I-S2EE Lab News and Updates

Devin Harris served as panelist in a National Academies of Science and Engineering workshop on Digital Twins

Devin Harris participated in a National Academies of Science and Engineering workshop titled: Opportunities and Challenges for Digital Twins in Engineering - A Workshop. The workshop was organized as a panel that included discussion across topics ranging from digital twins in oil and gas engineering, airframe sustainment, transportation, manufacturing, and more. During the workshop speakers discussed the definition of a digital twin and identify current methods, promising practices, and key technical challenges for their development and use. Additional topics of discussion included issues related to uncertainty quantification, data assimilation, and data visualization, as well as opportunities for translation of promising practices to and from other fields. The workshop served as one of three input gathering workshops organized as part of a larger National Academies’ study on research gaps and future directions for digital twins. View workshop here.

Devin Harris gives keynote presentation as part of 4th Annual Research Achievement Awards

Devin Harris delivered the keynote address, “Built to Last:​ The Evolving Role of Infrastructure in our Society ​… the Silent Partner in our Success.” at the fourth annual Research Achievement Awards, held at The Pavilion at the Boar’s Head Resort. The event honored faculty members across Grounds for their outstanding research and scholarship and provided an opportunity to present research on the built environments and the physical places where people live and interact, such as homes, office buildings and streets.

UVA VPR Melur “Ram” Ramasubramanian, Devin Harris, and UVA President Jim Ryan

Devin Harris - "Built to Last"

Devin Harris to Lead Multidisciplinary UVA Team on Digital Twinning Project

The Department of Engineering Systems and Environment’s Devin Harris, a professor of civil engineering, and collaborators from UVA Engineering’s Department of Computer Science and UVA’s School of Education and Human Development have received a new award for $300,000 under the National Science Foundation’s Early-Concept Grants for Exploratory Research Program.

The project, “Adaptive Digital Twinning: An Immersive Visualization Framework for Structural Cyber-Physical Systems,” aims to explore the power of artificial intelligence in the formation of digital twins for large-scale structural systems. Co-principal investigators on the project are assistant professor of computer science Brad Campbell, who holds a secondary appointment in electrical and computer engineering; assistant professor of computer science (teaching track) Panagiotis Apostolellis; and Jennifer Chiu, an associate professor of education. Harris and Campbell are members of UVA Engineering’s Link Lab, a multidisciplinary center for research in cyber-physical systems.

The team’s research will address the need to preserve existing, often aging, physical infrastructure systems on which society relies for essential needs – such as transportation, energy, water and sanitation, and communication – while modernizing these systems to serve as the smart and agile cyber-physical systems we need to meet demands of the future.

The NSF program funding the project, which is known by its acronym EAGER, is designed for untested but potentially transformative research approaches.

Devin Harris part of team awarded NCHRP project – 23-16

Devin Harris is part of a collaborative team that was awarded the NCHRP Project 23-16: Implementing and Leveraging Machine Learning at State Departments of Transportation. The team is a collaboration between EXP U.S. Services Inc. and the University of Virginia (School of Data Science and School of Engineering and Applied Science). Our team is led by Cody Pennetti (SEAS/Dewberry), with contributions by Michael Porter (SDS), Devin Harris (SEAS), Edna Aquilar (EXP), and INCATech.

The objective of this research is to advance the understanding and use of ML tools and techniques at state DOTs and other transportation agencies. The proposed research will aid state DOTs in transitioning to a more advanced state of practice by:

1. Demonstrating the feasibility and practical value of ML in the context of transportation systems, to better understand its application opportunities, implementation processes, and data requirements.
2. Identifying skills, capabilities, resource, and organizational capacities necessary to leverage ML.
3. Identifying and learning from existing applications at transportation agencies.
4. Providing insight into costs, benefits, and performance and limitations considerations.
5. Identifying and sharing ML frameworks, tools, guidance, and ML code for common use cases.

Dizaji paper accepted for publication in Journal of Civil Structural Health Monitoring

Congrats to Mehrdad Shafiei Dizaji for getting our paper accepted into the Journal of Civil Structural Health Monitoring. The paper is titled Full-Field Non-destructive Image-Based Diagnostics of A Structure Using 3D Digital Image Correlation and Laser Scanner Techniques and describes an experimental study on the field deployment of digital image correlation and handheld laser scanning to evaluate in-situ behavior of existing structures. The study evaluated the state of stress and behavior characteristics of a repaired bridge beam end, following the repairs adjacent members that had buckled out of plane during their repair.

Reference:

Shafiei Dizaji, M., Harris, D.K., Kassner, B. and Hill, J.C. (2021 – accepted). “Full-Field Non-destructive Image-Based Diagnostics of A Structure Using 3D Digital Image Correlation and Laser Scanner Techniques”, Springer -Journal of Civil Structural Health Monitoring.

Li, Alipour and Harris paper accepted in Automation in Construction

Congratulations to Tianshu Li for getting her second paper “Mapping Textual Descriptions to Condition Ratings to Assist Bridge Inspection and Condition Assessment Using Hierarchical Attention ” accepted for publication in Automation in Construction (Elsevier). This paper seeks to improve the accuracy and consistency of manually assigned condition ratings of bridges, by leveraging the narrative descriptions from bridge inspection reports as an untapped data source and proposes a data-driven framework to map natural language descriptions to quantitative ratings. A hierarchical architecture employing recurrent neural network encoders with an attention mechanism was developed using a collection of reports from the Virginia Department of Transportation, which outperformed a variety of baseline systems. Visualization of the resulting attention patterns was shown to provide interpretable insights which highlight potentially-overlooked indicators embedded in the narrative descriptions.

Reference:

Li, T., Alipour, M., and Harris, D. K. (2021 – accepted). “Mapping Textual Descriptions to Condition Ratings to Assist Bridge Inspection and Condition Assessment Using Hierarchical Attention”. Elsevier – Automation in Construction.

Li, Alipour, and Harris paper accepted in Advanced Engineering Informatics

Congratulations to Tianshu Li for getting our paper “Context-aware Sequence Labeling for Condition Information Extraction from Historical Bridge Inspection Reports” accepted for publication in Advanced Engineering Informatics (Elsevier). This paper leverages an untapped resource for bridge condition data and proposes a new method to extract condition information from it at a high level of detail. In this work,  a natural language processing approach was developed that analyzes textual descriptions from bridge inspection reports and extracts condition-related information. This work represent a first step in utilizing context derived from the rich, but unused, data inherent to bridge inspection reports to better inform bridge management decisions.

Reference:

Li, T., Alipour, M., and Harris, D. K. (2021 – accepted). “Context-aware Sequence Labeling for Condition Information Extraction from Historical Bridge Inspection Reports”. Elsevier – Advanced Engineering Informatics.

Tianshu Li successfully defends Ph.D. Dissertation

Congratulations to Dr. Tianshu Li for successfully defending her Ph.D. dissertation!!  Tianshu joined the MOBLab in 2017 and has explored a variety of strategies for evaluating infrastructure systems ranging from visual recognition to natural language process. We are thrilled to see her defend her dissertation successfully. Tianshu’s work has created novel approaches to using passive datasets to extract critical information on the performance of infrastructure systems. Looking forward to seeing more of her work published and shared with the academic community.  A special thanks to her committee: Professor Arsalan Heydarian (ESE), Professor Michael Porter (ESE), Professor Vicente Ordóñez Román (CS), and Professor David Lattanzi (GMU-CEIE) for their contributions to her work.

Title: Mining Domain Knowledge from Unstructured Multi-modal Data for Smart Bridge Infrastructure Management

Abstract:
Improving the efficiency of the bridge infrastructure management in the face of critical preservation challenges calls for the integration of automation concepts through the process of inspection, condition assessment, and deterioration prediction. The current experience-driven condition rating process requires extensive effort in training and quality control to ensure the consistency of the assigned ratings. Additionally, the lack of details in the currently available bridge condition database limits the performance of the data-driven models that extract knowledge from past experience to guide decision-making in future maintenance. Meanwhile, the inspection reports generated through the current infrastructure management practices only serve as records of activities, leaving the condition details and domain expertise buried in the reports without being fully exploited for further analysis. To that end, this study identifies visual and textual data from bridge inspection reports as an untapped resource of bridge condition information and mines domain knowledge from a large number of historical inspection reports for automatic condition rating and information extraction. First, to improve the accuracy and consistency of bridge condition rating, a data-driven framework was developed as a supportive tool for two applications: automated condition recommendation and real-time quality control. A hierarchical architecture employing recurrent neural network encoders with an attention mechanism was developed using a collection of reports from the Virginia Department of Transportation. Next, to fully exploit the multi-modal data from bridge inspection reports, a fusion approach is proposed for automated bridge condition rating using the visual and textual data from bridge inspection reports. This study further investigated the uncertainty of rating predictions under random disturbance and showed that and referring the uncertain predictions to human investigations can further improve the rating performance. Lastly, an information extraction (IE) framework is developed to extract bridge conditions from the inspection reports at a high level of detail. Results of this study show that the proposed method can be used to extract and create a condition information database that can further assist in developing data-driven bridge management and condition forecasting models, as well as automated bridge inspection systems. This dissertation is a collection of three manuscripts that describes the aforementioned research works. Through the presented research outcomes, this dissertation highlights the value of unstructured bridge inspection documentation in supporting automated condition assessment and information extraction for the smart bridge infrastructure system.

Tina Tang successfully defends her M.S. thesis

Congratulations to Tina Tang for successfully defending her M.S. thesis titled “Characterizing Shared Mobility Operator and User Behavior Using Big Data Analytics and Machine Learning” on April 22nd, 2021. Tina’s work represented a unique opportunity to explore the passive data derived from E-scooter usage in the Charlottesville community. We are very proud of her work and exploration of this topic and believe that it will be of great value to decision-makers on micro-mobility solutions in the Charlottesville area and beyond. A special thanks to her committee members: Brian Smith (Engineering Systems and Environment) and Andrew Mondschein (Urban Planning) for their contributions.

Title: Characterizing Shared Mobility Operator and User Behavior using Big Data Analytics and Machine Learning

Abstract: Towards more sustainable use of resources in cities, there is a rising trend in shared mobility for collaborative consumption. As a condition of working with cities, third party organizations managing shared vehicle fleets often have to provide public access to real-time data describing the location of vehicles. These datasets hold enormous value for monitoring and evaluating emerging transportation services; however, a major challenge for city planners and regulators remains extracting the value from streaming transportation data by leveraging analysis and visualization methods. E-scooters are an emerging shared mobility service that have been adopted in cities across the world, but, despite their popularity, cities are still searching for more effective monitoring methods in order to understand the benefits brought to their communities or lack thereof. Using real-time e-scooter data from Charlottesville, Virginia as a case study, this work aims to characterize operator and user behavior by using big data analytics and machine learning to extract important insights. Specifically, this work provides the following contributions via three analytical studies: (1) Study I demonstrates how e-scooter data can be harvested from streaming GPS traces and then aggregated and spatially joined with demographic, employment, and built environment data. A multiple regression analysis examining the relationships between these datasets revealed that e-scooter distribution was influenced by economic activity whereas e-scooter use was influenced by micro-transit need factors and built environment characteristics. (2) Study II presents data aggregation and visualization approaches for monitoring and evaluating e-scooter operator distribution decisions, showing that utilization is a suitable measure for planning and revealing that there is room for improvement for equitable fleet distribution. (3) Study III shows the efficacy of using Latent Dirichlet Allocation to characterize user trip behavior from an unstructured set of estimated e-scooter trips. Findings suggest that trip behavior differed significantly during periods with increased student population influxes. Charlottesville planners and regulators may use the results and methods presented in this work to make data-driven decisions for improving micro-mobility as a service for the community they serve.

Dizaji, Alipour and Harris paper accepted into Experimental Techniques

Congratulations to Mehrdad Shafiei Dizaji for getting our paper accepted for publication in Experimental Techniques (Springer). This paper presents a novel structural identification approach to localize unseen damage using measurements from full-field three dimensional digital image correlation within (3D-DIC) in collaboration with the finite element method. The approach provides a non-contact approach to identity unseen damage in structural components.

Reference:

Shafiei Dizaji, M., Alipour, M., and Harris, D. K. (2021 – accepted). “Image-based Tomography of Structures to Detect Internal Abnormalities Using Inverse Approach.” Springer – Experimental Techniques.