Journal of Computing in Civil Engineering

As facility owners and operators, public sector organizations (such as government agencies) that provide public facilities and services not only maintain the existing capital assets but also plan for future developments on an ongoing basis through a capital improvement program (CIP). The CIP identifies capital projects and is typically funded through several distinct funding sources with specific usage, timing, and reporting restrictions. The continuous nature of capital improvement planning compounded by the potential for scope or funding availability changes further makes it a challenge to retain the flexibility to add new and different funding sources to an existing funding allocation plan. Existing literature about this problem is sparse, and a systemic solution is yet to be proposed that enables an owner organization to meet its goals while maintaining fiscal responsibility toward its funding agencies. The financial digital twin (FinDT) framework developed in this study integrates fund allocation and accounting with project acquisition in real time and enables the owner to automate funding allocation tasks and to synchronize change management while ensuring compliance with funding restrictions. The goal of this framework is to leverage rule-based knowledge management to consolidate and continually optimize the efficient use of accurate financial information at all stages of the capital asset lifecycle. This is accomplished through automation and digital twin (DT) integration. A use case was developed to demonstrate and evaluate the FinDT, which was found to hold potential for more efficient information management and support for decision making through enterprise-level data integration. The framework developed in this study can be used to display other nongeometric attribute information in a building information model for the purposes of visualization and interactive, efficient decision making.

Accurate and timely assessment of pavement condition is critical to determine optimal maintenance plans. Due to the high costs of ground-based inspections, agencies often limit their monitoring to major roads and the condition of some elements of the road network remains unknown. Satellites, capable of rapidly collecting information over wide areas, can be a cost-effective alternative to monitor pavement condition. This wide coverage, however, comes at the expense of lower levels of accuracy. The objective of this study is to quantify the value of satellite-based information in optimal inspection and maintenance strategies. To account for the uncertainties associated with satellite observations, the system was modeled as a partially observable Markov decision process (POMDP) to determine optimal life-cycle inspection and maintenance policies. To estimate the value of information obtained from satellite inspections, two cases representing current pavement condition practices were simulated: (1) as an alternative to inspect highways, roads that are traditionally monitored with annual automated distress surveys, and (2) as an option to inspect local or ancillary roads, which are not typically monitored. Results indicate that satellite observations result in up to 6.5% reduction in cost if they are used to make maintenance and inspection decisions over the pavement life cycle. Savings are higher for nonmonitored roads compared to major roads that are annually inspected with automated distress surveys. Satellite information was found to become valuable at 70% level of accuracy when used in combination with more accurate systems.

Sustainability assessment tools have been widely used in the building sector to evaluate the performance of buildings. However, they do not account for the dynamic internal and external building changes that occur with time. Furthermore, these tools adopt a discrete approach by ignoring the trade-offs among the pillars of sustainability. Hence, sustainability assessment should progress toward defining the building’s performance against achievable or predefined sustainability targets. This process, commonly known as benchmarking, has found limited applications in the building sector due to the lack of standard methods and internationally comparable data inventory. The systems thinking approach is ideal for handling these challenges associated with building sustainability evaluation and benchmarking owing to its potential to handle different subcomponents and their dynamic interdependencies. Further, benchmarking requires large data sets and involves significant testing and experimenting with numerous scenarios. Hence, a viable solution is to model and simulate the building’s sustainability while integrating various scenario interventions to define standard benchmarks. Therefore, this study proposes a novel sustainability assessment and benchmarking (SAB) framework for the building sector using system dynamics (a branch of systems thinking) modeling and simulation approach along with multicriteria decision making (MCDM) methods. The SAB framework is demonstrated using the case of the affordable housing segment in India, for which a sustainability benchmarking scale is developed. The main findings of the case study show that renewable energy interventions above 80% are an efficient way to improve a building’s sustainability to the best performance levels. Additionally, the net-zero energy policy is not the only way to reach the highest level of sustainability because a combination of material, water, and energy interventions could also lead to improved sustainability performance. Builders, government agencies, and urban planners worldwide can adopt the SAB framework to define benchmarks and implement policies for improving the sustainability of buildings.

Rapid reconnaissance of building damage is critical for disaster response and recovery. Drones have been utilized to collect aerial images of affected areas in order to assess building damage. However, there are two challenges. First, processing many aerial images to detect and classify building damage based on a consistent standard remains laborious and complex, necessitating a new automated solution to achieve accurate building damage detection and classification. Second, drone operations during disaster response rely primarily on human operators’ experience and seldom use the obtained building damage information to optimize drone mission planning. Therefore, this study proposes a new method, which automates building damage reconnaissance with drone mission planning for disaster response operations. Specifically, a deep learning method is developed to detect and classify building damages using a newly labeled dataset consisting of 24,496 distinct instances of building damage. This deep learning method is validated, achieving 71.9% mean average precision. In addition, building damage information is modeled and integrated into mission planning, in order to optimize drones’ task assignments and route calculations. A tornado disaster in Tennessee is used as a case study to quantitatively evaluate this methodology. The present study concludes that optimal drone mission planning during disaster response can be augmented using accurate building damage information acquired from deep learning methods.

Abstract

Transportation agency personnel gain valuable knowledge through their work, but such knowledge is lost if it is not documented properly after the worker leaves the organization. The risk of losing institutional knowledge is a current problem at state departments of transportation, including the North Carolina Department of Transportation (NCDOT), due to high personnel turnover. State transportation agencies have implemented knowledge repositories in the form of lessons learned/best practices databases to address this problem. However, motivating end-users to use such databases is challenging. This paper addresses this challenge through novel artificial intelligence technology whereby a neural network–based language model is implemented as part of the NCDOT’s new knowledge management program: Communicate Lessons, Exchange Advice, Record (CLEAR). The CLEAR program encompasses a database of lessons learned/best practices and a website to access and search the database. The developed methodology involves training a language model on transportation construction texts and using that trained model in a novel algorithm enabling users to search the CLEAR database easily. The developed language-processing model provides an easily accessible interface to suggest the most relevant CLEAR data based on the end-user’s searched keywords. The model learns an inference model of construction domain–specific vocabulary extracted from various sources, such as contract documents, textbooks, and specifications, to make meaningful connections between lessons learned/best practices in the CLEAR database and project-specific knowledge. The developed model has been validated by project managers for projects at various life cycle stages. The automation of information retrieval is intended to encourage NCDOT personnel to use and embrace the CLEAR program as part of their routine work to improve project workflow. In the long run, the NCDOT will benefit from consistent usage of the CLEAR program and its high quality content, thereby leading to enhanced institutional knowledge and organizational innovation.

Practical Applications

The construction industry, with a particular emphasis on transportation construction, currently faces tremendous challenges in retaining and retraining existing personnel to ensure business continuity on projects. Knowledge gained on projects by project personnel can be lost forever if not properly documented. While knowledge repositories are effective toward ensuring the storing and retrieving of past knowledge, extant literature underlines the need to ensure continued participation by the end-users for the success of such repositories. This research effort uses natural language processing, a subfield artificial intelligence that deals specifically with text sources, as a means to quickly and accurately enhance the quality of search results being displayed to the end-users within the North Carolina Department of Transportation’s recently commissioned knowledge management program called CLEAR. As a result, end-users can stay motivated and embrace the CLEAR program, thereby ensuring its long-term success. In the long run, the consistent usage of the CLEAR program and the high quality content that is input to the CLEAR database by the NCDOT end-users will lead to enhanced institutional knowledge and internal organizational innovation.

To prepare graduates for the dynamic and complex working environment in the architecture, engineering, and construction (AEC) industry, educational institutions have distinctly emphasized the significance of communication skills. Field trips are a widely utilized strategy to provide students with a situated learning setting where students can communicate with professionals onsite. However, several challenges limit the application of this strategy, and virtual field trips become a potential and promising supplement or alternative approach. This study discusses the pedagogical use of a virtual field trip experience augmented with conversational virtual humans—iVisit-Communicate—to produce realistic field trips for students to practice their communication skills. This virtual experience is supported by immersive computing (i.e., 360-degree virtual space) and conversational virtual human technologies. iVisit-Communicate simulates two types of communication to provide student-professional verbal communication opportunities: guided scenario and conversational scenario. The paper provides a detailed description of the technical and pedagogical development of iVisit-Communicate components followed by a student-centered experiment, focusing on measures such as system usability, sense of presence, and communication skills of students. The contribution of this study is to provide an in-depth understanding of the significance of integrating communication opportunities in virtual field trips and illustrate how such on-the-site opportunities successfully facilitate students’ communication practices.

This paper presents a data set collected periodically on a construction site. The data set aims to evaluate the performance of simultaneous localization and mapping (SLAM) algorithms used by mobile scanners or autonomous robots. It includes ground-truth scans of a construction site collected using a terrestrial laser scanner along with five sequences of spatially registered and time-synchronized images, lidar scans, and inertial data coming from our prototypical handheld scanner. We also recover the ground-truth trajectory of the mobile scanner by registering the sequential lidar scans to the ground-truth scans and show how to use a popular software package to measure the accuracy of SLAM algorithms against our trajectory automatically. To the best of our knowledge, this is the first publicly accessible data set consisting of periodically collected sequential data on a construction site.