Georgia Department Of Transportation Research Articles

Ensemble Machine Learning Classification Models for Predicting Pavement Condition

Forecasting pavement performance condition is essential within the pavement management system to optimize decisions with regard to planning maintenance and rehabilitation projects. Accurate forecasts facilitate timely interventions and assist in formulating cost-effective asset management plans. Data-driven machine learning models that utilize historical data to improve forecasting precision have gained attention in the field of asset management. Although numerous studies have employed regression-based models to forecast pavement condition, transportation asset management often operates according to condition index ranges rather than exact values. Therefore, classification models are suitable for predicting pavement condition grades and determining the appropriate maintenance type for pavement assets. This research focuses on developing five machine learning classification models to predict pavement condition: random forest; gradient boost; support vector machine; k-nearest neighbors; and artificial neural network. To enhance prediction performance, these models are integrated using ensemble methods, including voting and stacking. The classification models are developed using a dataset from the Georgia Department of Transportation that documented the condition of asphalt pavements for predefined maintenance sections between 2017 and 2021. A voting ensemble model constructed with the two best-performing individual classification models reached the highest accuracy rate at 83%. Although the performance of individual models fluctuates, ensemble models consistently produce a top-tier performance regardless of the variations in data sampling. Therefore, ensemble methods are recommended for developing pavement condition prediction models to improve accuracy and achieve a more consistent quality of predictions. The findings of this research will provide transportation agencies with information to help them strengthen their forecasting practices in relation to pavement condition, thereby improving their maintenance planning and cost savings.

Assessing large-scale roadside tree removal using aerial imagery and crash analysis: A difference-in-differences approach

Transport agencies worldwide remove roadside trees in the name of traffic safety. In 2017, the Georgia Department of Transportation (GDOT), the agency managing state and federal highways in the US state of Georgia, began clear-cutting trees on all transport agency property statewide, amounting to thousands of hectares of cleared land. As justification, GDOT cited fatal traffic crashes with trees.In this paper, we quantified the extent of tree removal within a 60 m buffer along the five major interstates in Georgia, using aerial imagery from the United States Department of Agriculture (USDA) Farm Service Agency’s National Agricultural Imagery Project (NAIP). We used supervised image classification and change detection on aerial imagery between 2015 and 2021. We also collected pre- and post-treatment crash data and conducted a difference-in-differences analysis on sampled road segments to isolate the effect of tree removal on crash rates.Results showed that 28.4 sq km of previously forested land in the interstate corridors were cleared of trees, 25% of the total forested land in the corridors. Tree crash fatalities did not decrease.Findings provide insight into approaches to roadside safety like tree removal while helping assess impacts and plan for next steps. The lack of evidence for reduced tree crash fatalities indicate that transport agencies should consider measures other than large-scale tree removal to prevent fatal crashes with trees.

Environmental and Economic Efficiency of Stone-Matrix Asphalt with Surface Micro-Milling

The Georgia Department of Transportation (GDOT) has employed a combination of open-graded friction course (OGFC) and stone-matrix asphalt (SMA) layers on its Interstate highways for more than two decades. Given that the SMA layer was structurally intact, GDOT implemented micro-milling to solely remove the OGFC layer. Four conventional resurfacing cycles were considered to compare the environmental and economic impacts for a section of Interstate 95 using life-cycle assessment and life-cycle cost analysis, assuming 8-, 10-, 12-, and 14-year resurfacing intervals of asphalt concrete (AC) resurfacing. Within the pavement life cycle, the maintenance and rehabilitation (M&R), and the use stages were significantly influenced by the change of scenarios. The M&R stage revealed a twofold cost increase for the full-depth AC, as compared with SMA, over the pavement life, because of the latter’s relative improved structural performance and a delay in future rehabilitation. Furthermore, in the use stage, M&R affected the agency costs significantly, whereas the work-zone delay governed the user cost. The holistic evaluation of the SMA and micro-milling combination revealed positive outcomes in field performance, along with environmental and economic impacts. However, it is worth noting that the scenarios presented follow GDOT’s approach with a long-standing use of SMA and micro-milling. Given the extensive experience with the aforementioned combination, it was evident that the method provided significant benefits, especially in the long-term outlook, and may be applicable for locations that have conditions analogous to GDOT’s.

Evaluating Impact of Corrected Optimum Asphalt Content and Benchmarking Cracking Resistance of Georgia Mixtures for Balanced Mix Design Implementation

The Georgia Department of Transportation (GDOT) implemented the corrected optimum asphalt content (COAC) approach to reducing the allowable reclaimed asphalt pavement (RAP) binder contribution for improving mixture durability. The COAC of 60:40 implemented in 2019 allows 60% of the RAP binder to be credited to the total mix, requiring additional virgin asphalt binder equivalent to 40% of the RAP binder to be added to the volumetric optimum binder content. To further enhance the durability of asphalt mixtures, GDOT plans to implement the indirect tensile asphalt cracking test (IDEAL-CT) in the balanced mix design (BMD) method in conjunction with the Hamburg wheel tracking test that is currently specified in the GDOT specifications. To assist GDOT in the implementation effort, this study was conducted to (1) evaluate the effect of the COAC approach on the cracking and rutting resistance of asphalt mixtures using the proposed BMD test methods; and (2) benchmark the cracking resistance (i.e., CTindex) of plant-produced mixtures being produced in the state of Georgia for developing CTindex thresholds for future implementation. The results of this study suggest that the 60:40 COAC helps improve the cracking resistance of asphalt mixtures containing high RAP contents with minimum effect on their rutting resistance when compared with the corresponding virgin mixtures. In addition, because of statistical differences in the CTindex results between stone mastic asphalt (SMA), surface, and intermediate/base mixtures tested in this study, different CTindex thresholds are proposed for these three mixture groups for pilot projects. Further research is anticipated to refine these thresholds and verify with field performance for future implementation in mix design approval and acceptance testing.

Green rules and green tape: Streamlining the environmental review for transportation projects

• Streamlining rules and procedures using green tape improves performance. • Programmatic Categorical Exclusions create substantial time savings for infrastructure project delivery. • Streamlining project sub-disciplines creates performance gains for the entire project. • Green tape theory can be applied to diverse project teams in the outsourced environment. The Georgia Department of Transportation (GDOT) implemented a programmatic agreement as green tape to streamline environmental review and decrease project durations. We investigate the implementation and outcomes associated with this streamlining initiative to examine whether it functions as green tape and investigate whether it improves performance by reducing project durations. We use a mixed-methods approach, using qualitative analysis to explore green tape perceptions of project team members and a quasi-experimental, instrumental variable approach to estimate the time-savings the created for a sample of 397 infrastructure projects. Our results reveal the streamlining initiative operates as green tape and demonstrate the substantial performance gains green tape can yield. These findings suggest that streamlining one project discipline impacts performance across the entire project, creating substantial time savings. Furthermore, they demonstrate how green tape functions in diverse project teams and can be applied in public agencies to create performance gains in infrastructure project delivery.

Performance-Based and Evidence-Based Approach to Research Implementation at the Georgia Department of Transportation

This paper discusses the Georgia Department of Transportation’s (GDOT’s) performance-based and evidence-based approach to research implementation. Transportation agencies in the United States spend hundreds of millions of dollars on research, development, and technology transfer annually. From a performance-based standpoint, agencies will realize higher returns on investment and higher impacts of their research programs as research is implemented more effectively and efficiently. From an evidence-based standpoint, requesting evidence of research implementation as implementation deliverables from the outset of the project requires the project research team and other staff to think through and incorporate in the project plan explicit ways in which the research will be implemented. GDOT’s performance-based approach to research program management treats research implementation as part of an overall asset management business process. This process integrates technical, human, organizational, and external resources to encourage, track, and monitor research implementation activities toward achieving agency strategic objectives, using an evidence-based approach. The paper discusses the adoption of a performance-based and evidence-based process, and a research implementation management tool, and their application in the development of the fiscal year 2018 Annual Research Implementation Report as well as its impact within and beyond the agency. This paper is potentially useful to transportation practitioners and agencies that want to adopt a performance-based and evidence-based approach to augment return on research investment.

Environmental impact comparison of new micro-milling and thin overlay and conventional milling for sustainable pavement preservation

Micro-milling and thin overlay, a new pavement preservation treatment developed and first implemented by the Georgia Department of Transportation (GDOT) in 2007, as an alternative pavement preservation to conventional milling and resurfacing, is used to remove and replace a thin, deteriorated open-graded surface layer (2.2 cm) without disturbing the sound underlying layer. This paper focuses on quantifying micro-milling and thin overlay’s environmental impacts by comparing conventional milling and overlay with micro-milling and thin overlay to complement the previous economic impact study. The environmental impacts are quantified using a pavement construction decision tool entitled Pavement Life-Cycle Assessment Tool for Environmental and Economic Effects (PaLATE). The results show a 60% reduction in all environmental impacts assessed, including energy usage, water consumption, and CO2 emissions largely due to the savings in asphalt needed. Results have demonstrated that this new method is a promising pavement preservation alternative that is environmentally sustainable. This paper provides quantitative comparison information useful for transportation agencies in selecting sustainable pavement preservation alternatives.

Economic impact analysis of GDOT short line railroad infrastructure investment in Georgia

Short line railroads are critical to Georgia's transportation infrastructure and support the efficient movement of goods into and out of the State. Twenty-nine short line railroads operate in Georgia, six of which operate either partially or totally over 596 miles of rail track that the Georgia Department of Transportation (GDOT) owns. Based on the Bureau of Economic Analysis Regional Input-Output Modelling System (RIMS II) backward-linkage model, this study estimates the economic impacts associated with short line railroad revenues and with GDOT infrastructure spending on these six short line systems under different sets of assumptions. Among the economic impacts, and depending on the assumptions, the study finds that cumulatively, the six systems increase annual output in the range of $2.8 - $14.5 million, increase annual earnings in the range of $0.7 - $4.1 million, generate annual value-added in the range of $1.4 - $7.5 million, and annually add 10–93 jobs on average.

Analysis of timber transportation accident frequency, location, and contributing factors in Georgia, USA 2006-2016

ABSTRACTIn the US South, the overwhelming majority of timber is transported from forests to mills by tractor-trailers. Therefore, safe and efficient timber transportation is critical for wood supply chain efficiency. Vehicle accident reports (DOT-523) were collected from the Georgia Department of Transportation for all accidents involving logging vehicles and other heavy vehicles for the years 2006–2016. Accident location, frequency, contributing factors, and vehicle condition were analyzed for the periods 2006–2009, 2010–2012, and 2013–2016. Logging vehicle accidents outside the Atlanta metro declined by nearly 70% from 2006 to 2013, but increased by 14% from 2014 to 2016. Accidents per million tonnes of timber hauled in the state declined from 11 in 2006–2009 to 4.2 in 2013–2016 (p < 0.01). The most common contributing factors to logging vehicle accidents were “following too close” in 2006–2009 and “misjudged clearance” in 2010–2012 and 2013–2016. Logging vehicles involved in accidents were older and in worse mechanical condition compared to other heavy vehicles. Since 2010, more than half of all logging vehicle accidents have occurred in urban areas. An average of fewer than 5% of accidents occurred on Interstate Highways, but 41% of accidents occurred within 8 km of an Interstate Highway 2013–2016. Overall, timber transportation safety improved between 2006 and 2016, but opportunities exist for further improvement. Reducing average age of logging vehicles, proactive maintenance, and investing in technology such as onboard cameras and GPS tracking may reduce logging vehicle accidents and improve timber transportation efficiency.

Leveraging State DOT Technology and Practice for Cost-Effective Pavement Management by Local Government Authorities

To improve management of the pavement assets in Georgia’s local government authorities (LGs), there is a need to leverage the existing pavement management and maintenance technology and practice in the Georgia Department of Transportation (GDOT). However, due to the difference in organizational structure and resource limitation of LGs, GDOT’s technology cannot be directly adopted. To address this issue, a research team at Georgia Institute of Technology (Georgia Tech) developed the Computerized Pavement Condition Evaluation System for Counties and Cities (COPACES-CC) that is based on GDOT’s current practice on pavement management and maintenance. To facilitate the technology transfer (T2) from GDOT to Georgia’s LGs, the Georgia Tech research team has collaborated with GDOT’s Office of Research, Local Technical Assistance Program, Office of Local Grants, and Office of Maintenance. LGs’ decision makers and stakeholders were also engaged in this T2 activity. This paper presents the research need and development of COPACES-CC, a guided T2 process that was adopted in three-stage development and implementation strategies, and the key factors that are related to the components of a guided T2. Conclusions and future recommendations are offered. It is hoped that the successful experience in the T2 of COPACES-CC could be useful and beneficial to other state DOTs and LGs.

Identification of Site Characteristics for Proactive High-Friction Surface Treatment Site Selection using Sensor-Based, Detailed, Location-Referenced Curve Characteristics Data

The Georgia Department of Transportation (GDOT) has developed a proactive high-friction surface treatment (HFST) program for curve sites prone to run-off-road (ROR) crashes. Using crash data and a single-criterion, ball bank indicator (BBI) value, GDOT seeks to maximize the return on its HFST investment. GDOT has partnered with Georgia Tech to identify additional factors for its HFST site-selection (HFST-SS) decision-making process by leveraging high-resolution, full-coverage sensor data (e.g., GPS and LiDAR). This paper proposes a methodology to identify site characteristics that can be used in GDOT’s HFST-SS process by leveraging the sensor data and automatically extracting roadway curve features as follows: (a) roadway data collection using state-of-the-art sensing technologies, (b) automatic extraction of detailed site characteristics data and curve information, (c) curved-based roadway segmentation using the extracted curve information; (d) spatial integration of curve-site characteristics data (CSCD); (e) analysis of CSCD and ROR crashes to identify additional factors for HFST site selection. A case study using CSCD extracted from Georgia State Route 2 demonstrates the proposed methodology. Results show that on sharp curves having comparable site characteristics, vertical grades greater than 3% play an important role in ROR crashes. Therefore, a vertical grade greater than 3% could be considered as an additional HFST-SS factor along with the current BBI criterion.

Let sleeping bats lie: Analyzing institutional adaptation to environmental regulatory change through Adaptive Management theory

Employing a case of a state transportation agency, we examine how complex institutions which integrate outsourcing within a bureaucratic process adapt to environmental regulatory changes. In 2012, two endangered species of bats were located outside of their established ranges in northern Georgia. These discoveries required the Georgia Department of Transportation (GDOT) to comply with a new set of federal regulations relating to those species when developing its projects. This article examines how GDOT adapted to new and unforeseen regulations in the face of environmental uncertainty. Using archival and interview data, we describe how GDOT engaged in Adaptive Management (AM) to internalize environmental changes (i.e. sufficiently stabilize the situation so that the project can get back on track). We also examine the role of outsourcing in bureaucratic agencies as an avenue for AM and suggest extending the AM model to describe mediating actors in the adaptive process. Furthermore, we investigate the impact adaptation had on project outcomes by analyzing 81 bridge projects, which are most susceptible to these environmental shocks, from a sample of 429 transportation projects using multivariate regression. We show that GDOT engaged in initial decision-making, iterative learning, and collaboration through a multi-tiered communication structure. We then present evidence supporting the narrative that these strategies helped it mitigate the impact of subsequent environmental shocks and improve project outcomes over time through adaptation.

Understanding the emission impacts of high-occupancy vehicle (HOV) to high-occupancy toll (HOT) lane conversions: Experience from Atlanta, Georgia

ABSTRACTConverting a congested high-occupancy vehicle (HOV) lane into a high-occupancy toll (HOT) lane is a viable option for improving travel time reliability for carpools and buses that use the managed lane. However, the emission impacts of HOV-to-HOT conversions are not well understood. The lack of emission impact quantification for HOT conversions creates a policy challenge for agencies making transportation funding choices. The goal of this paper is to evaluate the case study of before-and-after changes in vehicle emissions for the Atlanta, Georgia, I-85 HOV/HOT lane conversion project, implemented in October 2011. The analyses employed the Motor Vehicle Emission Simulator (MOVES) for project-level analysis with monitored changes in vehicle activity data collected by Georgia Tech researchers for the Georgia Department of Transportation (GDOT). During the quarterly field data collection from 2010 to 2012, more than 1.5 million license plates were observed and matched to vehicle class and age information using the vehicle registration database. The study also utilized the 20-sec, lane-specific traffic operations data from the Georgia NaviGAtor intelligent transportation system, as well as a direct feed of HOT lane usage data from the State Road and Tollway Authority (SRTA) managed lane system. As such, the analyses in this paper simultaneously assessed the impacts associated with changes in traffic volumes, on-road operating conditions, and fleet composition before and after the conversion. Both greenhouse gases and criteria pollutants were examined.Implications: A straight before-after analysis showed about 5% decrease in air pollutants and carbon dioxide (CO2). However, when the before-after calendar year of analysis was held constant (to account for the effect of 1 yr of fleet turnover), mass emissions at the analysis site during peak hours increased by as much as 17%, with little change in CO2. Further investigation revealed that a large percentage decrease in criteria pollutants in the straight before-after analysis was associated with a single calendar year change in MOVES. Hence, the Atlanta, Georgia, results suggest that an HOV-to-HOT conversion project may have increased mass emissions on the corridor. The results also showcase the importance of obtaining on-road data for emission impact assessment of HOV-to-HOT conversion projects.

Influence of built environment on pedestrian crashes: A network-based GIS analysis

Built environment is an essential component of traffic safety. Using data from the Georgia Department of Transportation from 2000 to 2007, the authors investigated the influence of built environment on pedestrian crashes in DeKalb County, Georgia, United States. A network-based kernel density technique was used to locate the hotspots of pedestrian crashes. An environmental audit was then conducted to provide a qualitative assessment of the built environment within each hotspot. Bivariate analysis and negative binomial models were utilized to examine the influence quantitatively after the road was partitioned into short road segments. Results show more pedestrian crashes were likely to occur on segments with less gradient change and with more public transit stops, and be close to census tracts with more public transit users, fewer seniors, and more linguistically isolated households. The findings emphasize the need for more extensive prevention campaigns around the hotspots and consideration of pedestrian safety in future urban and environmental planning.

A cradle to gate LCA framework for emissions and energy reduction in concrete pavement mixture design

Abstract The plurality of life-cycle assessments for concrete pavements have used national or regional data to develop models to aid in decision related processes for designing concrete pavements. The findings of these studies effectively yield information in which global decisions can be made with respect to material selection and implementation. However, construction projects are dependent upon local supply and demand and require different levels of design complexity: freeze–thaw exposure, leaching potential, and alkali-silica reactivity. This paper presents a cradle-to-gate framework for design engineers and concrete ready-mix producers to implement in an effort to optimize mixture designs across economic, environmental, and mechanical performance criteria. The framework was assessed through the examination of a newly constructed highway in South Georgia. The case study proved that through the incorporation of 20% more Class F fly ash compared to the in-situ mixture design used on the pavement project site, all fresh and hardened concrete properties set by the Georgia Department of Transportation could still be exceeded, while the environmental efficiency would also increase by 23% with a total cost savings of 0.67 million dollars for the 11.52 linear miles (18.54 linear km) of pavement. Additionally, updated emission factors for input materials into concrete pavements, ranging from cement to aggregates, are presented using the most up-to-date data and techniques.

Critical Assessment of the Long-Term Performance and Cost-Effectiveness of a New Pavement Preservation Method: Micromilling and Thin Overlay

The Georgia Department of Transportation (DOT) has developed and implemented a new pavement preservation method that uses micromilling in conjunction with a thin overlay for cost-effective replacement of a deteriorated, thin, open-graded surface layer (¾ to 1¼ in.) without removing the sound underlying layer. The new method was first implemented in 2007 on a 15.3 mi section on I-75 near Perry, Georgia, and resulted in significant savings. In considering whether to adopt the new method as standard practice in pavement preservation, it is essential to evaluate the method’s actual, long-term performance. This is the first paper to report the actual, long-term performance of the new method. The study used 8 years of pavement condition data collected by the Georgia DOT and sensing data collected by Georgia Institute of Technology. Based on the data, the pavement on I-75 is still in good condition after 8 years (with 6% raveling in 2015). Based on the current condition and the Georgia DOT’s past experience with open-graded surface mixes, the Georgia DOT anticipates an additional 2 to 3 years of service for the I-75 project. This provides a service interval of 10 to 11 years, which is comparable to that of Georgia’s conventional pavement rehabilitative strategy for open-graded surfaced Interstate pavements, which typically ranges from 10 to 12 years. The equivalent uniform annual cost of the new method is $11,200 per lane-mile per year, which is 11% lower than the $12,600 per lane-mile for the conventional rehabilitation method computed on a 12-year service interval. This information can assist transportation agencies in making decisions about whether to apply the new pavement preservation treatment.

Managerial Practices, Trust in Leadership, and Performance

This research explores relationships among managerial practices, trust in leadership, teamwork, and organizational performance. It seeks to contribute to the existing knowledge of trust research by exploring several antecedents and outcomes. Based on the social exchange theory, the research examines whether managerial practices are associated with trust in leadership. Trust in leadership, then, is expected to improve teamwork and organizational performance by several leadership theories. Using data from an employee survey conducted for the Georgia Department of Transportation (GDOT) in 2010, we test the ideas by structural equation model analysis. Three kinds of trust in leadership—trust in department leadership, trust in leadership team, and trust in supervisor—are investigated. The analysis confirms that several managerial practices are substantially associated with the three kinds of trust, which in turn affects teamwork and organizational performance. Trust in department leadership is directly related to performance, whereas trust in leadership team and trust in supervisor are more likely to affect performance through teamwork.

I-85 High-Occupancy Toll Lane's Impact on Commuter Bus and Vanpool Occupancy in Atlanta, Georgia

In October 2011, Atlanta, Georgia, opened its first high-occupancy toll (HOT) lanes, which were converted from high-occupancy vehicle (HOV) lanes. In partnership with the Georgia Department of Transportation, the Georgia Institute of Technology established a research team to assess changes in vehicle throughput, vehicle occupancy, and passenger throughput associated with the I-85 HOV-to-HOT lane conversion. For the assessment of these measures, commuter bus ridership, which carries a significant portion of ridership, could not be collected through the applied efforts to collect field data. Moreover, the effects of ridership and vehicle throughput on vanpools, which also use the managed lanes, are unknown. The purpose of this research was to estimate the change in vehicle and person throughput of alternative modes before and after the HOV-to-HOT lane conversion. The results indicate that person throughput remained relatively stable for commuter buses, even with an increase in vehicle throughput. The vehicle throughput of vanpools was not substantial and increased slightly after the conversion. The commuter bus results were unexpected, as ridership was expected to increase because of the related travel time saving and reliability. Behavioral research is needed to understand the underlying effects of ridership to separate the underlying effects from external factors such as gas prices, travel times, employment, and others.

Microsimulation-Based Framework for Freeway Travel Time Forecasting

This paper presents a microsimulation-based framework for generating short-term forecasts of travel time on freeway corridors. The microsimulation model that was developed replicated freeway capacity drop and relaxation phenomena critical for modeling non–steady state conditions. This framework was evaluated offline on a real-world freeway corridor by using data from manual counts and the Georgia Department of Transportation's video detection system. The travel time forecasts were compared with ground truth travel time data; the comparison demonstrated the efficacy of this framework to produce realistic forecasts.

Collection, Screening, and Evaluation of Vehicle Occupancy Data

In 2008, the Georgia State Road and Tollway Authority and the Georgia Department of Transportation successfully applied to the U.S. Department of Transportation for seed funding under the Congestion Reduction Demonstration Program Grant to convert the congested I-85 high-occupancy-vehicle (HOV) lane to a high-occupancy-toll (HOT) lane. The facility was converted into a HOT lane and opened on October 1, 2011. This paper reports on the collection, processing, and filtering of the vehicle occupancy data and the factors that influenced vehicle occupancy along the I-85 corridor both before and after the conversion. The research effort used regression tree analysis techniques to identify data collector bias and then the effects of different factors such as season, data collection site, morning peak–evening peak, and lane type on vehicle occupancy. The study found that vehicle occupancy data on Monday were different from those data collected on Tuesdays, Wednesdays, and Thursdays. The study identified data collectors and sessions that had statistically different data from other sessions and filtered those data. The research effort then explored the factors affecting vehicle occupancy for the morning peak and the afternoon peak periods separately, given that related travel behavior characteristics were quite different. The research identified lane type (managed or general-purpose) as the most significant factor affecting occupancy, followed by pre- and postconversion of the HOV lane to HOT lane. The study also found that average vehicle occupancy decreased after conversion of the HOV lane. On the basis of the results of this study, the vehicle occupancy data were aggregated into the center stations, north of the GA-316 region and south of the I-285 region. The vehicle occupancy data are currently being applied to evaluate person throughput along the corridor.