In recent years, the frequency and severity of catastrophic events triggered by natural hazards have increased. Meanwhile, man-made hazards, such as terrorist attacks, and their impacts on infrastructure systems have gained increasing attention. These hazards (both natural and man-made) can cause catastrophic physical damage to transportation infrastructure systems that are essential to the wellbeing of the society. Moreover, the direct economic losses (e.g., physical damage to infrastructure) diffuse and expand continually through the disruption of economic activities between different regions and industries, resulting in enormous and complex indirect losses. A comprehensive investigation of total losses, including direct and indirect losses, requires the use of economic impact analysis models. However, most of the economic impact analysis methods and models introduced in the existing literature fail to incorporate the spatially distributed and networked nature of transportation infrastructures. To achieve a comprehensive and a realistic understanding of the economic impacts caused by the disturbances to the transportation infrastructure, the spatial distribution and the networked nature of transportation systems has to be accounted for, and realistic and locally relevant hazard scenarios must be incorporated into the economic analyses. This paper first provides a detailed account of the status-quo in economic modeling associated with impact analysis of transportation disturbances to identify the gaps in this domain. Next, focusing on the commuting related economic impacts of transportation disturbances as an example, the paper introduces a multidisciplinary framework designed to demonstrate an understanding on how to address the gaps. Preliminary results from a Los Angeles case study are presented.
The frequent natural hazards and their significant impacts in recent years have repeatedly highlighted the vulnerability of urban infrastructure systems and their lack of system resilience. The ever increasing interdependencies between urban infrastructure systems have further complicated the situation, causing considerable risks of failure propagation. Considering that the properties of urban infrastructure systems including their level of resilience are mainly determined at the design stage, the authors aim to propose a Quality Function Deployment (QFD) based conceptual framework for designing resilient urban infrastructure system of systems (SoS). As a preliminary effort towards this goal, this paper mainly focuses on developing the first matrix in the QFD based framework. Steps to identify resilience criteria and principles are presented, the questionnaire-based method to determine the main body of the first matrix is described, and the approach to work out practical design schemes is explained. Lastly, this paper summarizes the main strength of the proposed framework as well as its limitations, and discusses directions for future research.
Adaptive structures are sensed and actuated to modify internal forces and shape to maintain optimal performance in response to loads. The use of large shape changes as a structural adaptation strategy to counteract the effect of loads has been investigated previously. When large shape changes are employed, structures are designed to change shape as the load changes thus giving the opportunity to homogenize stresses. In this way, the design is not governed by peak loads that occur very rarely. Simulations have shown a significant amount of embodied energy can be reduced with respect to optimized active structures limited to small shape changes and with respect to passive structures. However, in these previous studies, the actuator layout was assigned a-priori. This paper presents a new method to search for an actuator layout that is optimum to counteract the effect of loads via large shape changes. The objective is to design the actuation system allowing the structure to ‘morph’ into shapes optimized to maximize material utilization for each load case. A combination of simulated annealing and the nonlinear force method is proposed to meet both the actuator placement problem and to determine appropriate actuator commands. A heuristic for near-neighbor generation based on the actuator control efficacy is employed to explore effectively the large search space. Case studies show the proposed method converges to the global optimum for simple configurations and generally produces actuator layouts enabling shape control even with a low number of actuators.
Construction, maintenance, and renewal of underground utilities ( e. g. gas, water and sewer pipes, and electrical and telecommunication cables) impose a large lifecycle cost on utility companies and the citizens of urban areas. Integrating all the utilities in a Multi-purpose Utility Tunnel ( MUT) can reduce post-construction costs of accessibility, inspection, maintenance, protection, and social costs. Although the high construction cost is an obstacle for promoting MUTs, post-construction cost savings make MUTs an economic alternative solution considering the total lifecycle costs. On the other hand, sharing the lifecycle cost of MUTs is another issue, particularly for the high initial investment in construction. Public-Private Partnership ( PPP) is a promising approach for sharing the lifecycle cost of MUTs among public and private investors and facilitating the development of MUT projects. This paper aims to develop a model for PPP financing and cost-sharing of MUT projects using game theory. The PPP investors need to investigate different scenarios for lifecycle cost-sharing and choosing the best strategy considering their financial situation. The proposed model is based on three steps: ( a) selection of MUT cost allocation method for MUT lifecycle phases, ( b) cost adjustment based on risk and benefit cost factors, ( c) game model of MUT cost-sharing based on an entity covering a part of the MUT construction cost using loans/bonds with a penalty that is used as a reward to the other entity. The results show that this model can improve the fairness of MUT cost allocation and be used as an analytical tool for the MUT project stakeholders to choose appropriate financing strategy based on anticipated lifecycle benefits and costs.
When using Building Information Modeling (BIM), it is expected to increase interorganizational collaboration in the construction industry. To reach this goal, it is necessary to use open standards for model exchange such as the Industry Foundation Classes (IFC). The non-profit organization buildingSMART defines this standard. IFC provides object types and attributes, which describe properties of objects. The IFC standard do not address possible values for attributes; but beside coordinated names for object types and attributes, coordinated values for attributes are required. This is necessary for the evaluation of digital models for different purposes. Several approaches in the fields of ontology and classification systems address this issue. However, at present time there is no clear distinction between prerequisites that modeled objects have to fulfill and products, which satisfy these requirements. Suppliers or construction companies at the interface between planning and construction propose products for the modelled requirements of the planning process. This paper describes an approach that bases on a dictionary, which provides standardized object types, attributes and values. A project database and a database with products, called supplier database, use this dictionary as a basis. Resulting from the modeling phase, the project database is ultimately mapped with the supplier database to propose products that fulfill the requirements. The approach presented in this paper can be regarded as a next step in using a standard that already exists: the buildingSMART Data Dictionary.
Currently, visual inspection techniques, especially closed-circuit television (CCTV), are commonly utilized for sewer pipe inspection. Computer vision techniques are applied for automated interpretation of CCTV images to identify pipe defects. However, conventional computer vision techniques require complex handcrafted feature extraction and large amount of image pre-processing. In this study, a deep learning based approach is developed for sewer pipe defect detection using faster region-based convolutional neural network (faster R-CNN). 3000 images were collected from CCTV inspection videos of sewer pipes, among which 85% were used for training and validation and 15% are for testing. The detection model was trained and evaluated in terms of mean average precision (mAP), missing rate, detection speed and training time. The proposed approach is demonstrated to be applicable for detecting sewer pipe defects accurately with a high mAP and low missing rate. In addition, the initial model was improved by investigating the influence of dataset size, initialization network type and training mode, as well as network hyper-parameters on model performance. The improved model achieved a mAP of 83% and fast detection speed. This study has the potential for addressing similar object detection problems in the architecture, engineering and construction (AEC) industry and provides references when designing the deep learning models.
Urbanization and aging of society are two converging trends of current demographic changes. The intensified human activities associate with the formation and dynamic of Urban Heat Island (UHI) which is harmful to health. Looking into the correlation between UHI effect and the land surface coverage of everyday spaces is curtail and significant. Aided by spatial analysis and spatial statistic functions based on Geographical Information System (GIS) and data extraction and processing methods enabled by the Remote Sensing (RS) platform, this paper detected the distribution of land surface temperature and traced its changes alongside dynamics of land surface coverage in the selected typical areas of Hong Kong. Findings of this paper will be significant for Hong Kong planners, architects and housing officials to consider when deciding on the next step of Hong Kong urban planning and housing development.
The application of building information modeling ( BIM) in early design phases requires the support of different levels of detail ( LOD). This allows scaling to be supported as an important activity of designing. Furthermore, to achieve well-performing solutions in terms of energy efficiency, it is necessary to consider energy performance in early design stages. Therefore, this paper presents a multiLOD modeling approach for the early phases of building design that integrates energy performance prediction based on component-based machine learning ( ML) using artificial neural networks ( ANN). A model structure with three adaptive LOD definitions is proposed to support the design process by a digital model that supports flexible scaling back and forth. By linking the ML models to the elements in this structure, components are formed that support quick and flexible modeling and energy performance prediction in the early building design process. The transformation rules flexibly link the ML components to all LOD. This approach was illustrated and validated by a test case with a medium-sized office building. The early design states of the case were reconstructed for the application of the method. For validation purposes, the results of the ML predictions for 60 different design configurations were compared to those of a conventional parametric full-detail simulation model. This comparison showed that the average error was no higher than 3.8% for heating and 3.5% for cooling.
A number of studies have assessed the energy consumed and carbon dioxide emitted by construction machinery during earthwork operations. However, little attention has been paid to predicting these variables during planning phases of such operations, which could help efforts to identify the best options for minimizing environmental impacts. Excavators are widely used in earthwork operations and consume considerable amounts of fuel, thereby generating large quantities of carbon dioxide. Therefore, rigorous evaluation of the energy consumption and emissions of different excavators during planning stages of project, based on characteristics of the excavators and projects, would facilitate selection of optimal excavators for specific projects, thereby reducing associated environmental impacts. Here we describe use of artificial neural networks (ANNs), developed using data from Caterpillar's handbook, to model the energy consumption and CO2 emissions of different excavators per unit volume of earth handled. We also report a sensitivity analysis conducted to determine effects of key parameters (utilization rate, digging depth, cycle time, bucket payload, horsepower, load factor, and hauler capacity) on excavators' energy consumption and CO2 emissions. Our analysis shows that environmental impacts of excavators can be most significantly reduced by improving their utilization rates and/ or cycle times, and reducing their engine load factor. We believe our ANN models can potentially improve estimates of energy consumption and CO2 emissions by excavators. Their use in planning stages of earthworks projects could help planners make informed decisions about optimal excavator(s) to use, and contractors to evaluate environmental impacts of their activities. Finally, we describe a case study, based on a road construction project in Sweden, in which we use empirical data on the quantities and nature of the materials to be excavated, to estimate the environmental impact of using different excavators for the project.
With the advancements in 3D-imaging technology, new quality control methods can be developed and applied in the construction industry. Strictly stipulated tolerances put contractors under pressure to assure a high level of quality and accuracy. Thus, tolerance control has become imperative in the construction industry. Industrial assemblies are usually fabricated through stepwise processes which are being automated at a rapid pace. New components are added to partially-built assemblies composed of one or more components. Consequently, for each step, geometrical control can be applied to prevent any ongoing deviation from being propagated throughout the completion of the assembly. To do so, an intermediate 3D-model is derived from the initial design CAD file. Utilizing the Minimum Required Model (MRM) as opposed to the whole model enables easier quality control when using a 3D-imaging device and makes the control of the orientation and alignment faster, more accurate, easier and safer. This paper proposes a novel method to derive the MRM for stepwise quality control of a construction assembly using 3D-imaging technologies. Developed for use in the piping industry, the method uses the complete 3D-model and the Piping Component File (PCF), which includes the overall assembly in a generic text file format, to reduce the boundaries of the model based on the step currently being assembled. This process saves time by reducing the volume of the required scan to be processed and is more accurate than using manual tools for measuring the alignment. The algorithm for the derivation of the MRM is evaluated on 95 different scenarios and the results are compared in terms of reduced level of complexity and reduced principal length.
In situations where rapid decisions are required or a large number of design alternatives is to be explored, numerical predictions of construction processes have to be performed in near real-time. For the design assessment of complex engineering problems such as mechanised tunnelling, simple numerical and analytical models are not able to reproduce all complex 3D interactions. To overcome this problem, in this paper a novel concept for on-demand design assessment for mechanized tunnelling using simulation-based meta models is proposed. This concept includes: (i) the generation of enhanced simulation-based meta models; (ii) real-time meta model-based design assessment in the design tool, and; (iii) the implementation within a unified numerical and information modelling platform called SATBIM. The capabilities of this concept are demonstrated through an example for the evaluation of tunnel alignment design and the assessment of the impact of tunnelling on existing infrastructure. Moreover, meta models are used for fast forward calculation in sensitivity analyses for the evaluation of the importance of model parameters. The concept proved its efficiency by assessing the design alternatives in real-time with the prediction error of less than 3% compared to complex numerical simulation in presented example.
The paper advocates the use of Linked Data approaches as a means of overcoming the limitations of conventional monolithic data modeling in the context of the de-facto heterogeneity of information systems and data models in the diverse fields of the digital built environment. It enriches the discussion by focusing on one specific sub-domain – the application of digital methods in road design, construction and operation which involves the exchange, management and querying of spatio-semantic data that typically stems from different data sources and involves diverse software systems and data models. It argues that semantic web techniques significantly simplify the integration of heterogeneous data models. In a case study, the German road data exchange standard OKSTRA is linked with the Dutch CB-NL and RWS object type libraries. Doing so, it is shown how nationally well-established and widespread standards can be integrated in order to perform cross-country querying of road data. As a basis for the cases study, OKSTRA was transferred into an OWL-based data format, resulting in the creation of okstraOWL. The paper discusses in detail strategies for realizing semi-automated mapping with the Dutch CB-NL and RWS object type library, including a detailed analysis of the advantages and limitations of the different options. Major emphasis is placed on finding spatially related entities. Finally, the paper discusses the capabilities of linking the data sets by presenting a number of exemplary SPARQL queries for answering real-world questions that require the integrated analysis of data sets in different data models.
Presented herein is a big-data driven methodology for the detection of roadway anomalies, utilizing smartphone-based data and image signal streams. The methodology uses a vibration-based method and artificial intelligence for the detection of vibration-inducing anomalies, and a vision-based method with entropic texture segmentation filters and support vector machine (SVM) classification for the detection of patch defects on roadway pavements. The presented system pre-processes video streams for the identification of video frames of changes in image-entropy values, isolates these frames and performs texture segmentation to identify pixel areas of significant changes in entropy values, and then classifies and quantifies these areas using SVMs. The developed SVM is trained and tested by feature vectors generated from the image histogram and two texture descriptors of non-overlapped square blocks, which constitute images that includes ‘‘patch’’ and ‘‘no-patch’’ areas. The outcome is composed of block-based and image-based classification, as well as of measurements of the patch area.
Building information modeling (BIM) brings interoperability to the architecture, engineering, construction and facility management (AEC/FM) industry by providing an information backbone throughout a building’s lifecycle. Information interoperability or data mapping is critical for seamless information sharing between BIM and FM, because the approaches to information representation in BIM and in FM systems are different. The geometric and semantic information in BIM can be represented by IFC schema, while the FM data is commonly represented in a corresponding relational database. Even though COBie is a neutral data format to deliver information from BIM to FM, the data structure of COBie does not incomplete to represent all information from BIM to FM. Therefore, there is a lack of an appropriate solution to smoothly integrate BIM with FM. This paper aims to address the information interoperability issue and propose an ontology-based methodology framework for data mapping between BIM and FM. Based on the proposed framework, an ontology approach is developed as a tool to facilitate the facility knowledge management and to improve the data mapping process. Finally, a case application about facility maintenance activity is implemented to demonstrate the feasibility of the ontology approach.
This paper proposes using Deep Neural Networks (DNN) models for recognizing construction workers' postures from motion data captured by wearable Inertial Measurement Units (IMUs) sensors. The recognized awkward postures can be linked to known risks of Musculoskeletal Disorders among workers. Applying conventional Machine Learning (ML)-based models has shown promising results in recognizing workers' postures. ML models are limited - they reply on heuristic feature engineering when constructing discriminative features for characterizing postures. This makes further improving the model performance regarding recognition accuracy challenging. In this paper, the authors investigate the feasibility of addressing this problem using a DNN model that, through integrating Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) layers, automates feature engineering and sequential pattern detection. The model's recognition performance was evaluated using datasets collected from four workers on construction sites. The DNN model integrating one convolutional and two LSTM layers resulted in the best performance (measured by F1 Score). The proposed model outperformed baseline CNN and LSTM models suggesting that it leveraged the advantages of the two baseline models for effective feature learning. It improved benchmark ML models' recognition performance by an average of 11% under personalized modelling. The recognition performance was also improved by 3% when the proposed model was applied to 8 types of postures across three subjects. These results support that the proposed DNN model has a high potential in addressing challenges for improving the recognition performance that was observed when using ML models.
A continuous generation and evaluation of design variants characterize the conceptual architectural design stages. Variants comparison is a crucial process for making design-detailing decisions. Objectifiable criteria, including results of simulations and analysis, used for evaluation and comparison of design variants can legitimize decisions as the design process proceeds. A major challenge today is the management of design information and collaboration among several actors in a building project. Yet, a large portion of the Architecture, Engineering and Construction (AEC) industry deals with conventional methods to exchange design information. The growing use of building information models is promising but even the most recent developments and practices still rely heavily on human-readable protocols and issue management systems. Considering the potential of schematized computer-readable communications to be analyzed and used for future references and case-based reasoning systems. This paper proposes a novel minimized communication protocol that aims to introduce a computer-readable, yet adaptive universal method which works on schematized information exchange requirements (templates) for different use cases. The concept is demonstrated using an example scenario.
Navigation support is of significant importance for fire evacuation and rescue due to the complexity of building indoor structures and the uncertainty of fire emergency. This paper presents a framework of a multi-user voice-driven building information model (BIM)-based navigation system for fire emergency response. Classes of the navigation system is first defined being consistent with the open BIM data standard (i.e. Industry Foundation Classes, IFC). A string-matching method is then developed to generate a navigation query from each voice navigation request based on the Levenshtein distance and Burkhard and Keller (BK)-tree of a fire navigation associated lexicon. With the semantic information of the location in the navigation query, the spatial geometric information of the location is extracted from the BIM model and visibility graph-based route plans for multiple users are generated. To deliver the route planning to building users in an intuitive and direct manner, patterns of different voice prompts will be designed to automatically broadcast the navigation route step by step. Finally, the proposed navigation system will be validated with virtual reality (VR) based experiment.
Designers need to compare numerous design options in the process of designing an energy-efficient building. There are two impediments in this process, first probabilistic prediction of energy requirement at an early stage of design with uncertain design parameters and, second, selecting a design option based on the probabilistic energy prediction. The paper presents an integration of machine learning energy prediction model with building information modelling (BIM) tool to make probabilistic energy prediction, i.e. ranges of values. Wilcoxon rank-sum test is useful in this situation, which is capable of comparing alternatives based on probabilistic energy predictions. The tool has been developed to extract information from the BIM model, make probabilistic energy prediction using the Monte Carlo method, and perform statistical analysis. It has been found that BIM integrated machine learning model can make energy prediction of six design alternatives in 30-35 seconds with no additional modelling efforts. Higher uncertainty in the design parameters will result in larger uncertainty in the energy prediction, and the test may not be able to suggest the better option even using statistical comparison. This will require the more precise value of design parameters, i.e. reduced uncertainty. Different uncertainty levels in the design parameters have been tested to which extent they are sufficient to make a selection of the energyefficient option. It is observed that uncertainty levels that are suitable for decision-making depend on the combination of design options to be compared. It is possible to differentiate among alternatives with high uncertainty in the design parameters if they are entirely different else more precise definition of the design parameters is required. This research provides a method to select a better option among the developed options based on energy performance at the early stage of design.
