Search

The purpose of this paper is to examine the opportunities and barriers of Integrated Human-Machine Intelligence (IHMI) in civil engineering. Integrating artificial intelligence's high efficiency and repeatability with humans' adaptability in various contexts can advance timely and reliable decision-making during civil engineering projects and emergencies. Successful cases in other domains, such as biomedical science, healthcare, and transportation, showed the potential of IHMI in data-driven, knowledge-based decision-making in numerous civil engineering applications. However, whether the industry and academia are ready to embrace the era of IHMI and maximize its benefit to the industry is still questionable due to several knowledge gaps. This paper thus calls for future studies in exploring the value, method, and challenges of applying IHMI in civil engineering. Our systematic review of the literature and motivating cases has identified four knowledge gaps in achieving effective IHMI in civil engineering. First, it is unknown what types of tasks in the civil engineering domain can be assisted by AI and to what extent. Second, the interface between human and AI in civil engineering-related tasks need more precise and formal definition. Third, the barriers that impede collecting detailed behavioral data from humans and contextual environments deserve systematic classification and prototyping. Lastly, it is unknown what expected and unexpected impacts will IHMI have on the AEC industry and entrepreneurship. Analyzing these knowledge gaps led to a list of identified research questions. This paper will lay the foundation for identifying relevant studies to form a research roadmap to address the four knowledge gaps identified.

Construction inspection is a critical component of INDOT’s quality assurance (QA) program. Upon receiving an inspection notice/assignment, INDOT inspectors review the plans and specifications to identify the construction quality requirements and conduct their inspections accordingly. This manual approach to gathering inspection requirements from textual documents is time-consuming, subjective, and error-prone. This project addresses this critical issue by developing an inspection requirements database along with a set of tools to automatically gather the inspection requirements and provide field crews with customized construction checklists during the inspection with the specifics of what to check, when to check, and how to check, as well as the risks and the actions to take when noncompliance is encountered. This newly developed toolset eliminates the manual effort required to acquire construction requirements, which will enhance the efficiency of the construction inspection process at INDOT. It also enables the incorporation of field-collected data to automate future compliance checking and facilitate construction documentation.

Acceptance testing is a critical aspect of the quality control and quality assurance (QC/QA) program to ensure the reliable long-term performance of pavement. A typical acceptance testing specification includes acceptable quality characteristics (AQCs), testing methods, number of samples, sample locations, and acceptance criteria. In the current practice, Indiana Department of Transportation (INDOT) accepts pavement by sampling and testing materials with a pre-determined, very low frequency at random locations, leading to a significant problem: testing results are not truly “representative” of the project because sampling is neither based on a statistical foundation, nor on the reliability concept. This study developed a systematic guideline that has addressed the aforementioned problem of material acceptance testing in four aspects: identifying key material properties for testing, selecting sample locations, designing acceptance criteria, and determining optimal sample size. Key material properties that are critical to the pavement long-term performance are identified by comparing with sensitive material properties in MEPDG. A random sampling mechanism was devised based on two spatial indices to control the spatial pattern of samples to minimize the influence from spatial autocorrelation. Risk-based acceptance criteria was proposed based on statistical methods to control the agency’s risk at a desired level given a specific sampling and testing strategy, based on which optimal sample size is determined from a risk perspective. Cost analysis approaches were developed to estimate the total cost of acceptance testing by integrating the risk of making incorrect decision and enable the determination of optimal sample size from a cost perspective. Additionally, quality control chart was exploited as a complementary tool to ensure the consistency of the pavement quality of a project. The results of this study were validated using real data from INDOT projects, and a web t

Construction inspection is a critical component in the quality assurance (QA) program to ensure the quality and long-term performance of pavements. Over the years, INDOT has been developing and modifying its standard specification to set requirements for construction inspection and material testing. With the retirement of experienced employees, INDOT is challenged with the lack of knowledge to effectively inspect the critical elements of construction results/deliverables such as pavement, soil embankment, and bridge (decks). There is a critical need for INDOT to allocate limited resources to the riskiest areas and equip construction inspectors with necessary knowledge to conduct inspection, ensure the quality of construction results, and minimize risks to INDOT. This study developed a risk-based inspection guide that has addressed the aforementioned problems of shortage in staffing and loss and lack of knowledge by providing answers in aspects of what, when, how, and how often to inspect. A comprehensive list of testing and inspection activities were extracted from INDOT’s material testing manual, INDOT’s standard specification, and the QA implementation at the Ohio River Bridge (ORB) project. This list was narrowed down to a core set of items based on survey responses and interviews with INDOT domain experts. Testing and inspection activities in the core set were aligned with the construction process. The risk associated with each inspection activity was assessed by considering both the probability of failure and consequence severity of failure in four dimensions: cost, time, quality, and safety. A composite risk index was developed as a single measure for the overall risk. All inspection activities were prioritized based on the composite index. For implementation, a linking mechanism was developed to link inspection activity, pay item, and check items (extracted from specification). This linking mechanism aligns with the business process of construction inspection

This report describes a study of intelligent compaction (IC) technologies, within the context of actual construction projects, for its potential as a component of INDOT’s QC/QA for soils. The output from an IC-equipped roller compaction equipment is a real-time area mapping of the compacted lift stiffness as captured by the IC measure. Data was collected to evaluate the correlation between each of two IC measures—compaction meter value (CMV) and machine drive power (MDP)—and in situ embankment quality test measures, the chief in situ test being the dynamic cone penetrometer (DCP) test which INDOT uses for soil embankment acceptance testing. Researchers sought to understand how well the IC measures might assess embankment quality as currently evaluated by the in situ measures. Window-averaged IC measures were compared with the in situ DCP test points. For CMV, a variable correlation was found between the average CMV and DCP values from 74 in situ locations. Also, a limited head-to-head comparison of CMV and MDP with the in situ measures provided some indication that MDP should be studied further. Lessons were learned regarding the elimination of bias in future correlation studies, critical provisions to facilitate best data quality, and important aspects of data management. IC technology holds promise for monitoring the consistency of the soil compaction effort and flagging weak areas in real time during compaction operations. However, further insight is needed regarding the correlation of the DCP measure with both types of IC measures for various soil characterizations and field moisture conditions.

This study was in two parts. The first part established and demonstrated a framework for pavement data integration. This is critical for fulfilling QC/QA needs of INDOT’s pavement management system, because the precision of the physical location references is a prerequisite for the reliable collection and interpretation of pavement data. Such consistency is often jeopardized because the data are collected at different years, and are affected by changes in the vendor, inventory, or referencing system or reference points. This study therefore developed a “lining-up” methodology to address this issue. The applicability of the developed methodology was demonstrated using 2012-2014 data from Indiana’s highway network. The results showed that the errors in the unlined up data are significant as they mischaracterize the true pavement condition. This could lead to the reporting of unreliable information of road network condition to the decision makers, ultimately leading to inappropriate condition assessments and prescriptions. Benefits of the methodology reverberate throughout the management functions and processes associated with highway pavements in Indiana, including pavement performance modeling, optimal timing of maintenance, rehabilitation, and reconstruction (MRR), and assessment of the effectiveness of MRR treatments and schedules. The second part of the study developed correlations for the different types of pavement distresses using machine learning algorithms. That way, the severity of any one type of distress can be estimated based on known severity of other distresses at that location. The 2012-2014 data were from I-70, US-41, and US-52, and the distress types considered are cracking, rutting, faulting, and roughness. Models were developed to relate surface roughness (IRI) to pavement cracks, and between the different crack types, with resulting degrees of confidence that varied across the different crack types and road functional classes. In addition, for eac

State Transportation Agencies (STAs) use quality control/quality assurance (QC/QA) specifications to guide the testing and inspection of road pavement construction. Although failed materials of pavement rarely occur in practice, it is critical to have a sound decision framework to assist in making data-driven, informed decisions regarding failed materials because such decisions have profound impacts on the long-term performance of the pavement and the operation and maintenance costs of the responsible highway agencies. A performance-related specification (PRS) is a quality acceptance (QA) specification that specifies the acceptable levels of key acceptance quality characteristics (AQCs) that are directly related to fundamental engineering properties, which in turn, determine the long-term performance of the constructed end products. Two PRS tools, PaveSpec for Portland Cement Concrete Pavement (PCCP) and Quality Related Specification Software (QRSS) for QC/QA Hot Mixed Asphalt (HMA) pavement, were investigated in this study to develop decision frameworks for PCCP and HMA pavement to assist the decision-making regarding failed materials at INDOT. A large number of simulations of various scenarios in the context of INDOT pavement construction were conducted to fully develop and implement the decision framework. For PCCP, the newly developed decision framework based on PaveSpec was validated using data from an INDOT construction project. The framework is readily implementable to assist INDOT in making informed decision regarding failed materials for PCCP. For QC/QA pavement, it was found that QRSS is not an appropriate PRS tool to estimate the long-term performance because of its limitations, the misalignment between QRSS process and INDOT practice, and erroneous simulation results.

Accurate and complete construction records and as-built data are the key prerequisites to the effective management of transportation infrastructure assets throughout their life cycle. The construction phase is the best time to collect such data. Assets such as underground drainage and culverts are visible and physically accessible only during construction. For assets such as guardrails, signals, and pavement, it is safer and more efficient to collect data during construction than after construction when the road segment is open to traffic. The purpose of this project was to conduct a synthesis study to 1) assess the current status at INDOT regarding the collection of asset data during the construction phase and the use of such data in the operation and maintenance (O&M) phase, and 2) develop a framework for INDOT to leverage the construction inspection and documentation process to collect data for assets. Data needs during O&M were identified through rounds of meetings with relevant INDOT business units. The current practice in construction documentation was investigated in detail. A survey of state highway agencies (SHAs) was conducted to assess the state-of-the-practice. A practical framework was developed to leverage the construction inspection and documentation practice to collect asset data that are needed in O&M. The framework uses specific pay items—construction activities that result in physical structures—as the bridge to connect plan assets (i.e. physical structures specified in the design documents) to their corresponding counterparts in the asset management systems. The framework is composed of 1) a data needs component for determining the information requirements from the O&M perspective, 2) a construction documentation module, and 3) a mapping mechanism to link data items to be collected during the construction documentation to data items in the asset management systems. The mapping mechanism was tested and validated using four priority asset classes—u

SPR No. 750, Construction inspection is a critical component of the quality assurance/quality control (QA/QC) program at the South Carolina Department of Transportation (SCDOT). A pressing challenge SCDOT has been facing is to balance the increasing demand to rebuild the statewide transportation systems with declining resources available for inspection. This project aims to address this critical issue by developing a risk-based inspection program that includes (1) the identification and assessment of risks associated with construction activities at SCDOT, (2) the development of inspection strategies that correspond to the level of risk, (3) the identification of quality requirements and inspection activities that are applicable to pay items, and (4) the design and implementation of a digital, risk-based inspection system. Two rounds of surveys were conducted to first identify and rank the risk factors and then assess the risks in aspects of likelihood, severity (cost, time, safety, and quality), and performance (short-/mid-/long-term). An aggregate risk score was calculated to categorize each pay item into high, medium, and low levels of risk. The inspection strategies (i.e., inspection priority and frequency and documentation frequency) were designed to correspond to the level of risk. The quality requirements and inspection activities for pay items were extracted from the corresponding sections in the standard specification, construction manual, supplementary specifications, supplementary technical specifications, and standard drawings. A database was designed to host the risk and inspection information. Tools were developed to assist project engineers and field inspectors in retrieving inspection activities in the digital form of checklists. The main deliverable is the resulting digital inspection system with tools. This system allows for the retrieval of all the related risk information, inspection strategy, and inspection activities/checklists for any given pay item. This s

SPR-3707, Accurate and complete construction records and as‐built data are the key prerequisites to the effective management of transportation, infrastructure assets throughout their life cycle. The construction phase is the best time to collect such data. Assets such as underground, drainage and culverts are visible and physically accessible only during construction. For assets such as guardrails, signals, and pavement, it is safer and more efficient to collect data during construction than after construction when the road segment is open to traffic., The purpose of this project was to conduct a synthesis study to 1) assess the current status at INDOT regarding the collection of asset, data during the construction phase and the use of such data in the operation and maintenance (O&M) phase, and 2) develop a, framework for INDOT to leverage the construction inspection and documentation process to collect data for assets. Data needs during, O&M were identified through rounds of meetings with relevant INDOT business units. The current practice in construction, documentation was investigated in detail. A survey of state highway agencies (SHAs) was conducted to assess the state‐of‐the‐practice., A practical framework was developed to leverage the construction inspection and documentation practice to collect asset data that are, needed in O&M. The framework uses specific pay items—construction activities that result in physical structures—as the bridge to, connect plan assets (i.e. physical structures specified in the design documents) to their corresponding counterparts in the asset, management systems. The framework is composed of 1) a data needs component for determining the information requirements from, the O&M perspective, 2) a construction documentation module, and 3) a mapping mechanism to link data items to be collected during, the construction documentation to data items in the asset management systems. The mapping mechanism was tested and validated, using four priority asset classes—underdrains, guardrails, attenuators, and small culverts—from an INDOT construction project. The, testing results show that the newly developed framework is viable and solid to collect asset data during the construction phase for O&M, use in the future, without adding extra workload to construction crews. The framework can reduce/eliminate the duplicate data, collection efforts at INDOT, leading to savings and efficiency gains in the long term.

SPR-3805, State Transportation Agencies (STAs) use quality control/quality assurance (QC/QA) specifications to guide the testing and inspection of, road pavement construction. Although failed materials of pavement rarely occur in practice, it is critical to have a sound decision, framework to assist in making data-driven, informed decisions regarding failed materials because such decisions have profound impacts, on the long-term performance of the pavement and the operation and maintenance costs of the responsible highway agencies., A performance-related specification (PRS) is a quality acceptance (QA) specification that specifies the acceptable levels of key acceptance, quality characteristics (AQCs) that are directly related to fundamental engineering properties, which in turn, determine the long-term, performance of the constructed end products. Two PRS tools, PaveSpec for Portland Cement Concrete Pavement (PCCP) and Quality, Related Specification Software (QRSS) for QC/QA Hot Mixed Asphalt (HMA) pavement, were investigated in this study to develop, decision frameworks for PCCP and HMA pavement to assist the decision-making regarding failed materials at INDOT. A large number of, simulations of various scenarios in the context of INDOT pavement construction were conducted to fully develop and implement the, decision framework., For PCCP, the newly developed decision framework based on PaveSpec was validated using data from an INDOT construction project., The framework is readily implementable to assist INDOT in making informed decision regarding failed materials for PCCP. For QC/QA, pavement, it was found that QRSS is not an appropriate PRS tool to estimate the long-term performance because of its limitations, the, misalignment between QRSS process and INDOT practice, and erroneous simulation results.