format may lower costs and errors during inspection and assessment and support future needs, for example, immediate simulations regarding performance assessment, automated maintenance planning, and mixed reality inspections. This study focused on the concept for modeling damage information to support bridge reviews and structural analysis. Starting from the definition of multiple use cases and related requirements, the data model for damage information is defined independently from the subsequent implementation. In the next step, the implementation via an established standard is explained. Functional tests aim to identify problems in the concept and implementation. To show the capability of the final model, two example use cases are illustrated: the inspection review of the entire bridge and a finite-element analysis of a single component. Main results are the definition of necessary damage data, an object-oriented damage model, which supports multiple use cases, and the implementation of the model in a standard. Furthermore, the tests have shown that the standard is suitable to deliver damage information; however, several software programs lack proper implementation of the standard.

Abstract

Advanced information technologies provide an opportunity to radically improve the business process of the architecture, engineering, construction, and operations (AECO) industry. The concept of interoperability, characterized as the representation of information and seamless exchange across interdisciplinary system domains, significantly influences the industry to leverage emerging digital technologies. However, just as representation and exchange contribute, an integrative method for organizing lifecycle information plays a vital role in achieving optimized interoperability. In a practical setting, breakdown structures have been used extensively to organize diverse types of project information for decades and are a vital tool for technical information integration. Nevertheless, their complex characteristics, user-specific ways of construction, and availability of heterogeneous classifications to refer to make integration less coordinated. To address this problem, this paper presents an integrative concept (named unified breakdown structure) and proves its viability. First, fundamental principles and characteristics of breakdown structures were discussed to examine their complexity. Then, an integrative concept was presented, and its viability was examined through an investigation of several cases incorporated with key assessment variables. The findings have revealed that the proposed approach is feasible enough to integrate lifecycle information and be used as a practical integration mechanism.

Abstract

This paper proposes an approach to integrating facility lifecycle information by introducing a separate set of breakdown structures (named primary breakdown structures). This set of structures will help keep a consistent facet (i.e., identifiers used to construct breakdown structure) across the remaining secondary breakdown structures, thus making information integration more coordinated and exchange smoother. The case analysis validates the proposed concept, showing it to be a useful approach in terms of maximizing the integration within and across disciplinary domains in the industry. For practitioners, the approach minimizes the effort and time spent translating and exchanging the complex range of facility information by keeping homogeneous facets to construct information. Furthermore, as broader data linkage through common data environments is becoming more common, the proposed approach provides a meaningful insight toward simplifying the complexity of constructing secondary breakdown structures in an integrative manner. Additionally, the introduced approach would help the industry efficiently utilize advanced technologies through the advent of optimized integration mechanisms.

Floods are the most common and damaging natural disaster worldwide in terms of both economic losses and human casualties. Currently, policymakers rely on data collected through labor-intensive and, consequently, expensive street-level surveys to assess flood risks. We propose a laborless and financially feasible alternative: a framework that can effectively and efficiently collect building attribute data without manual street surveys. By utilizing deep learning, the proposed framework analyzes Google Street View (GSV) images to estimate multiple attributes of buildings simultaneously—including foundation height, foundation type, building type, and number of stories—that are necessary for assessing flood risks. The proposed framework achieves a 0.177-m mean absolute error (MAE) for foundation height estimation and classification F1 scores of 77.96% for foundation type, 83.12% for building type, and 94.60% for building stories, and requires less than five days to predict the attributes of 0.8 million buildings in coastal Louisiana.

Advances in technology are leading to more and more devices integrating sensors capable of acquiring data quickly and with high accuracy. Point clouds are no exception. Therefore, there is increased research interest in the large amount of available light detection and ranging (LiDAR) data by point cloud classification using artificial intelligence. Nevertheless, point cloud labeling is a time-consuming task. Hence the amount of labeled data is still scarce. Data synthesis is gaining attention as an alternative to increase the volume of classified data. At the same time, the amount of Building Information Models (BIMs) provided by manufacturers on website databases is increasing. In line with these recent trends, this paper presents a deep-learning framework for classifying point cloud objects based on synthetic data sets created from BIM objects. The method starts by transforming BIM objects into point clouds deriving a data set consisting of 21 object classes characterized with various perturbation patterns. Then, the data set is split into four subsets to carry out the evaluation of synthetic data on the implemented flexible two-dimensional (2D) deep neural framework. In the latter, binary or greyscale images can be generated from point clouds by both orthographic or perspective projection to feed the network. Moreover, the surface variation feature was computed in order to aggregate more geometric information to images and to evaluate how it influences the object classification. The overall accuracy is over 85% in all tests when orthographic images are used. Also, the use of greyscale images representing surface variation improves performance in almost all tests although the computation of this feature may not be robust in point clouds with complex geometry or perturbations.

Extreme wind events can pick up loose and small objects on the ground, and once the objects become airborne, they negatively impact surrounding communities due to the collision impact. In this regard, jobsites and laydown yards that involve construction materials such as gravel piles and crushed rocks could be the main sources of potential windborne debris during extreme wind events. To analyze safety risk and predict the damage imposed by jobsite debris to nearby built environments, a new computer vision-based risk assessment based on geometric digital twins of jobsite debris is proposed for the reliability analysis on glazing systems of dwellings located on nearby jobsites. The impact of a gravel pile in a railroad jobsite on nearby buildings and residential environments was studied based on extreme wind event scenarios, and the failure risk of the building glazing system was computed. The risk associated with jobsite debris during extreme wind events and their impact on neighboring communities are analyzed through three computing modules: (1) satellite imagery-based terrain modeling to study 3D characteristics of the at-risk built environment; (2) analyzing visual data from Google Street View to assess the risk associated with glazing panels of dwellings in the communities; and (3) analyzing visual data from a jobsite to quantify the impact of jobsite debris, to associate its safety risk to neighboring communities. The proposed method can provide an immediate heads up for those who reside nearby jobsites, allowing to take required preemptive actions to protect their habitation against potential windborne debris. Practitioners will also be informed of such jobsite debris-related risk before extreme wind events to better secure their jobsites for the risk mitigation.

Offshore wind energy is a promising option for emission-free power generationbecause its potential can satisfy the entire US energy demand. However, deep offshore wind energy is mostly left untapped due to the high levelized cost of energy (LCOE) of floating offshore wind turbines (FOWTs). To address the challenge, individual blade pitch control (IPC) of each blade is necessary to reduce fatigue due to the nonlinear dynamics involving unbalanced nonstationary wind/wave loading. In this paper, a machine learning control (MLC) method is proposed that utilizes a genetic program to selectively evolve promising control law candidates from simulated FOWT sensor data. The proposed method utilizes unique MLC features including an efficient selection of design load cases (DLCs) to accelerate the evolution, DLC-based elitism to identify promising candidates, and the use of internal state feedback to learn effective controller properties across all the target DLCs. These features are used to reduce evaluation time on the complex nonlinear FOWT model. The feasibility of the method is demonstrated by reducing fatigue and ultimate loading by 41% under the representative DLCs provided by the Aerodynamic Turbines with Load Attenuation Systems (ATLAS) competition hosted by the Advanced Research Projects Agency–Energy (ARPA-E). Unlike the methods founded upon black-box models [e.g., artificial neural networks (ANNs)], the proposed method provides interpretable results; thus, it can contribute to learning important design features to guide the future controller design. For example, the best individual driven during the feasibility study exhibits a nonlinear proportional controller to a sigmoidlike platform pitch signal, which can be a starting point for more advanced IPC design for FOWTs. The proposed method will improve the cost-effectiveness of FOWTs and be applied to similar nonlinear control problems, such as unmanned aerial vehicle (UAV) control.

Guardrails are critical boundary infrastructures protecting against road departures and traffic collisions. The presence and condition information of in-service guardrails are essential for transportation agencies to perform necessary repair or replacement operations on time. Unfortunately, most current practices still rely on manual field surveys or windshield inspections that can be time-consuming, labor-intensive, and subjective. This study proposes an automated, network-level guardrail detection and tracking model based on 3D local features captured in mobile LiDAR data. The 3D local features, including corrugation, vertical profile, connectivity, and continuity of the guardrails, are introduced to extract guardrail status through four key sequential and corresponding steps, including Difference of Normals (DoN)-based segmentation, vertical profile-based filtering, guardrail-associated point re-population, and guardrail tracking. The proposed method is evaluated in two sections on State Route 113 and State Route 9 in Massachusetts. It shows promising performance with high precision rates of 95.6% and 95.5% and excellent length covering rates of 97.9% and 100.0%, respectively. The proposed method will provide a reliable and efficient means for transportation agencies to inspect and evaluate their critical guardrail infrastructure on a network level.

The advancement in sensors, robotics, and artificial intelligence has enabled a series of methods such as simultaneous localization and mapping (SLAM), semantic segmentation, and point cloud registration to help with the reality capture process. To completely investigate an unknown indoor space, obtaining a general spatial comprehension as well as detailed scene reconstruction for a digital twin model requires a deeper insight into the characteristics of different ranging sensors, as well as corresponding techniques to combine data from distinct systems. This paper discusses the necessity and workflow of utilizing two distinct types of scanning sensors, including depth camera and light detection and ranging sensor (LiDAR), paired with a quadrupedal ground robot to obtain spatial data of a large, complex indoor space. A digital twin model was built in real time with two SLAM methods and then consolidated with the geometric feature extraction methods of fast point feature histograms (FPFH) and fast global registration. Finally, the reconstructed scene was streamed to a HoloLens 2 headset to create an illusion of seeing through walls. Results showed that both the depth camera and LiDAR could handle a large space reality capture with both required coverage and fidelity with textural information. As a result, the proposed workflow and analytical pipeline provides a hierarchical data fusion strategy to integrate the advantages of distinct sensing methods and to carry out a complete indoor investigation. It also validates the feasibility of robot-assisted reality capture in larger spaces.

Abstract

Floor plan vectorization is an emerging research area in geographic information science and computer vision. However, automated recognition of building elements remains a challenge. This work proposes a method that combines the advantages of classical graphics with deep learning. Specifically, a morphological template is introduced to optimize topological relations, enhance completeness, and suppress conflicts. Bezier curves are utilized to represent irregularity contributing to improving visual effects and experimental accuracy. Thus, the proposed method can be directly practiced to boost performance and correct pseudo-samples in self-training. Experiments demonstrate that the proposed method achieves a considerable improvement in CVC-FP and R2V benchmarks. Additionally, our approach outputs instances with consistent topology, enabling direct modeling into Industry Foundation Classes (IFC) or City Geography Markup Language (CityGML). Hopefully, this work can serve as a new baseline for further study.

Practical Applications

Automatic floor plan recognition has many potential applications. From one perspective, structural elements (walls, doors, and windows) are the most typical objects as they illustrate the primary layout of a building and convey essential information to deploy other components. Identifying structural elements is particularly crucial because it provides design, investigation, and assessment representations. From another perspective, retrieving the room types plays a critical role given that it offers the semantics of a scene. Identifying these elements can avoid tedious secondary measurements and even recover the structure from an advertising paper or poster, facilitating the subsequent application deployment. The applications in Building Information Modeling reconstruction (such as 3D models from 2D maps, architectural arrangement, structural redesign, virtual reality, indoor navigation and modeling, 3D reconstruction from interior photographs, bearing structure analysis, renovation, refurbishment, plan illustration interpretation, apartment price estimation, and accessibility for visually impaired people) could provide substantial support for the smart city.

Several studies have used surrogate safety measures obtained from microsimulation packages, such as VISSIM, for safety assessments. However, this approach has shortcomings: (1) microsimulation models are developed considering specific rules that tend to avoid collisions; and (2) existing models do not realistically model road users’ behavior and collision avoidance strategies. Moreover, the majority of these models rely on the single-agent modeling assumption (i.e., the remaining agents are considered components of a fixed and stationary environment). Nevertheless, this framework is not realistic, which can limit the models’ representation of the real world. This study used a Markov Game (MG) for modeling concurrent road users’ behavior and evasive actions in near misses. Unlike the conventional game-theoretic approach that considers single-time-step modeling, the MG framework models the sequences of road user decisions. In this framework, road users are modeled as rational agents that aim to maximize their own utility functions by taking rational actions. Road user utility functions are recovered from examples of conflict trajectories using a multiagent adversarial inverse reinforcement learning (MAAIRL) framework. In this study, trajectories from conflicts between motorcyclists and pedestrians in Shanghai, China, were used. Road user policies and collision avoidance strategies in near misses were determined with multiagent actor–critic deep reinforcement learning. A multiagent simulation platform was implemented to emulate pedestrian and motorcyclist trajectories. The results demonstrated that the multiagent model outperformed a Gaussian process inverse reinforcement learning single-agent model in predicting road user trajectories and their evasive actions. The MAAIRL model predicted the interactions’ postencroachment time with high accuracy. Moreover, unlike the single-agent framework, the recovered multiagent reward function captured the equilibrium concept in road user interactions. The multiagent model enables greater understanding of road users’ behavior in conflict interactions and captures the nonstationariness in the environment.

Abstract

Recent infrastructure inspection has used deep-learning models to enhance and augment typical inspection tasks such as detecting and quantifying damage. One of the issues with this trend is that deep-learning models typically require a significant amount of data. In a data domain such as structural inspection, publicly accessible data are difficult to find, making the advancement of this research slower. Therefore, we set out to acquire bridge inspection data by selectively extracting candidate images from hundreds of thousands of bridge inspection reports from the Virginia Department of Transportation. Using this rich source of diverse data, we refined our collected data to develop four high-quality, easily extendable, publicly accessible datasets, tested with state-of-the-art models to support typical bridge inspection tasks. The four datasets: labeled cracks in the wild, 3,817 image sets of semantically segmented concrete cracks taken from diverse scenery; 3,817 image sets of semantically segmented structural inspection materials (concrete, steel, metal decking); 440 images of finely annotated steel corrosion condition state (good, fair, poor, severe); and 1,470 images of fatigue-prone structural steel bridge details (bearings, gusset plates, cover plate terminations, and out-of-plane stiffeners) for object detection. To ensure the extendibility of the datasets, the authors have proposed annotation guidelines to maintain consistent growth through annotation collaboration. Researchers can use these trained models and data for auxiliary inspection tasks such as damage detection, damage forecasting, automatic report generation, and, coupled with the assistance of unmanned aerial systems, for autonomous flight path planning and object avoidance. The procedures, concepts, and repositories provided in this paper will help to set a course for the advancement of better detection models using high-quality accessible and extendable datasets.

Practical Applications

Researchers can use these datasets and trained models for auxiliary inspection tasks such as damage detection, damage forecasting, automatic report generation, and, coupled with the assistance of unmanned aerial systems, for autonomous flight path planning and object avoidance. The models may also be used as starting points for researchers to extend or begin their own use-case datasets for this domain or similar domain of data. The authors have presented three main reasons for including annotations guidelines for any public dataset: to establish consistent annotations for the training and testing data; to clearly present the expected model prediction; for guiding future collaborators. Annotation guidelines are a practical tool for any researcher procuring a dataset for supervised machine learning.