Computing in Civil Engineering 2017: Information Modeling and Data Analytics

Railroad management entities in the U.S. are developing new tools to improve the management of railroad bridge networks, in order to comply with new federal regulations on bridge safety and to increase their profitability. Decisions about maintenance, repair, and replacement (MRR) actions are currently prioritized by rating the bridges based on structural inspections and predictions about the estimated costs of operations. This study proposes a framework for the management of railroad bridge networks that: (1) utilizes a consequence-based management approach that considers relationships between displacements, serviceability levels and bridge MRR decisions; and (2) minimizes the expected value of total network costs by determining the best MRR decisions based on an annual MRR budget. Through this study, two different optimization methods are explored in two different scenarios: (I) mixed integer linear programming (MILP) when the impact of bridge location on costs is insignificant resulting in linear objective function and constraints; and (II) genetic algorithm (GA) when the impact of bridge location on costs is significant resulting in nonlinear objective function and constraints. A case study of a network comprised of 100 railroad bridges is used to demonstrate the proposed framework. The results show that scenario I leads the optimum MRR decision to replace more bridges. On the other hand, scenario II leads the optimum MRR decision to more repair or maintain groups of bridges which are closer to each other.

From 2011-2015, employers in the field of facility management (FM) recorded 144,600 injuries, with 39,950 cases resulting in 31+ days away from work. Case study analysis shows that many incidents were avoidable had the FM worker followed appropriate hazard mitigation steps. In an effort to improve efficiency and productivity in FM, BIM for facility management (BIMFM) has been an area of focus for many researchers and large public and private companies. While most BIMFM research focuses on the use of BIM models for O&M information storage and retrieval, this research presents a mechanism to improve worker safety during the FM phase. Currently, safety information for FM staff is fragmented in multiple repositories, in various formats, and is often uncategorized. Utilizing qualitative data collection methods, safety properties are established to mitigate three common hazard types for FM staff; falls, contact with/struck by, and harmful environments and substances. Utilizing these safety properties, an asset safety identification tool (ASIT) is designed in order to standardize asset specific safety properties. Using this tool, standardized safety properties are loaded into specific assets within a BIM model, mitigating fragmentation and streamlining information retrieval. This paper discusses the development of the safety properties and focuses on how the ASIT is utilized for documenting standard safety protocols in an effort to automate the connection of safety protocol information and assets based on asset parameters.

This study proposes a simulation-based and risk-aware multi-objective optimization framework to determine the capital structure for private financing in infrastructure projects to align the interests between the lender and the shareholder. Monte Carlo simulation is used for project valuation and conditional value at risk is introduced to measure the inherent risk associated with the expected project performance. A multi-objective optimization problem is formulated, where the decision variables are the amounts of diversified debt instruments, the first objective is to maximize the net present value while minimize the conditional cash flow at risk from the shareholder’s perspective, and the second objective is to maximizes the rate of return on the loans while minimize the default risk of shareholders from the lender’s perspective. Fast non-dominated sorting genetic algorithm NSGA-II is used to solve the problem. The case of Indiana Toll Road is presented as a numerical study to prove the concept.

Building information models are increasingly being integrated with building automation systems for making decisions associated with building performance and facility management and maintenance. Currently, there has also been an increased amount of interest generated towards construction operations building information exchange (COBIE). A data exchange format for facility management data handover, COBIE allows extracting asset information from a BIM model and provides a provision of importing this facility data into a buildings computerized maintenance management system (CMMS). Recent development in sensors and monitoring devices in building controls with wireless connection capabilities offers a huge potential for applications in smart buildings and building maintenance. This paper describes the development of an application that integrates sensor data collected using a wireless sensor network, and reports the data via the virtual model of the building. The application was developed to aid the facility management personnel with early detection of issues and maintenance checks. The data collected from the sensors is integrated with COBIE equipment information affected by this sensor data. The integration of equipment information and sensor data is accessible to onsite personnel and facility managers allowing easy access to field conditions and maintenance information during maintenance checks.

Forming semantically-rich 3D models of buildings from point clouds conveys vital as-built information that support applications such as construction quality assessment and control, construction progress tracking and infrastructure asset management. Conventional scan-to-BIM techniques rely on registration of scanned data with as-designed BIM objects. These methods are unable to handle cases where site-specific building element models are not readily available. Thus, this study proposes an unsupervised recognition framework to automatically identify generic building elements such as columns and beams from unstructured point clouds. The point cloud is first partitioned into multiple sections corresponding to each building floor. Next, column elements are identified using point cloud clustering and classification based on the bounding box dimensions. Beam elements are identified using plane projection and line fitting. The identified building elements are represented by volumetric bounding boxes in the resulting 3D model. The proposed method is validated using two separate datasets of laser-scanned building point clouds.

Maintaining facilities and the assets within the facilities is a critical task for large institutional owners. Many owners leverage computerized maintenance management systems (CMMS) to store, manage and query data related to their assets and work orders focused on maintaining the depreciating assets. Many CMMS applications have limited functionality to visualize this data relative to geographic and other spatial information. CMMS data can be associated with geospatial information to enable additional planning tasks related to location requirements. Recent advances in reality modeling allow us to quickly generate a highly detailed, geospatially accurate 3D digital model of a large area with multiple buildings at a low cost by leveraging digital images. This research first developed taxonomy of virtual 3D city model use cases in diverse domains. Then the value of visualizing facility asset information in a virtual 3D city model was explored. A case study was implemented through the creation of a large-scale 3D reality mesh of the University Park Campus at Penn State. A framework to integrate facility asset data with the reality mesh was developed. The taxonomy provided a better understanding of the scope of applications that virtual 3D city models can offer for model users. Integrating the reality mesh with CMMS data allowed for the demonstration of better visualization of maintenance scheduling decisions, along with improved maintenance personnel distribution.

Large institutional owners face a daunting task of assessing the condition of their facilities and forecasting maintenance and replacement costs of these assets. Computerized systems, generally known as asset management systems (AMS) are emerging in the market that provide maintenance budget forecasts based on building inventory data and conditions assessment. The main challenge that large institutional owners face is how to efficiently develop asset inventories that form the basis for AMS calculations. In response, this paper presents a research project wherein the research team developed a workflow for building information modeling (BIM) data transfer to AMS used for facility management (FM). This work provides a way for owners to leverage legacy BIM data to create building inventories or to specify BIM deliverables that will provide asset data for their intended AMS. This paper presents applied action research that developed a BIM to FM-AMS workflow as a proven practice that leads to sustainable facility management. The study results indicate that information exchange should require classification systems to utilize a consistent language between BIM systems. In addition, this paper presents peer institutional efforts and customization of classification systems that were studied to present a data crosswalk between these systems. The study results are expected to support owners and facility managers to understand the logic behind BIM information exchange practices so that they can adopt the best practice that fits the needs and goals of their organization.

During the planning process of a building, complex boundary conditions as well as technical and financial restrictions need to be considered. Consequently, the development and analysis of different design options are essential to find the optimum design for clients. At early planning stages, possible modifications may refer to building topology, geometric details or semantic characteristics. However, the manual development and evaluation of design options represents a time-consuming process. Despite a high impact on the following planning process, relatively little effort is spent on the investigation of alternative designs. Building information modeling (BIM) offers new prospects in the context of facilitating the creation and evaluation of design options. Due to the high number of project participants, the use of a neutral data format (e.g., industry foundations classes (IFC)) represents a prerequisite for project success. As information concerning changes, modifications as well as possible design options are not supported by IFC, individual models are required. However, this approach may result in inconsistency by using multiple model-specific identifiers referring to identical objects. Furthermore, creating models that consist of partially identical or even entirely identical objects leads to redundant data. In this paper, concepts for the formal modeling and management of design options using IFC are presented. The focus lies on the creation and management of alternative designs within a consistent and efficient BIM model. Following the object-oriented approach, possible design options are created by defining explicit option objects for building elements. Consequently, the required information is directly linked to the common BIM model.

Tower crane planning in pre-construction phase has long-lasting impacts on project cost and schedule. Current approaches and tools being used by industry to facilitate tower crane planning can be time-consuming, difficult to disclose all constraints, and challenging in visualizing and comprehending alternatives. Hence, tower crane planning experiential knowledge needs to be formalized to develop an efficient approach and effective tools to support tower crane planning. This research aims at formalizing criteria for tower crane planning by examining standards published by regulatory institutes and synthesizing knowledge of professional experts. Learning from the criteria and the knowledge inherent in current procedures, we devised an enhanced planning approach leveraging building information modeling (BIM) and optimization models. Finally, this paper demonstrates the utility of the formalized criteria and the proposed approach by applying them to a case study on a building construction project. Using a combination of parametric modelling, BIM, and optimization techniques, tower crane planning can be less time consuming and more accurate. Visualization capabilities and improved information quality in the planning process are also beneficial. For future work, the proposed approach for tower crane planning can be further analyzed to develop benchmarking metrics that can be applied, in general, to any building project, despite project unique characteristics.

During an emergency event, such as fire hazard or an incident of hijacking in a building or public facility, scout teams are assigned to assess the condition for the planning of response actions. To enhance the preciseness, responsiveness and timeliness of the evaluation of indoor status, an approach to determine, beforehand or in real-time, patrol routes for scouting is needed. In this paper, the authors present an efficient approach to find interior patrol routes in constructions with large scales or highly geometric complexity using building information modeling (BIM) and the multiple traveling salesman problem (MTSP). To reduce the computational time, a solution algorithm is developed to meet the objective. The results show that the proposed method performs well in both benchmark instances and very-large cases. The authors expected to help related agencies to establish a standard operation protocol for patrolling mission of emergency rescue.

With the increasing implementation of building information model (BIM) in operations and maintenance (O&M) management for existing buildings, various new data detection technologies for constructing an as-is BIM have been proposed and developed in past decades. In particular, extracting information from existing CAD documents is still a significant research area in the as-is BIM construction process. Data saved in CAD formats are unstructured and consequently it is hard to extract information and recognize objects in a systematic approach. With the ultimate goal of developing an effective and applicable approach to assist constructing as-is BIM objects, a novel semi-automatic approach is developed to detect structural components from CAD drawings. This approach mainly consists of a data processing part and a data management part. The data processing part defines the location of a structural component through recognizing special symbols in a floor plan and then extracting data from the floor plan using the optical character recognition (OCR) algorithm. The data management part analyzes and takes meaningful information from the extracted data based on predefined outlines. This paper first summarizes state-of-the-art information detection and analysis methods from CAD drawings. Then, the methodology and the prototype application developed in Matlab are introduced and discussed. Moreover, a preliminary set of tests using an office building and the analysis results from the tests are discussed as a case study from the perspectives of applicability and accuracy. Lastly, future works and limitations are also addressed.

Aiming at releasing the potential of BrIM model for improving the efficiency of bridge design and structural analysis, this paper analyzed the combined parametric method of BrIM models and built up structure properties for FEM, thereby established a three-class modeling frame and a new method of bridge analysis based on BrIM. To overcome the gap between BrIM and FEM in bridge structural analysis, the information of bridge structures and special analysis characteristics are set up to establish the reservoir of bridge structure analysis. The mesh controls of the bridge structural information is standardized and preset. The processing generates the corresponding switch file which file type is compatible with a variety of FEM analysis software tools. The seamless transition from BrIM to FEM greatly accelerated the design and analysis of bridges. The smooth parameter flow from the BrIM to FEM is also formed for quick information modification and mechanical analysis of bridge structures.

A post-earthquake fire simulation method of buildings considering the seismic damage of sprinkler systems is proposed, to quantitatively assess the influences of such damage on the building fires. Specifically, a modeling approach that converts building information model (BIM) to the model of fire dynamics simulator (FDS) is firstly designed, to create the high-fidelity models of buildings and sprinkler systems in FDS; a method for predicting the seismic damage of sprinkler components is developed based on the next-generation performance based design method FEMA P-58; according to the seismic damage of components, a method for predicting the seismic damage of the whole sprinkler system is proposed to determine the worst case of seismic damage. A post-earthquake fire of a six-story office building is simulated using the proposed simulation method, which indicates that the seismic damage of the sprinkler system has a significant influence on the spreading of post-earthquake fire.

In an industry foundation classes (IFC) building information model (BIM), proper space boundaries (SBs), especially 2^nd-level SBs, are needed to properly construct building surface geometry for building energy analysis. This study aims to develop a framework to achieve a rule-based automatic validation of IFC 2^nd-level SBs. Two types of errors, namely semantic and syntactic errors and geometric errors, which 2^nd-level SBs in IFC instance files may often encounter, are first analyzed. Corresponding to the two types of errors, the framework consists of two modules: 1) the semantic and syntactic validation module, which checks whether the IFC model complies with semantic constraints on 2^nd-level SB definition and the IFC schema; 2) the geometric validation module, which checks whether the 2^nd-level SBs are geometrically correct. For the syntactic and semantic validation, validation rules are identified and supporting validation tools are investigated. Rules for the geometric validation are also identified and the rule interpretation and execution method are discussed.

Reliable estimation of construction project costs requires generation of accurate quantity take-offs. Quantity take-offs are, traditionally, experience-based exercises that are often tedious and time-consuming. Although the BIM platform has been used to improve accuracy and efficiency of quantity take-offs, its widespread use is limited by the intellectual property issues associated with the transfer of complete BIM models or standardized IFC files from owners to contractors. Instead, owners may provide 3D model review files, such as Navisworks, that are capable of integrating 3D models created by various modeling platforms, to contractors. However, this platform does not explicitly provide essential 3D model data, such as component type, shape information, and geometric dimensions, that are essential for generating quantity take-offs. This significantly limits the use of the Navisworks platform, requiring estimators to manually create necessary data for millions of model items. Here, a novel computational approach that addresses these limitations, allowing quantity take-offs to be generated quickly, accurately, and cost-effectively, is proposed. The proposed method was cross-validated by comparing the obtained results to manual measurements. A practical case study based on an industrial construction project in Alberta, Canada, was conducted to test the functionality of the proposed method.

As a proven technique for improving project performance, benchmarking is regarded as a crucial component of a mature project delivery. However, it is generally accepted that incorporating benchmarking into an organization’s routine is a cumbersome and time-consuming endeavor because it requires a variety and vast amount of project information that is typically manually provided by project stakeholders. Furthermore, existing benchmarking systems do not generate critical benchmarks in real time, which results in reactive rather than proactive strategies to support performance enhancements. To address the aforementioned shortfalls, this research proposes a model-based benchmarking framework which leverages information stored in 3D models. This research focuses on healthcare projects which were benchmarked using a comprehensive list of performance metrics developed by the Construction Industry Institute (CII) in collaboration with a number of leading healthcare organizations. To achieve our objective, model contents were analyzed to explore the types of geometric and attribute data contained in 3D models by investigating a series of models collected from several healthcare organizations. The outcomes of content analysis were used to develop software requirements necessary to create a model-based benchmarking system. The requirements are further illustrated in mock-ups of a prototype system. The model-based benchmarking system will enable real-time benchmarking which can be efficiently implemented with less manual inputs.

This paper demonstrates how linked data can be used to provide a semantically rich connection between the domains of building information modeling (BIM) and geographic information systems (GIS). Expertise from both domains is necessary for the planning of infrastructure projects such as roads, tunnels, bridges, and railways, as these projects usually require the consideration of widely diverging scales. As BIM and GIS rely on different types of data modeling standards, a conversion between the data models will inevitably result in data loss. To overcome this problem, we propose to utilize the concept of Linked Data which allows the original data to coexist and provide coherence by establishing references between the corresponding entities of both standards. The approach is illustrated using exemplary models of shield tunnels in both, the IFC and the CityGML format.

Manual cost estimation is a task that requires experience; it is tedious, cumbersome and error-prone. As a result, manual estimation results are influenced by each estimator’s judgement or opinion. The use of computing technology can help automate this cost estimation process, reducing the subjectivity in estimation, improving the efficiency and the accuracy of cost estimation. This has been shown by the success of various building information modeling (BIM) software products. Nevertheless, existing cost estimation automation methods in these tools are still limited in terms of: (1) interoperability-in spite of their compliance with the industry foundation classes (IFC) schema, the detailed implementation in complying with IFC is largely proprietary; and (2) full automation-manual input is needed in various ways to fulfill the information need of the cost estimation automation. To address these two limitations, in this study, the authors developed a method and algorithms for fully automated cost estimation in wood construction, leveraging the fundamental geometric representation of wood building object representations in IFC models for quantity takeoff, and a price computation using a cost database generated from RSMeans Assemblies Cost Data 2015. The developed automated cost estimation method and algorithms were implemented in java programs and tested in a case study wood building project in Kalamazoo, MI, area. By comparing with estimates from existing BIM tools, improvement in interoperability and a reduction in manual input were observed using the authors’ newly developed method and algorithms.

According to the European Union report, 40% of the total energy was consumed in the building sector. Over 80% of energy was consumed during building’s operational stage. In addition, energy embodied in construction materials is also huge, e.g., producing cement is an energyintensive process. Therefore, a proper building design help reduce building’s environmental impact during its lifecycle. Due to the uncertainty of the relationship between individual design parameters and building performance, the extent of impact cannot be well-understood. In this study, main design parameters that affect the building performance are analyzed based on a BIM platform. Simulation of energy consumption is run for different scenarios in terms of operational energy consumption. Energy embodied in typical construction materials are calculated for each scenario. Sensitivity analysis is applied to find the extent of impact for the selected design parameters in terms of both operational and embodied energy. It is found that although the material of windows have highest energy coefficient, windows are the most energy-efficient and important parameter to be considered in term of life cycle energy consumption.

With building information modeling becoming mature, this technology provides several new possibilities for improving the checking processes of the design planning regarding the applicable codes in the construction industry. Code compliance checks are essential since they standardize the functions a building must fulfill finally. Currently, these controls are cumbersome and error-prone and performed to a large extent manual by the planning consultants as well as authority officers. To enable an automation, the contents of the applicable guidelines have to be translated into a digital format, which can be read, interpreted and applied by machines. Various approaches focused on a textual or hard-coded translation of the guidelines. To increase the involvement of non-programming users, we introduced the visual code checking language, which uses a visual notation to represent the content of directives machine- as well as human-readable. In the presented paper, the authors present a refinement of the VCCL to increase the coverage rate of the translation capabilities.

The last few years has been subject to an unprecedented growth in the application of 3D data for representing built environment and performing engineering analyses such as site planning, and condition assessment. Despite the growing demand, generating semantic rich 3D CAD models, particularly when building and structural systems are exposed remains a labor-intensive and time-consuming process. To address these limitations, this paper presents an end-to-end procedure to produce semantic rich 3D CAD models from point cloud data at a user defined level of abstraction. The procedure starts by segmenting a point cloud while considering local context using a multi-scale region growing algorithm. A Markov-Random-Field optimization labels segments based on their semantic categories. This step reduces the over-segmentation produced during the segmentation stage by compositing similarly labeled segments into super segments. The interconnectivity among these super-segments are reasoned and b-splines and solid geometrical representations are fit to produce 3D NURBS surfaces and cylindrical elements, respectively. Experimental results on real-world point clouds show an average fit error of 6.33 E-01 mm making the method the first to include beams and columns in an automated Scan2BIM process.

Improving energy efficiency and reducing environmental impact are two major priorities worldwide. The building construction industry represents a considerable share of global energy consumption and carbon emission. The majority of information required for sustainability evaluation of a building is usually specified at early stages of projects. Therefore, decision makers can potentially estimate energy consumption and environmental impacts of a building in the preliminary design phases in order to attain the best results in terms of sustainability. During the past decade, building information modeling (BIM) has become widely adopted in the architectural, engineering, and construction (AEC) industry, and the trend is still growing. Sustainability evaluation is one of the applications of BIM that can be implemented from early stages of building projects. Although, using BIM for extraction of information required for sustainability assessment shortens the evaluation process, the lack of interoperability between BIM and sustainability evaluation software makes the information exchange process tedious and error-prone. This paper reviews the current status of the application of BIM for sustainability assessments and identifies issues and shortcomings. It also investigates potentials of BIM standards such as IFC, as BIM’s open standard file format, and the level of development (LOD) Specification for addressing these challenges. Finally, a three-level IFC/LOD-based LCA analysis approach is presented, which can enable designers to estimate the embodied energy of buildings from early stages of projects, and refines such estimation as a design progresses by adding more reliable content to the model.

Localization for the resources and workers on construction sites is significant and essential for effective construction projects management. Since global positioning system (GPS) usually does not work indoor, indoor localization is critical for jobsite indoor resource and labor management. However, due to the complex and variable environment situation on construction jobsites, accuracy and quick deployment remains to be two primary challenges for practical and wide adoption of localization technology on jobsites. As radio signal strength index (RSSI) based indoor localization techniques have relatively low costs and easy deployments comparing to other techniques (e.g., Ultra Wide Band, Ultrasonic), RSSI-based localization technologies have attracted much attention in construction industry. However, the accuracy of existing RSSI-based technologies is still not satisfactory. Therefore, the authors proposed to improve the accuracy of RSSI-based indoor localization technology by utilizing the information extracted from BIM in the localization algorithm. A lab with BIM model is used as the testbed for the algorithm’s validation at 19 discrete locations and one continuous moving path. The test results indicate that the localization accuracy can be improved by filtering up to 25.6% of the infeasible positions based on building information modelling (BIM).

Four-dimensional (4D) building information modeling (BIM)–3D BIM plus time–is used in various aspects of construction project management, including construction planning, constructability review, communications with clients, and collaborations among project participants. An emerging area of 4D BIM in the current practice is progress control, whereby 4D BIM models are manually updated to capture the construction progress as project proceeds. This paper presents a Web-based framework that harnesses information from daily reports to update the 4D BIM models and visualize the project progress. Leveraging the advancements in the Web and database technologies, the newly developed approach automates the updating and visualization process. A prototype system was developed and validated using the project schedule and BIM models from a real construction project.

Building information modeling (BIM) has been widely adopted in architecture, engineering, construction, and facilities management (AEC/FM) industry. Recently, 3D BIM models are commonly used; however, there is still some debates going on upon the necessity of as-built BIM models linked with operations and maintenance (O&M) information of building’s components. Owners are always faced with the dilemma of the benefits of real-time data driven BIM application in their projects’ lifecycle. In this paper, the timesaving benefits of BIM models linked with real-time operational information in facilities management (FM) practices during a building’s life cycle are studied. Real-time data driven BIM model application in operations control and maintenance, the two main responsibilities of FM departments, is studied in this paper. Several scenarios, tested in a building owned by USC facilities management department, are designed to determine the timesaving benefits of BIM model utilization. These scenarios are divided into two major categories: operations control and maintenance. Data acquisition from CRAC units, Chillers, and AHUs for operations control and two maintenance scenarios of occupant dissatisfaction from office temperature and leakage in chilled water lines have been studied in this paper. The time needed to accomplish the explained scenarios has been measured using two available approaches (Traditional vs. real-time data driven BIM). Based on achieved results, possible timesaving benefits of 6D-BIM model application are explored and presented.

Building information modeling (BIM) has become a standard practice in the architecture, engineering, construction, owner and operator (AECOO) industry, and has gained a critical mass of adoption. However, for companies to fully harvest BIM’s potential and sustain its momentum, the key is to cultivate a competent workforce that has the desired BIM knowledge, skills and abilities (KSAs) to perform job tasks in both organizational and project contexts. This paper reports a research undertaken by the academic interoperability coalition (AiC) using a Delphi method to explore the body of knowledge (BOK) of BIM. The BIM BOK is defined as the systematically curated concepts and nomenclature, knowledge, best practices, standards and outcomes pertaining to BIM and its implementation. The Delphi study is built upon a comprehensive and multi-dimensional system architecture to address both macro and micro BIM uses, and to classify and categorize the BIM BOK. This three-round Delphi study produced an inclusive and living inventory of BIM BOK line items, with built consensus on their practicality associated with the various user groups at different competency levels in specific implementation contexts. The BIM BOK makes three major contributions: 1) it establishes the foundational and holistic knowledge representation for BIM competency assessment for workforce across the AECOO industry; 2) it provides a roadmap and common curriculum to bridge the current gap between academic learning outcomes and workplace performance requirements via corporate training and BIM education; and 3) it standardizes the level of expectation and benchmarks job task performance for emerging BIM job titles, which forms baseline performance measurement for BIM education accreditation, and professional credentialing and certification.

Investigating the drivers for building information modelling (BIM) adoption by organisations has attracted a significant interest in recent years. However, there are still some important limitations to overcome in this areas: (1) drivers and factors for BIM adoption are disjointedly identified and dispersed across many studies−this is caused by both the specialised theoretical lenses embraced by researchers and the use of non-systematic approaches for identifying pertinent studies; and (2) inadequate attention to the meaning of key terms and concepts (i.e. readiness, implementation, diffusion, adoption) underpinning this area of investigation and their overlap. This paper addresses these shortcomings and presents an exhaustive set of the drivers (i.e. innovation characteristics, internal environment characteristics, and external environment characteristics), key factors and their potential determinants that affect BIM adoption decisions. These findings can be used to: (a) investigate the power or influence exerted by the different drivers, factors and determinants on the organisational adoption of BIM within markets with different macro diffusion dynamics−e.g. with and without BIM mandate; (b) facilitate BIM diffusion planning by setting actions that target specific drivers and factors; and (c) develop a conceptual model for BIM adoption by organisations.

This paper presents the development of a framework for leveraging the 3D nature of spaces, their attributes, and sensor location data in conjunction with building management system (BMS) sensor data to allow both users and operators to more easily access building data in the operations and maintenance of facilities. A case study, using data from various sensors of an operational facility and data from the building information model (BIM) used in the construction of the facility, is presented in a spatial data visualization framework. The initial prototype focuses on 3D data visualization by building occupants to enable exposure and access to facility data. The framework offers an initial system to integrate two typically disparate data streams to aid in building life-cycle data integration as well as increase transparency and awareness of operations for building occupants.

Operation and maintenance of commercial facilities rely on complex building information to locate assets and troubleshoot equipment failures. Although building information models (BIM) provide an integrated information repository for searching for facility information, currently more than 90% of existing buildings in the U.S. were still built with 2D drawings. Manually reconstructing BIM, especially the mechanical systems, has been proven to be very labor-intensive. Hence, previous research studies have investigated automated 3D reconstruction approaches using point clouds data obtained by laser scanner techniques or 2D photos. But due to the needs for a line of sight, these approaches are not applicable for constructed buildings. Researchers have also developed approaches that convert 2D drawings to spatial building models, but the previous studies are limited to generate architectural components such as the wall, doors, and windows. Thus, this paper investigated challenges and approaches to automatically generate models for mechanical systems in buildings using 2D drawings. We analyzed the contents and characteristics of the mechanical components that are represented in drawings and developed a set of classification and algorithms that support the automated recognition of the spatial information and metadata of the mechanical systems. A software framework was proposed to utilize the developed computational approach to recognize 2D drawings for BIM reconstruction. The results of the prototype demonstrated that more than 80% of ducts can be recognized from various drawings in DXF format.

Analyzing the value of buildings and how this value is affected by the various design decisions is vital for delivering value-adding buildings. The value of a building to a specific stakeholder is a function of the stakeholder’s personal value system (the personal values of that stakeholders and their order of priority) and the building properties. Building information modeling (BIM) can, thus, facilitate the execution of the value analysis process, in an efficient and accurate manner, because it allows for a rich representation of building properties. To achieve such efficient and accurate value analysis, there is a need to develop a BIM-integrated automated value analysis system that can directly retrieve value-specific information from a building information model and use it for assessing the value of a building to its stakeholders, based on their own personal value systems. To address this need, this paper proposes an automated value analysis system that is integrated with BIM. The proposed system is composed of three modules: (1) a module for soliciting information about a stakeholder’s personal value system; (2) a module for retrieving value-specific design information (i.e., relevant and sufficient information for value analysis including building objects and their properties) from a BIM model; and (3) a module for quantifying and aggregating the value of a building to a stakeholder based on the stakeholder’s personal value system and the building properties. The proposed system was evaluated using a case study that analyzes the value of a two-story apartment building to its stakeholders. The preliminary results show that the proposed system could support more effective and efficient value-sensitive decision making by allowing stakeholders to explore and identify high-value design alternatives efficiently–with less time and effort.

Open data standards (e.g. LandXML, TransXML) have been widely recognized as a solution to the interoperability issue in exchanging digital data in the transportation sector. Since these schemas include rich sets of data types covering a wide range of disciplines across all project phases, model view definitions (MVDs) which define subsets of a schema are required to specify what types of data to be shared in accordance with a specific exchange scenario. The traditional method for generating MVDs is time consuming and tedious as developers have to manually search for entities and attributes names that semantically match to the data exchange requirements. This paper presents a computational method that automatically maps users’ keywords to semantics-equivalent data labels (classes and attributes) in LandXML data schema. The study employs a lexical database of civil engineering terms to interpret users’ intention from their keywords. The study also introduces a context-aware entity search algorithm that is able to find equivalent or most similar entities for a given keyword. The developed method has been experimented on a set of keywords extracted from an asset management manual. The experiment results show that the design algorithm is successful in generating partial LandXML branches from keywords.

To help prevent noise-induced hearing loss (NIHL), a permissible exposure limit for construction noise has been set to 90 dB over an eight-hour period by the Occupational Safety and Health Administration (OSHA). The National Institute for Occupational Safety and Health (NIOSH) has further set the recommended exposure level to be 85 dB for an eight-hour period. However, construction workers still lack enough awareness of the potential noise hazard in their workplace and the damage that long-term exposure to noise can cause. The feasibility of noise mapping has been approved in improving intuitive noise related awareness especially in urban traffic planning, however, little research of its benefits has been explored in construction worker safety training. Building information modeling (BIM) is increasingly applied in construction safety related analysis and training, while integration of noise hazard into BIM has not been employed. Therefore, this paper presents an innovative framework of integrating wearable sensing data into BIM for construction workplace noise hazard prediction in safety training. The objective of the framework is to predict and visualize noise spatial distribution in BIM based on collected scattered data from wearable noise sensors. A case study is conducted in an indoor renovation project to validate the feasibility of the proposed framework and evaluate its performance.

Construction industry is a labor-intensive industry. Sentiment or mood of participants in the construction industry is a key issue in this business. To analyze this issue, using traditional ways such as questionnaire survey to collect data is both time- and cost-consuming. Recently, with the rapid development of social media services, data can be collected and extracted for sentiment analysis to provide officials and managers with fresh perspectives on participants in the construction management. In this paper, a sentiment analysis systematic framework is proposed. This system collected user messages from social media sites, establishes and compare different clusters emotion dictionaries by time duration and location. This paper generated valuable information and knowledge in the construction domain. As an initial trial, this study selected social media of Weibo because of its wide usage in China. Four clusters which include construction workers, construction companies, construction unions, and construction media were analyzed. For each user, the crawler is used to collect the Weibo messages from his/her Web page. On average, there are 135 messages collected for each user. This research then analyzed these data in the following aspects to dig out sentiments behind data: hourly, daily, monthly, and locations. Detailed findings, benefits and barriers to incorporating social media data analytics in the construction industry, along with future research, were discussed. This paper benefits the academia by testing an alternative way of studying the construction population, which further will help decision makers better understand the real situations of the construction industry.

Around 40% of the energy used in buildings is wasted as a result of inefficient operating and maintenance of the HVAC systems. Various approaches, such as automated fault detection and diagnosis (AFDD) and retro-commissioning are used to detect faults and minimize waste in HVAC systems. However, either they aim to identify failures (e.g., AFDD) after they occur or wait for the assessment period to arrive (e.g., retro-commissioning), both of which result in energy waste until problems are discovered. The facilities management industry needs approaches that help them to continuously monitor their systems, observe changes in expected pattern of behavior and pinpoint such changes before failures or large scale equipment replacements occur. The advances in the data science field provide opportunities to be leveraged towards information extraction from building operation data and investigation of the expected and actual behavior of systems. The objective of this paper is to enable automated analysis of the behavioral patterns of major HVAC equipment over time and assist in detection of variations in the system behavior that are not expected. As an initial case analysis, performance data (e.g., air temperature, valve status, air flow, etc.) of an air handling unit in a campus facility has been analyzed to identify the frequently occurring behavioral patterns and associated energy use. The initial results reveal the characteristic behavior of the system and the variations of these patterns over years to help the facility operators to pinpoint unexpected changes in system behavior and eliminate energy waste associated with such changes.

When building systems and occupants use energy, they create data—much of it unstructured and characterized as long running time series. Energy data captured at the plug level offers an opportunity not only to analyze highly granular building activities, but also to infer information about the behavior of occupants. Previous work examining occupant behavior typically seeks to understand how individual occupant schedules can be better modeled to improve the efficiency of building system operations, and therefore they treat individual actions as entirely self-contained. However, individual behavior—including that which draws power in buildings—is highly influenced by the inherent spatial and social network structures of occupants. Therefore, understanding the underlying spatio-social occupant network in a commercial building is integral to driving more energy efficient operations. Doing so is challenging since network relationships are highly complex and difficult to directly measure using traditional methods (e.g., surveys) and will require a deep understanding of occupant behavioral patterns. In this paper, we propose an automated methodology for inferring occupant behavioral patterns by classifying raw plug load data ascribed to individual occupants into energy use states. Our method utilizes a Gaussian Mixture Model to model occupant energy use variability and probabilistically classify energy use data into one of three generalized states. We present preliminary results of our classification algorithm using empirical data from a fifty-person commercial office building in San Francisco, California.

Automatic detection and classification of defects in infrastructure surface images can largely boost its maintenance efficiency. Given enough labeled images, various supervised learning methods have been investigated for this task, including decision trees and support vector machines in previous studies, and deep neural networks more recently. However, in real world applications, labels are harder to obtain than images, due to the limited labeling resources (i.e., experts). Thus we propose a deep active learning system to maximize the performance. A deep residual network is firstly designed for defect detection and classification in an image. Following our active learning strategy, this network is trained as soon as an initial batch of labeled images becomes available. It is then used to select a most informative subset of new images and query labels from experts to retrain the network. Experiments demonstrate more efficient performance improvements of our method than baselines, achieving 87.5% detection accuracy.

Parking signs recognition and localization aim to extract and digitize accurate on-street parking restrictions. Today’s visual data collection, annotation, and analysis practices are still costly, prone to error, and cumbersome as performed manually. While online street-level imagery databases contain updated panoramic images of all signs, their potential for understanding on-street parking restrictions at scale has not been fully explored. The key benefit of these databases is that once the parking signs are detected, accurate geographic coordinates of the detected signs can be automatically determined and visualized within the same platform. This paper evaluates the application of a computer vision-based method for parking signs recognition from street-level imagery aimed at facilitating the parking in dense cities. The method extracts images and leverages a sliding window mechanism to detect potential candidates for parking signs. A Histogram of Oriented Gradients for each candidate is formed and trained. Hard negative mining approach is then used at final stage to remove the remaining false positives. The potential detections of images from multiple viewpoints are then combined to locate the position of the signs on a map. Experimental results with high accuracy demonstrate the potential of leveraging street-level images and provide a viable solution for digitizing at scale all parking signs to help drivers understand parking rules and avoid fines.

Bridge inspection reports provide a large amount of technically-detailed bridge condition and maintenance history data/information that could be utilized for advancing the understanding of bridge deterioration for supporting improved maintenance of bridges. However, the valuable information buried in these inspection reports remains unexploited. There is, thus, a need for information extraction (IE) methods to automatically extract information about existing bridge deficiencies and performed maintenance actions from these reports for further analysis. However, existing IE methods are limited in their ability to extract information from such highly-technical text, with varying levels of technical detail, text patterns, and quality. To address this gap, this paper proposes an ontology-based IE framework that extracts entities and relations (i.e., dependency relations) about bridge deficiencies and maintenance actions, and represents the extracted information in a structured way. The proposed IE framework is composed of two primary components: (1) a name entity recognizer for term identification, and (2) a relation extractor for relationship association. This paper focuses on presenting the proposed similarity-based dependency parsing (DP) methodology for automated relation extraction. The proposed DP methodology utilizes a transition-based DP model to capture sentence-level dependency configurations, and represents each configuration with a distributed representation. Each unlabeled configuration is compared to labeled configurations in the distributed representation space by a cosine-similarity measure, and is then labeled with its most similar labeled configuration’s transition. The proposed DP methodology achieved a configuration-based accuracy of 78.0%.

Sustainability assessment has drawn much attention worldwide. However, current sustainability assessment models/systems are usually focused on a single scale, which does not allow for capturing the interdependencies across the different scales. Also, existing sustainability assessment models/systems typically require high levels of data collection and complex measurements, which is usually costly and time-consuming. To address these gaps, the authors propose a data-driven multiscale sustainability assessment and prediction (DM-SAP) framework, which (1) focuses on the building, neighborhood, and city scales, and (2) takes a data-driven approach that relies on machine learning. The machine learning approach focuses on two primary objectives: (1) learning from history (previous building, neighborhood, and city data) to predict and assess sustainability metrics (e.g., energy consumption) in an efficient and reliable manner, and (2) feature selection to identify the minimum amount of data that would be sufficient for reliable assessment. This paper focuses on presenting the feature selection and the machine learning-based model development for predicting residential building energy consumption for supporting multiscale sustainability assessment. The LASSO algorithm was used for feature selection. Three machine learning algorithms were implemented and tested. The prediction performance was evaluated in terms of coefficient of variation. The 2009 Residential Energy Consumption Survey (RECS) by the U.S. Energy Information Administration (EIA) was utilized for model training and testing. The testing results showed reasonable prediction performance.

The transportation project decision making process has long been criticized for resulting in frequent delays for the development of important projects. For large-scale transportation projects, late identification of stakeholder concerns have been identified as a major contributor to the delay. In order to help transportation decision makers better identify stakeholder concerns in the early project stage, this paper proposes a methodology for stakeholder opinion classification, which classifies the opinion expressions extracted from stakeholder comments on large-scale transportation projects into different concern groups. In developing the proposed methodology, both supervised methods and unsupervised methods were evaluated and compared. For the supervised methods, several machine learning algorithms were tested. For the unsupervised methods, different word-embedding methods were investigated. All methods were trained on a dataset of 1,369 comment sentences, and were tested on a dataset of 376 comment sentences. Both, the training and testing datasets were randomly selected from a comment collection for eight large-scale transportation projects nationwide. This paper focuses on presenting and discussing the proposed stakeholder opinion classification methodology and the experimental results.

Leakages in gas pipelines are associated with significant safety, economic, and environmental concerns. In 2014, there were 113 gas distribution pipeline incidents across the United States. The annual loss rate from all downstream components of the natural gas (NG) system in the Boston region, MA, was estimated to be around $90 million worth of NG fuel in 2015. Environmental concerns further support this effort given that methane (the major component of NG) is a potent greenhouse gas. To address these issues, numerous research studies have developed sensing technologies and hydraulic simulations for detecting leaks in pipeline networks. However, most of these methods lack integration of the detected leaks and infrastructure properties to provide feedback and support risk-informed decisions about a given pipeline network. Through analysis of a citywide pipeline leakage and maintenance data set for a mid-size city in the United States, authors have developed a statistical model relating pipeline infrastructure properties and detected leaks on segments over multiple periods of surveys. A calibrated generalized linear model provides the expected number of leaks for pipeline segments, based on factors such as pipe age, material, and diameter. This model reveals information about (1) influential network characteristics contributing to leakages, and (2) hot-spot leak regions. The identified influential factors and hotspot leak regions can utilize decisions on optimal sensor placement or leak surveying and hence improve maintenance activities.

Improving the energy efficiency of residential and office buildings is an important way of reducing energy consumption and CO₂ emissions and eliminating their associated adverse environmental impacts. On the other hand, people spend the majority of their times in buildings and demand comfortable, healthy, and productive indoor environments. Building energy conservation and occupant comfort are usually considered to be conflicting objectives. Towards exploring ways of reducing building energy consumption without sacrifice in thermal comfort, this paper aims to study the relationship between energy use mode, thermal comfort, and energy consumption in office buildings. To conduct an initial experimental study, a floor of an office building in Philadelphia, PA, was instrumented and monitored. Cooling energy consumption, indoor temperature, occupant feedback on thermal comfort, and outdoor temperature data were collected for three months. A clustering algorithm was used to identify three energy use modes: saver, balanced, and spender. The correlation between building operation efficiency and thermal comfort was investigated, and thermal comfort levels under different energy use modes were compared. The results show that there is a very weak correlation between building operation efficiency and thermal comfort, and that there is not a significant difference in thermal comfort across the three energy use modes. The findings of this study may help in identifying building operation strategies that reduce energy consumption while remaining thermally comfortable.

A large portion of the injuries incurred on construction sites is due to the lack of posture awareness among labors and crew while performing highly physical tasks. Most of these injuries are caused by bad and inexpert poses. Posture and gesture awareness is therefore a critical issue that can substantially decrease the number of injuries on construction sites. This paper presents a framework for identifying the level of expertise based using a machine learning-based classification algorithm. Expert and inexpert performance are identified, in order to detect unsafe and/or unproductive postures taken workers, and avoid injuries to improve productivity. The proposed framework has two major components: (1) codebook generation based on a sensor-based body joints model representing the poses taken by the labors, and (2) training a support vector machine (SVM)-based classifier for identifying expert vs inexpert performance. A set of experiments, with twenty-one masons with varying levels-of-expertise, is designed for verification and validation of the proposed methodology. An inertial measurement unit (IMU) suit, with 17 sensors, is used for data collection. While the method is applicable to any types of construction trades, this study focuses on masonry brick laying tasks. Results show that the classifier is capable of identifying the two forms with a reasonable accuracy. Moreover, SVM linear kernel classifier generates the most accurate results with lowest computational cost in classification. The low computational cost of this classifier makes it feasible for on-the-field deployment.

Availability of 3D underground information models is key to designing and managing urban infrastructure construction projects. Buried utilities information is often registered by using different types of location data with different uncertainties. These data variances are, however, not considered in many existing visualization approaches. This study, therefore, aims to enable 3D-representation of uncertainties in underground utility data. To this end, the research team identified four different location parameters (unknown, standard, estimated, surveyed location), and explored how these could be integrated into models that visualize utilities in 3D. As proof of concept, we developed a first 3D uncertainty model for an urban road intersection and implemented this in a handheld augmented reality application. As a next step, we designed an improved 3D visualization model that incorporates standard, estimated, and surveyed location data to estimate utility locations, and to generate cylinder shapes that represent the uncertainty buffer around a visualized utility. The presented approach eventually allows city engineers to better estimate design and construction workspaces. We conclude the paper by elaborating both the contributions and consecutive research steps.

Laser scanning has been widely used in as-built 3D modeling for construction and infrastructure management. When laser scanners can produce dense point clouds that capture cm-level features in minutes, efficient 3D point cloud processing, storage, and visualization remain challenging due to gigabytes of 3D imagery data. A practical solution is to compress the point cloud while keeping enough geometric information (e.g., edges, corners). Existing compression method uses uniform sampling that reduces data densities everywhere without considering that geometric complexities of different parts of a scene may deserve varying data densities. However, limited studies quantitively assess the deviations between pre- and post-compression point clouds. This research examines a laser-scanning data compression method that enables automatic compression of a point cloud with varying subsampling rates per geometric complexities of parts of data and a user-specified compression ratio (the ratio of the removed point cloud to the original point cloud, the segment with the highest value is the most compressed). Compared with qualitative methods, this method quantifies the relationship between conserved geometric details, subsampling rates, and geometric complexities. First, the developed approach takes segmented point clouds as inputs (either manual or automatic segmentation results) that contain data segments of significantly different geometric complexities depending on the shapes underlying the data segments. Next, surface smoothness and curvature are calculated to quantify the geometric complexities of point cloud segments. Then per total compression ratio, the developed approach assigns various sub-compression ratios to the segments per their geometry complexities (e.g., complicated segments are assigned higher compression ratio to keep more data and vice versa). Finally, this approach evaluates compression results by accessing deviations between the original and compressed point clouds. Results from a case study indicate that the proposed method can achieve a high compression ratio and conserve more information than the uniform sampling approach.

Infrared thermography (IR) is a modern, non-destructive evaluation technology for monitoring and assessing civil infrastructure conditions. It mainly relies on measuring the infrastructure surface temperature to identify any potential defects. Currently, most of the existing research studies in IR rely on 2D thermal images which are time-consuming and labor-intensive. This paper describes a case study that examines the use of both infrared and visual sensing in recording thermal and spatial conditions of a subway tunnel segment in the city of Montreal, Canada. In the case study, both thermal and visible images of the infrastructure conditioning data were collected separately. Next, the visible images were used to generate a 3D point cloud model by applying the structure from motion approach. In parallel, each set of overlapping thermal images were stitched to form a thermal panoramic image that covers a large surface area with an accurate temperature representation. The stitched thermal images were finally mapped to the 3D point cloud in order to produce both thermal and metric measurements of a subway tunnel segment. The results of the proposed framework demonstrate that 3D thermal modeling using visual and infrared sensing is able to generate geometric and thermal information of indoor infrastructure environments. Furthermore, this approach is affordable in terms of cost and time.

Bridge infrastructure is aging, and maintenance backlog is steadily rising due to low budgets. Therefore, accurate and timely condition assessment of bridges can support decision-makers to effectively monitor and assess structural condition for repairing/replacing elements before major damage or collapse state is reached. Improved techniques, such as laser scanning, have enhanced inspection and monitoring of infrastructure, but assessment and interpretation of the collected data remain to be challenging due to uncertainties present in them. Presence of these uncertainties calls for a need to look at condition assessment from a probabilistic perspective as opposed to the deterministic approach followed in the current practice. Given that there are different types of uncertainties stemming from various sources, to facilitate any analysis incorporating these uncertainties, would require representing them in a computer-interpretable manner. In this paper, we present a representation model that covers parameters associated with defects, scan environment and related uncertainties, which will allow us to do a probabilistic condition assessment.

Dense three-dimensional (3D) point clouds of infrastructure systems, generated from laser scanners or through multi-view photogrammetry, have significant potential as a source of nondestructive evaluation information. The growing maturity of these techniques make them capable of reconstructing photorealistic 3D models with high precision at the millimeter scale adequate for inspection and evaluation practices. Manual analysis of these rich point clouds is often time consuming and labor intensive and does not provide explicit information on structural performance and health conditions, highlighting the need for new techniques to efficiently analyze these models. This paper presents a new multiresolution change analysis approach for tracking small movements in point clouds over time through localized spatial analytics. This technique uses a combination of a direct point-wise distance metric in conjunction with statistical sampling to extract structural deformations. By identifying and tracking these changes, mechanical deformations can be quantified along with the associated strains and stresses. These measurements can then be used to assess both service conditions and remaining system capacity. The results of a series of laboratory experiments designed to test the proposed approach are presented. The findings indicate measurement accuracy on the order of +/- 0.6 mm, making it suitable for accurate and automatic geometrical analyses and change detection in a variety of infrastructure inspection scenarios. Ongoing work seeks to connect this technique to automated finite element model updating, and to field test the methodology.

Point cloud-based reconstruction is a crucial process for process monitoring and as-built modeling, but still a challenging task in construction applications. Due to the finite perspective of a scan, multiple scans from different locations are required to obtain full scene coverage. Then, a registration process is required to transform all point clouds into a common coordinate frame. In this paper, a fully automated target-less point cloud registration method is presented. For this approach, a laser scanning device and a co-registered digital camera are required. This method consists of three major steps for an automatic registration of point clouds. First, it detects common features on the corresponding overlapped images from different locations. Second, it achieves coarse registration using the extracted features on images. Lastly, the fine registration is attained by the iterative closet point (ICP) algorithm. The proposed novel method was tested on a real construction site and the test results of automatic registration of all different scan positions were successfully verified in terms of accuracy.

Aging nuclear power plants are susceptible to the damage onset. Regular inspection of reactor internal components is needed to ensure safe operations. Current practice involves operators watching inspection videos that are captured under the water, and manually labeling the cracks on metallic surfaces of nuclear reactors. This approach is time-consuming, tedious, and subjective. On the other hand, most of the prevalent autonomous crack detection methods were designed for concrete surfaces, which could not detect the cracks on metallic surfaces since they are typically very tiny with low contrast. In addition, the state-of-the-art methods tend to detect scratches, welds, and grind marks on the surfaces as cracks, which cause a large number of false positives. To resolve the above issues, a novel autonomous crack detection method is proposed based on texture analysis and Bayesian data fusion. In each video frame, local binary pattern (LBP) is used to analyze texture and detect crack patches to form crack bounding boxes. After that the crack bounding boxes in different frames are fused with Bayesian decision theory, which dramatically increase the robustness and reliability of detection. To optimize the computation speed, integral histogram and 2-stage support vector machine (SVM) are applied. The proposed method is evaluated using a thousand of frames from several inspection videos. The results showed that it is accurate and efficient, where the other methods could not perform well.

Climate-related disasters have increased in terms of occurrence and intensity over the past decades. After the events occurred, the “cleaning up” stage of a significant amount of waste is a prerequisite step for search-and-rescue and damage recovery missions, which requires extensive efforts in terms of time and cost. For effective and efficient disaster waste management, accurate information on the amount and location of the waste should be delivered in a very timely manner. Most existing studies have developed statistical models to predict the amount of the waste using the waste records generated during the past disaster. Although the models were widely used for disaster recovery, they exhibited some limitations. They cannot be applied for the countries that do not have historical data of disaster waste. Thus, this paper proposes a 3D information-based quantification and localization method of disaster waste, using unmanned aerial vehicle (UAV). The proposed method consists of five main processes: image acquisition, point cloud generation, waste dump detection, volume estimation, and data visualization. To validate the feasibility of the proposed approach, experiments were performed on waste disposal sites in Seoul, Korea, to measure the volume of solid waste dumps through dense point cloud analysis. In particular, two algorithms were applied in order to extract a group of points representing a waste dump based on the clustering method and calculate the volume of the selected waste dump. A UAV was used to capture 3D images of the site and the images were transformed into point clouds, eventually generating more than 863 million points from the 111 images. The results confirmed the error rate of 7.04% in the volume measurement, promising the potential for wider implementation of the method developed to conduct disaster waste management more effectively.

A mass casualty incident (MCI) is a single incident which triggers a large number of emergency medical demands requiring different treatment based on the severity of injury. The time of treatment is an important indicator to measure the service quality of emergency medical care. However, limited by the number of ambulances and capacity of hospitals, an MCI disturbs the proper functioning of the usual pre-hospital health care system, and the zone where the MCI occurred usually is challenged by traffic congestion when a great number of demands occurred in a concentrated short period of time. This study aims to propose a traffic flow problem considering congestion of routes and hospitals. In our work, we develop a mixed integer quadratic programming, where the objective function is quadratic. The proposed method is aimed to improve the survival of victims in an MCI.

A vision based method was introduced to detect human patterns. The purpose of the research is to assist the decision maker while considering the installation of automated external defibrillators (AEDs). We first recorded the video of people in the objective building. During the process of detection, we adopt the histogram of oriented gradient (HOG) to detect the locations of people at each time interval. HOG is the descriptor which could describe the shape of objects in the image, based on the orientation or gradient of each pixel. After the extraction of the human volume data, we further analyze the patterns of the human and the distribution of human in the building. The analysis would help to plan and evaluate locations of AEDs based on the real potential demand distribution and could optimized the performance of AED during an emergency.

Incidents occurring on a freeway system can cause significant impact over traffic flows, such as long-duration and large-scale congestion and potential danger to the driving environment. Hence, prompt incident response is critical for freeway management. The objective of this study is to investigate the deployment of freeway incident response teams through the optimal partitioning of freeway segments, so as to attain the efficiency of dispatch and routing. Based on the analysis of historical incident data and response records, this study first visualizes the spatiotemporal pattern of incident occurrence over a GIS map. The factors that are significantly associated with the frequency of incident occurrence are further identified, and the possible causality is discussed thereupon. Considering the association and the structure of the freeway network, this study develops a partition problem, where each incident response team is located near a certain ramp and generally responsible for a set of freeway segments. Upon the expected incident occurrence during different planning horizons or time-of-day periods, the partition problem determines the freeway segments belonging to each incident response team in a time-dependent context, seeking to minimize total mileages traveled while balancing the workloads over the response teams. The visualization of the partitioning result can be used by a control center, as the reference for online dispatching and routing incident response teams. Combined with real-time freeway monitoring within a GIS-based deployment system, the time-dependent partitioning enable freeway administrators to more flexibly and efficiently address various incident scenarios, such as trip chains for multiple incidents of different emergency levels, based on better understanding of freeway incident characteristics.

Large transportation hubs are difficult to navigate, especially for people with disabilities such as those with visual or mobility impairment, autism spectrum disorder (ASD), or simply those with navigation challenges. The primary objective of this research is to design and develop a novel cyber-physical infrastructure that can effectively and efficiently transform existing transportation hubs into smart facilities capable of providing better location-aware services. We investigated the integration of a number of novel Internet of Things elements, including video analytics, low-cost Bluetooth beacons, mobile computing, and 3D semantic models, to provide reliable indoor navigation services to people with traveling challenges, yet requiring minimum infrastructure changes since our approach leverages existing cyberinfrastructures such as surveillance cameras, facility models, and mobile phones, and incorporates a minimum number of new and small devices such as beacons to achieve reliable navigation services. The two unique features of our solution are the use of 3D digital semantic models and crowd analysis with surveillance cameras for providing the best available paths. We have started a pilot test with people with disabilities at a multi-floor building in New York City to demonstrate the effectiveness of our proposed framework.

The world is rapidly urbanizing, and for the first time in history over 50% of the world’s population reside in urban areas. This rapid urbanization brings about tremendous challenges at the intersection of governance, infrastructure and the environment. Advanced sensing and data analytics techniques have been developed in the context of so called “smart cities” with the goal of providing insights on how urban systems could be designed and managed more effectively. However, the proliferation of data from heterogeneous sources makes interoperability and mining of such urban data streams difficult. Facilitating the extraction of insights that support data-informed policymaking and program recommendations will require frameworks to integrate such heterogeneous data streams. In this paper, we introduce a novel data integration framework that utilizes an RDF (resource description framework) model to integrate disparate urban data streams based on geo-relationships that are iteratively learned from semantic information and the structure of relational databases. The development of our framework was driven by interviews and observations of city officials responsible for managing and integrating urban data and a review of the various types of disparate datasets generated from sources like departmental databases, sensors, and crowdsourcing. Finally, we apply our proposed framework to an urban data scenario in order to demonstrate the applicability and usefulness of the framework.