Speakers & Presentations

More Information Coming Soon!

More Information Coming Soon!

Xiaopeng Li, University of Wisconsin-Madison, Director of the U.S. Tribal and Rural Autonomous Vehicles for Efficiency, Livability and Safety Program

More Information Coming Soon!

ABSTRACT 

Several studies have shown that female occupants are more susceptible to severe injuries and fatalities than males. For example, one study reported that females have a 22% higher risk of head injury, 45% higher risk of neck injury, 26% higher risk of chest injury, and 80% higher risk of leg injury with compared to males. The American Association of State Highway and Transportation Officials (AASHTO) Manual for Assessing Safety Hardware (MASH) provides guidelines for crash testing and evaluating roadside safety hardware. MASH safety criteria include occupant compartment intrusion limits and the flail space model when assessing roadside safety hardware. Occupant compartment intrusion limits, or the amount of interior vehicle crush toward an occupant, vary based upon location within the vehicle, such as the windshield, roof, floorboard, toe pan, door, etc. Intrusion limits established by MASH have been shown to strongly predict injury risk, with crashes exceeding these limits significantly more likely to result in injuries. However, even intrusions below the thresholds can cause significant injuries, especially for females.

The flail space model assumes unrestrained occupants with longitudinal and lateral traversal distances prior to occupant impact with stiff internal vehicle structures. Longitudinal and lateral traversal distances of 24 in. and 12 in. respectively, were determined with a 50th-percentile male dummy. The time at which the occupant traverses 24 in. longitudinally or 12 in. laterally, which is the theoretical time of occupant impact with interior vehicle structures, establishes the longitudinal and lateral occupant impact velocities (OIVs). The flail space model assumes that the occupant remains in contact with the vehicle after initial impact. Therefore, the occupant is subjected to the same accelerations and velocity changes as the vehicle. As such, the occupant ridedown acceleration (ORA) is determined from the peak longitudinal and lateral accelerations after the occupant impacts the vehicle interior. The limits for OIVs and ORAs are 40 ft/s and 20.49 g, respectively. 

Due to known differences in male/female injury and fatality risk in the event of a crash, it was desired to evaluate the flail space model with varying longitudinal and lateral traversal distances reflective of the 5th- and 50th-percentile female and 95th-percentile male seated positions. To do so, 244 previous crash tests across 11 system types were reevaluated with the alternative traversal distances. Of those tests, 26%, 44%, and 57% indicated 5th-percentile female, 50th-percentile female, and 95th-percentile male, respectively had same or less severe OIV when compared to 50th-percentile male. Further, 95%, 98%, and 89% of tests indicated the 5th-percentile female, 50th-percentile female, and 95th-percentile male, respectively had same or less severe ORA when compared to 50th-percentile male. Differences in which 5th- and 50th-percentile female exhibit higher OIVs and ORAs than 50th-percentile male were minor.

To further evaluate MASH occupant risk criteria, the Hybrid III 5th percentile female, 50th percentile male, and 95th percentile male dummy models were seated in vehicle models and impacts into roadside safety hardware were simulated, and dummy injury criteria were compared to MASH criteria. 

PRESENTER

Brandon Perry

Brandon Perry received his Bachelor's degree in Biological Systems Engineering at the University of Nebraska-Lincoln (UNL) and Master’s degree in Mechanical Engineering from the University of Virginia (UVA). He is currently a Research Engineer at UNL’s Midwest Roadside Safety Facility (MwRSF) where his focus is developing crashworthy roadside safety hardware and evaluating designs with computer simulations and full-scale crash tests.

ABSTRACT 

The American Association of State Highway and Transportation Officials (AASHTO) Manual for Assessing Safety Hardware (MASH) provides guidelines for crash testing and evaluating roadside safety hardware. Roadside safety hardware includes bridge rails, longitudinal barriers such as W-beam guardrail and cable barriers, end terminals for longitudinal barriers, crash cushions, sign supports, luminaire poles, etc. Evaluation criteria for full-scale vehicle crash testing are based on three factors: (1) structural adequacy, (2) occupant risk, and (3) post-impact vehicle trajectory. Criteria for structural adequacy are intended to evaluate the ability of roadside barrier to contain and redirect impacting vehicles. In addition, controlled lateral deflection of the test article is acceptable. Occupant risk evaluates the degree of hazard to occupants in the impacting vehicle. Post-impact vehicle trajectory is a measure of the potential of the vehicle to result in a secondary collision with other vehicles and/or fixed objects, thereby increasing the risk of injury to the occupants of the impacting vehicle and/or other vehicles. 

Occupant compartment intrusion limits, or the amount of interior vehicle crush toward an occupant, vary based upon location within the vehicle: 

• Roof: ≤ 4.0 in.,
• Windshield: must not tear or deform more than 3 in.,
• Side window: must remain intact (no shattering),
• A- and B-pillars: ≤ 5 in. resultant deformation and ≤ 3 in. lateral deformation,
• Toe pan, and front side door area (above the seat): ≤ 9 in.,
• Side front panel, front side door (below the seat), floor pan, transmission tunnel: ≤ 12 in.

The flail space model assumes unrestrained occupants with longitudinal and lateral traversal distances prior to occupant impact with stiff internal vehicle structures. Longitudinal and lateral traversal distances of 24 in. and 12 in., respectively. The time at which the occupant traverses 24 in. longitudinally or 12 in. laterally, establishes the longitudinal and lateral occupant impact velocities (OIVs). The flail space model assumes that the occupant remains in contact with the vehicle after initial impact. Therefore, the occupant is subjected to the same accelerations and velocity changes as the vehicle. As such, the occupant ridedown acceleration (ORA) is determined from the peak longitudinal and lateral accelerations after the occupant impacts the vehicle interior. The limits for OIVs and ORAs are 40 ft/s and 20.49 g, respectively. 

Challenges and examples of roadside safety hardware design, testing, and evaluation, the evolution of the vehicle fleet, and compatibility of battery electric vehicles (BEVs) with current roadside safety systems will be discussed. 

PRESENTER

Brandon Perry

Brandon Perry received his Bachelor's degree in Biological Systems Engineering at the University of Nebraska-Lincoln (UNL) and Master’s degree in Mechanical Engineering from the University of Virginia (UVA). He is currently a Research Engineer at UNL’s Midwest Roadside Safety Facility (MwRSF) where his focus is developing crashworthy roadside safety hardware and evaluating designs with computer simulations and full-scale crash tests.

ABSTRACT

This paper focuses on traffic safety and operational performance of rural, minor approach stop-controlled intersections with free right-turn (FRT) ramps. Studies on the guidelines, safety, and operational analysis of FRT ramps are limited. Therefore, this paper formulates a comprehensive framework of FRT ramp studies and applies it to Nebraska. 

As of 2023, 79 FRT ramps exist at 68 rural highway intersections in Nebraska. FRT ramps may be located on three-legged or four-legged intersections and may be on the minor, the major, or both minor and major approaches of the same intersection. This study compared 68 rural FRT ramp intersections with 24 similar non-FRT rural intersections to identify differences in crash frequency, crash rate, and crash severity using 2010 to 2019 crash data from the Nebraska Department of Transportation (NDOT). The crash analysis did not show any statistically significant differences between the two intersection groups. This result is identical to a 1995 Nebraska-based study of rural FRT ramp intersection safety.

The research further assessed the operational impacts of FRT ramps. Using a calibrated and validated VISSIM microsimulation model, traffic operations at FRT ramps and non-FRT intersections were modeled and analyzed to study 324 scenarios across varying traffic and roadway geometry. 

As no statistically significant safety differences between the intersection groups were found, this research performed cost-benefit analysis based on the operational outcomes. Assuming a 20-year lifespan, using the operation data, this cost-benefit analysis was conducted for combinations of discount rates (4%, 6%, 8%), major road annual average daily traffic (AADT) (5,000, 10,000, 15,000), minor road AADT (2,500, 5,000, 7,500), percentage of right-turning traffic (10%, 25%, 50%), FRT ramp radius (650, 1,200, 1,800 ft), and speed limit (45, 55, 65 mph). Using the results, this study provides guidelines for NDOT for FRT ramp construction, reconstruction, or removal. Traffic agencies in Nebraska and the Midwest may make more informed decisions about FRT ramps using the guidance in this paper. However, the widely applicable methodology presented here for determining the feasibility of the FRT ramp can be applied in other locations in the United States without appreciable loss of generality.

PRESENTERS

MM Shakiul Haque

Dr. MM Shakiul Haque is a post-doctoral research associate of Mid-America Transportation Center, University of Nebraska-Lincoln.  He earned his Ph.D. in Civil Engineering from the University of Nebraska-Lincoln and his M.S. in Civil Engineering from Lamar University. 

His research focuses on transportation operation, with emphasis on microsimulation modeling of transportation systems and evaluation and improvement of operational efficiency, mobility analysis involving Intelligent transportation System, and intersection/interchange safety.

Jon Camenzind

Jon Camenzind is a Transportation Engineer-in-Training (EIT) at HDR. He earned both his M.S. and B.S. in Civil Engineering from the University of Nebraska–Lincoln. His research experience includes railroad-crossing data analysis, intersection inventory development, and safety evaluations of free-right-turn ramps. His career combines transportation research with the practical application of transportation engineering principles in the consulting industry. In addition to his academic and professional experience, Jon served as a tutor for the University of Nebraska–Lincoln Athletic Department.

Aemal J. Khattak

Dr. Aemal Khattak is a Professor of Civil and Environmental Engineering at the University of Nebraska-Lincoln and Director of the Mid-America Transportation Center, which is the U.S. Department of Transportation’s Federal Region 7 University Transportation Center.

He earned his Ph.D. in Civil Engineering from North Carolina State University and his M.S. in Civil Engineering from Pennsylvania State University. His research focuses on transportation safety, with particular emphasis on highway-rail grade crossing safety, data-driven risk modeling, and the use of advanced technologies and community-based strategies to reduce transportation hazards.

ABSTRACT

Bridge rehabilitation and replacement decisions are commonly guided by structural condition, functional classification, and traffic demand. However, it remains unclear whether differences between bridge width and the width of the approaching roadway influence traffic safety. This study examined the relationship between relative bridge width (RBW), defined as bridge width minus approach roadway width, and crash frequency on Nebraska bridges.

Five years (2020–2024) of Nebraska Department of Transportation (NDOT) crash data were integrated with statewide bridge inventory, roadway, and traffic data for approximately 9,300 bridges. A GIS-based bridge influence-zone method was developed to improve bridge-crash attribution by incorporating bridge decks together with upstream and downstream stopping-sight-distance transition areas. A negative binomial model was estimated using natural log-transformed vehicle miles traveled (VMT) as the exposure measure while accounting for urban location, functional classification, bridge length, out to out deck width, and RBW.

The preferred model substantially improved model fit compared with alternative exposure formulations. Traffic exposure exhibited an elasticity of 0.559, indicating that expected crashes increased less than proportionally with VMT. RBW exhibited a negative association with crash frequency (β = −0.00562, p = 0.078), suggesting that each additional foot of bridge width relative to the approaching roadway was associated with an estimated 0.56% reduction in expected five-year crashes and an average marginal reduction of 0.0026 crashes per bridge over five years. Urban bridges, collector and minor arterial facilities, longer bridges, and wider bridge decks were associated with higher crash frequencies.

Overall, the findings suggest that Transportation agencies may evaluate RBW as a practical safety-screening measure for bridge rehabilitation and widening decisions while demonstrating the importance of appropriate traffic-exposure specification when developing bridge safety performance models.

PRESENTERS

Aemal Khattak

Dr. Aemal Khattak is a Professor of Civil and Environmental Engineering at the University of Nebraska-Lincoln and Director of the Mid-America Transportation Center, which is the U.S. Department of Transportation’s Federal Region 7 University Transportation Center.

He earned his Ph.D. in Civil Engineering from North Carolina State University and his M.S. in Civil Engineering from Pennsylvania State University. His research focuses on transportation safety, with particular emphasis on highway-rail grade crossing safety, data-driven risk modeling, and the use of advanced technologies and community-based strategies to reduce transportation hazards.

MM Shakiul Haque

Dr. MM Shakiul Haque is a post-doctoral research associate of Mid-America Transportation Center, University of Nebraska-Lincoln.  He earned his Ph.D. in Civil Engineering from the University of Nebraska-Lincoln and his M.S. in Civil Engineering from Lamar University. 

His research focuses on transportation operation, with emphasis on microsimulation modeling of transportation systems and evaluation and improvement of operational efficiency, mobility analysis involving Intelligent transportation System, and intersection/interchange safety.

Aobo Wang, Oregon State University 

More Information Coming Soon!

ABSTRACT

One of the most challenging tasks on highway design-build projects is providing an effective Maintenance of Traffic (MOT) plan that allows construction to occur and preserves motorist level of service. Traditionally, transportation agencies struggled with making the most out of their MOT plans. First, MOT plans are often based on historical traffic, but fail to account for travelers’ changing their trips during construction, which leaves missed opportunities to keep working during good operations and help shorten construction duration. Second, performance of alternative detour routes can sometimes be subjective, where citizen complaints drive whether countermeasures are deployed. In the past, the cost of installing roadside devices to capture comprehensive data was very high, but innovative tools using Big Data are changing that.

This presentation discusses a real-world use case with the I-494/I-35W design-build project in Minnesota, which is a massive freeway reconstruction project in the Minneapolis-St. Paul region. On freeways and arterials, the design-build team is utilizing Big Data to monitor traffic performance all hours on all routes, looking for periods where certain thresholds are exceeded. Results can inform many conversations, including whether traffic operations could sustain expanded MOT hours or whether additional strategies are required to help support that.

The Minnesota Design-Build team is using the Iteris ClearGuide platform to gather and assess traffic operations, but other Big Data platforms can apply to these use cases. The ClearGuide platform assists traffic engineers to ensure no delay exceeds the mandated thresholds; identify congestion hotspots; observe effects of roadway closures and construction staging; and measure corridor performance during and post construction activities.

PRESENTERS

Christos Achillides

Christos Achillides is a Vice President with over 21 years of Traffic Engineering experience. Christos' experience includes signal system optimization studies using Signal Performance Measures (SPM), ITS/transportation systems design and operations, MOT plans for construction projects, system integration, and signal warrant studies.

Adam Danczyk

Adam is a principal engineer with over 20 years of ITS planning, design, and deployment experience. He has been involved in a wide variety of technical matters involving high-profile transportation technology topics across the United States, including truck parking systems, connected and automated vehicles, and leveraging big data platforms to deliver insights to influence business decisions in traffic operations.

ABSTRACT

Transportation agencies often face decisions about selecting appropriate interchange types or converting existing facilities during highway reconstruction projects. While diamond interchanges, either stop-controlled (DIstop) or signalized (DIsig), are common, an alternative, the diverging diamond interchange (DDI), is gaining recognition for its potential to improve traffic flow and safety. However, there is a lack of comprehensive comparative analyses to select an appropriate interchange type, accounting for factors such as annual average daily traffic (AADT), traffic turning proportions, time-of-day, and associated costs. 

This research is conducted in the Midwestern state of Nebraska. As Nebraska progresses toward completing its expressway system, updated, defensible decision-making is essential for converting interchanges. Nebraska Department of Transportation’s (NDOT) existing guidelines, developed in the 1990s, no longer reflect current operational performances or modern interchange designs.

This study collected geometric and operational traffic data for the three interchange types. VISSIM microsimulation models were calibrated to imitate the real world traffic movements observed from the field data. Then the calibrated model created and analyzed  6,144 traffic scenarios for DIstop, DIsig, and DDI, considering factors such as AADT for arterial and ramp approaches, various traffic turning proportions for both approaches, and different time-of-day. Based on simulated traffic performance, predictive models were developed to evaluate interchange performance, providing insights into how thousands of traffic conditions influence the effectiveness of different interchange types. 

This research further conducted benefit-cost analysis using the construction/retrofit costs of interchange options and comparative operational performances. Based on the outcomes of benefit-cost analysis, this study also developed predictive models to assess the economic feasibility of different interchange choices. The outcomes of this study provide transportation agencies, practitioners, and researchers with detailed, context-sensitive, data-driven models and guidelines for interchange decisions that extend beyond the limited, generalized recommendations found in the existing literature. 

PRESENTERS

MM Shakiul Haque

Dr. MM Shakiul Haque is a post-doctoral research associate of Mid-America Transportation Center, University of Nebraska-Lincoln.  He earned his Ph.D. in Civil Engineering from the University of Nebraska-Lincoln and his M.S. in Civil Engineering from Lamar University. 

His research focuses on transportation operation, with emphasis on microsimulation modeling of transportation systems and evaluation and improvement of operational efficiency, mobility analysis involving Intelligent transportation System, and intersection/interchange safety.

Aemal J. Khattak

Dr. Aemal Khattak is a Professor of Civil and Environmental Engineering at the University of Nebraska-Lincoln and Director of the Mid-America Transportation Center, which is the U.S. Department of Transportation’s Federal Region 7 University Transportation Center.

He earned his Ph.D. in Civil Engineering from North Carolina State University and his M.S. in Civil Engineering from Pennsylvania State University. His research focuses on transportation safety, with particular emphasis on highway-rail grade crossing safety, data-driven risk modeling, and the use of advanced technologies and community-based strategies to reduce transportation hazards

ABSTRACT

Actuated signal controller is widely used across the United States and incorporating Volume-Density Control (VDC) features in modern signal controllers serve to facilitate actuated phase termination through gap-outs and max-outs, can improve intersection traffic operations by minimizing pre-matured phase endings and inefficient phase dwelling. Despite their widespread use, the effects of key VDC parameters, including added initial, maximum initial, time to reduce, time before reduction, and minimum gap on signal operations have not been comprehensively explored in the context of complex intersection operations, and it is unclear how adjustments to these parameters influence traffic signal control operational and safety performance. This study revisits and investigates VDC parameter configurations through Hardware-In the Loop Simulation (HILS) experiments. The study first establishes the operational need for VDC by comparing signal performance with and without VDC parameters configured, examining the resulting gap-out and max-out termination patterns. It then evaluates various VDC parameter selections across a range of traffic volumes, vehicle compositions (including truck percentages), approach grades, and driver behavior factors in VISSIM to identify considerations for effective VDC configuration. The findings aim to provide practitioners with actionable guidance for configuring VDC parameters to achieve desired gap-out and max-out ratios across diverse intersection geometric and traffic conditions, supporting more informed signal timing decisions.

PRESENTER

Tsigereda Mossie

Tsigereda 'Rose' Mossie is a concurrent master's and PHD student in Civil Engineering, specializing in Transportation Engineering, at Oregon State University. Rose's current research investigates the influence of volume density control parameters on traffic signal operations at an isolated intersection through Hardware-in-the-Loop (HILS) Simulation system. Her research interests include traffic operations, intelligent transportation systems, traffic simulation, and transportation data analytics. She is interested in expanding her expertise in transportation simulation and applying those skills to develop practical, data-driven solutions that support active traffic management. Originally from Ethiopia, Rose hopes to contribute to the development of traffic management strategies that improve the efficiency and reliability of transportation systems.

Li Zhao, University of Nebraska-Lincoln 

More Information Coming Soon!

Trilce Encarnación, University of Missouri-St. Louis 

More Information Coming Soon!

Geoffery Eyram Agorku, University of Arkansas 

More Information Coming Soon!

Omar Ahmad, University of Iowa 

More Information Coming Soon!

ABSTRACT

Wildlife–vehicle collisions (WVCs) are a persistent safety problem on Missouri roadways, causing injuries, fatalities, and economic loss while fragmenting habitat and restricting wildlife movement. Missouri recorded 17,609 total WVCs from 2019–2023 and ranked 17th nationally for WVCs in 2024 and 2025, highlighting the issue’s scale and urgency. The challenge is amplified by Missouri’s 33,811 miles of state-maintained roads and 10,454 bridges and culverts, which can impede or enable safe wildlife passage. To address this risk at a statewide scale, the Missouri Department of Transportation (MoDOT) received a grant from the Federal Highway Administration’s Wildlife Crossing Pilot Program to complete the Missouri Statewide Wildlife Vehicle Collision Study. The effort assembled a Technical Advisory Committee to guide methods and implementation and produced a practical blueprint for planning and delivering wildlife crossing accommodations that improve safety and connectivity.

The Study was conducted in two phases. Phase One focused on the analyses of available datasets to model and map current and future conditions, identify and rank wildlife-vehicle conflict hot spots where wildlife and drivers are at risk of WVCs and where wildlife need to move across roads, and created a list of top statewide wildlife-vehicle conflict locations where mitigation was most warranted. In Phase Two, the wildlife-vehicle conflict locations were prioritized based on various evaluation criteria, and the TOP 10 locations were visited to help develop recommended mitigation measures to both reduce crashes with wildlife and provide wildlife connectivity. A website was developed to inform the public of the study (www.modot.org/wildlife-vehicle-collision-study) with the opportunity for public comment, along with posts to Twitter/X, YouTube, Instagram, and Facebook pages.

Recommended solutions emphasize cost-effective retrofits rather than new, large wildlife overpasses, reflecting Missouri’s non-migratory wildlife movement patterns. Typical concepts include exclusion fencing that funnels animals toward existing bridges or culverts with a suitable path beneath the roadway, improving safe passage and reducing collision potential. Across the top ten segments, conceptual mitigation costs (including ongoing maintenance) total $11.6 million, with an estimated $43.4 million in benefits over 50 years.

To sustain and improve safety outcomes, MoDOT implemented the Roadkill Observation and Data System (ROaDS) to standardize reporting and enable citizen-science participation to capture unreported events. Public engagement was further supported through a project website with opportunities for comment and related outreach. Together, the study’s analytics, prioritization framework, and implementable retrofit concepts provide a replicable, safety-centered pathway to reduce WVCs statewide while enhancing wildlife connectivity for 25 priority species.

PRESENTERS

Caleb Knerr

Caleb Knerr is a Senior Environmental Specialist with the Missouri Department of Transportation (MoDOT) and the Improve Interstate 70 program Environmental contact specializing in wetland delineations for Clean Water Act Section 404 permitting with U.S. Army Corps of Engineers; threatened and endangered species surveys, habitat assessments, and coordination with U.S. Fish and Wildlife Service; Migratory Bird Treaty Act compliance; stormwater best management practices reviews; and freshwater mussel surveys for MoDOT projects. Caleb is also MoDOT's wildlife crossing project contact for the Missouri statewide wildlife-vehicle collision (WVC) hotspot analysis and prioritization project and the I-70 WVC mitigation projects funded by the FHWA Wildlife Crossing Pilot Program (WCPP) and INFRA grant.

Ian Waters

Ian Waters is an Environmental Scientist with HDR and has 11 years of experience in ecological surveying, environmental compliance, and project delivery. His expertise includes NEPA documentation, permitting, natural resource assessments, and compliance with federal, state, and local environmental regulations. Ian is highly experienced in Kansas and Missouri flora and fauna survey methodologies and leverages 6 years of geospatial data collection and GIS analysis to support environmental planning, infrastructure development, and regulatory compliance.

Tayler MacDonald

Tayler MacDonald is a Senior Environmental Specialist for the Missouri Department of Transportation (MoDOT), covering environmental permitting for impacts from transportation projects to threatened and endangered species, wetlands and streams, and migratory birds. Tayler also facilitated the completion of the 2026 Missouri statewide wildlife-vehicle collision (WVC) hotspot analysis and prioritization project and works to incorporate WVC mitigation considerations into MoDOT projects for improved wildlife and human safety.

Omid Armantalab, William Brown, Li Zhao, Wissam Kontar, University of Nebraska-Lincoln 

More Information Coming Soon!

ABSTRACT

Overview

Understanding crash patterns and identifying effective countermeasures have always been challenging. It requires a deeper look into the roadway, behavioral, and environmental contexts that shape crash outcomes. This study proposes a novel data-driven methodology that combines machine learning (ML), clustering, and spatial risk scoring to evaluate and prioritize countermeasures at both intersection and corridor levels. The approach was designed to be practical and transferable so that cities and counties can easily apply it under the Safe System Approach framework.

Methodology

This methodology was implemented in USDOT-funded Safe Streets for All (SS4A) projects in Eureka, Kansas, and El Dorado, Kansas. Findings successfully identified locally relevant countermeasures and revealed behavioral and roadway risk patterns that traditional screening methods often overlook.

• Crash data from local transportation networks were analyzed to identify variables most associated with crash severity using ML model (random forest).
• Using the selected factors derived from ML model—roadway type, control type, lighting, surface conditions, and driver behaviors—similar types of crashes were grouped together through several unsupervised clustering algorithms.
• Each resulting crash cluster represented a unique crash context or “crash typology” where crashes shared similar characteristics while remaining distinct from other clusters.
• Detailed examination of cluster profiles allowed identifying the dominant contributing factors, crash scenarios, and overrepresented conditions for each cluster.
• Countermeasures were then developed for each cluster by combining insights from relevant literature, FHWA and NCHRP guidance, and practical engineering judgment and field experience.

A severity-based scoring system was also developed to quantify crash risk by assigning greater importance to more severe crashes while still considering less severe and property-damage-only crashes. The weighting structure was derived from crash severity cost estimates used in Kansas Department of Transportation safety analyses and engineering judgement further calibrated with local crash characteristics.

• Fatal and disabling injuries were combined because both often share similar contributing conditions.
• To capture the spatial dimension of risk, DBSCAN clustering was used to aggregate crashes at intersections and roadway segments separately. The cumulative risk score at each location was normalized by its crash proportion within the same facility type to create a ranked High-Injury Network (HIN) highlighting the top high-risk intersections and corridors.

Results

Integrating the cluster-level profiles with the HIN helped determine which crash types dominated each high-risk location. Moreover, this method evaluated the countermeasures for HIN while also prioritizing the countermeasures for each intersection and segment. For example, if a site showed 50%, 30%, and 20% of total crashes from clusters 1, 2, and 3 respectively, countermeasure priorities would follow the same order—giving highest priority to strategies effective for Cluster 1, followed the other clusters. This proportional approach ensured that safety improvements directly reflected the dominant crash patterns present at each site, leading to more targeted and efficient countermeasure implementation.

Conclusion

The approach provides a scalable, transferable, and reproducible framework for cities seeking to implement data-informed, human-centered safety improvements consistent with Vision Zero and the Safe System principles.

PRESENTER

Saumik Sakib Bin Masud

Dr. Saumik Sakib Bin Masud is a Traffic Engineer at JEO Consulting Group specializing in transportation safety, the Safe System Approach, and data-driven decision making. His work focuses on developing Safe Streets and Roads for All (SS4A) Safety Action Plans, conducting highway safety and operational analyses, evaluating roadway and interchange alternatives, and applying predictive safety methods, machine learning, and advanced analytics to support evidence-based transportation planning, infrastructure investment, and policy decisions. He earned his Ph.D. in Transportation Engineering from the University of Kansas, where his research advanced the understanding of human factors and automated driving systems through machine learning and driving simulation. He continues to conduct applied research in transportation safety, intelligent transportation systems, and artificial intelligence, bridging research and practice to improve roadway safety and mobility. Dr. Masud is an active member of the ITE Safety Council, the ITE Vision Zero Standing Committee, and ASCE, and regularly presents his work at national and international conferences while contributing to innovative, data-driven transportation safety solutions.

ABSTRACT

Adaptive cruise control (ACC) vehicles are the first generation of automated vehicles. While fully automated vehicles are expected to benefit traffic flow, field experiments have shown that commercially available ACC vehicles may instead degrade it by reducing string stability and roadway throughput. To mitigate these effects, existing studies adjust the ACC control algorithm or introduce additional control inputs; however, few have examined the transition between the cruise control (CC) and ACC modes without modifying the ACC control algorithm itself, leaving the impacts of ACC vehicles incompletely understood. Although microscopic car-following models effectively describe the driving behavior of ACC vehicles, they capture only part of the dynamics that shape traffic flow, as they represent the ACC mode alone and omit the CC mode. To address this gap, we propose a unified dynamical model of CC and ACC that interpolates continuously between the two modes through a sigmoid weighting function, and improve traffic flow by designing the mode switching.

Based on this new model, we conduct an equilibrium and string stability analysis of the platoon, revealing the trade-off among safety, throughput, and string stability. The optimal switching threshold is designed under throughput-priority and safety-priority criteria, and compared against the threshold adopted by commercial ACC vehicles. Numerical experiments show that, with a properly designed switching threshold, the throughput increases by up to 58.6% and the average speed variation, a measure of speed oscillations, decreases by up to 39.7% relative to the commercial baseline. We conclude that the excessively large switching threshold of commercially available ACC vehicles is a likely cause of their negative impact on traffic flow, and that this impact can be mitigated by properly reducing the threshold toward a safer and more string-stable regime.

PRESENTER

Mingfeng Shang

Dr. Mingfeng Shang is an Assistant Professor at Rochester Institute of Technology. He received his B.S. in Civil Engineering from Southwest Jiaotong University in 2017, his M.S. in Civil Engineering from the University of Illinois Urbana-Champaign in 2018, and his Ph.D. in Civil Engineering from the University of Minnesota in 2024. His research focuses on transportation cyber-physical systems, heterogeneous driving behavior, traffic flow modeling for connected and automated vehicles, and transportation data analytics. Prior to joining RIT, he was an Assistant Research Professor at the University of Arizona, where he was affiliated with the Center for Applied Transportation Sciences and the Arizona Transportation Institute. Dr. Shang is a recipient of the ITS Arizona 2025–2026 Young Professional Scholarship, as well as the Doctoral Dissertation Fellowship, the Hsiao Shaw-Lundquist Fellowship, and the Matthew J. Huber Award from the University of Minnesota.

Mohammad Elayan and Wissam Kontar, University of Nebraska-Lincoln 

More Information Coming Soon!

ABSTRACT

Emerging automated and electric vehicles are reshaping traffic operations, safety, and energy use, but their system-level benefits depend on how individual vehicles behave in real traffic. This presentation will discuss a series of research efforts on AI-enabled microscopic traffic modeling and control, with a focus on integrating machine learning, traffic-flow theory, and transportation domain knowledge. First, I will introduce RACER, a physics-informed AI car-following model that incorporates rational driving constraints to improve both prediction accuracy and behavioral consistency. I will then discuss how neural-network-based driving models can be customized for adaptive cruise control, allowing automated vehicles to pursue different objectives such as traffic smoothing, safety, energy efficiency, and human-guided driving behavior. The presentation will also highlight recent work on socially adaptive automated vehicle control using Social Value Orientation, which provides a data-driven way to model how automated vehicles may balance egoistic and cooperative driving objectives in mixed traffic. In addition, I will discuss ongoing work on electric vehicle car-following behavior and its implications for mixed traffic dynamics and sustainable mobility. Finally, I will briefly explore how large language models may support transportation research, including driver behavior interpretation, car-following model development, pedestrian safety analysis, and human-in-the-loop intelligent transportation systems. Together, these studies demonstrate how AI can move beyond black-box prediction toward trustworthy, interpretable, and application-oriented transportation models for future mobility systems.

PRESENTER

Tianyi Li

Tianyi Li is an Assistant Professor of transportation engineering at Saint Louis University. He earned a Ph.D. degree in Transportation Engineering from the University of Minnesota and received a Master of Science in Transportation Engineering from the University of Washington, and a B.S. degree in Civil Engineering from Iowa State University. Tianyi Li is a three-time recipient of the President Dwight David Eisenhower Graduate Fellowship from the Federal Highway Administration (2021, 2022, 2023). He also serves as an Associate Editor for the IEEE Open Journal of Intelligent Transportation Systems.

ABSTRACT

A review of the literature identifies eight distinct categories of models used to simulate traffic flow: safe-distance, stimulus-response, optimal velocity, desired measures, gas-kinetic, kinematic, multi-class, and momentum-based. This presentation connects the fundamentals of fluid mechanics to each model, highlighting the equations typically solved, whether the model is applied to single or multiple vehicles, and the solver type required for computation. The discussion covers both microscopic and macroscopic traffic flow, distinguishing between models that involve straightforward computation using ordinary differential equations and those that require more complex numerical methods to solve partial differential equations.

This review, combined with an evaluation of microscopic leader-follower data from the Waymo Open Dataset, leads to new insight into simulating the follower vehicle. Specifically, observed oscillatory behavior in follower velocity relative to spacing motivates the inclusion of a bounded diffusion term in a safe-distance model. Historical context is provided on the use of diffusion in car-following models, with emphasis on how this approach aligns with and diverges from prior work. The modified model is evaluated across human-driven vehicle (HDV)-HDV, HDV-autonomous vehicle (AV), and AV-HDV datasets, with model predictions of distance, velocity, and acceleration compared to the data.

Connecting fluid mechanics with traffic flow theory enables a direct mapping between microscopic and macroscopic models. Results highlighting the translation of microscopic driver behavior to macroscopic traffic flow patterns are presented. By the end of this presentation, the audience will gain a clearer understanding of the interconnectedness of these models and how a fluid mechanics perspective can enhance the interpretation of traffic flow data.

PRESENTER

Christopher Depcik

Dr. Christopher Depcik is a Professor of Mechanical Engineering at the University of Kansas (KU), with a courtesy appointment in Aerospace Engineering. His research focuses on reacting flow modeling and energy systems, with an emphasis on reduced-dimensional (1-D and 1+1-D) formulations of chemically reacting flows and fluid mechanics. His work spans topics including catalytic reactors, desiccant wheels, traffic flow, and spirit maturation. His research group has published more than 130 peer-reviewed papers, and he has been recognized among the top 2% of researchers worldwide in the Elsevier Scopus rankings. He has received numerous departmental, school, university, national, and international awards for service, research, and teaching. He is a Fellow of ASME and SAE.

ABSTRACT

Resilience frameworks in transportation largely inherit their assumptions from natural hazards, which are random, bounded, and reasonably well characterized by historical data. Digitization strains that inheritance. Signal control, traffic management, connected vehicles, and electronic payment now depend on networked systems whose failures are chosen rather than drawn from a distribution. This presentation argues that cybersecurity does not simply add another hazard to existing resilience models but unsettles the premises on which they rest. An adversary adapts, selects targets, and may remain undetected, complicating both probabilistic reasoning and the recovery timelines that most resilience metrics assume. Drawing on work in cyber-physical systems and infrastructure resilience, the talk considers what changes when intelligent threats are treated as a design case: how disruption cascades across digital and physical boundaries, why organizational capacity may matter as much as technical hardening, and where research and agency practice should turn next.

PRESENTER

Kevin Heaslip

Kevin Heaslip, Ph.D., P.E., is Professor of Civil and Environmental Engineering and Director of the Center for Transportation Research (CTR) at the University of Tennessee, Knoxville. He directs FERSC, a $10M USDOT Tier 1 University Transportation Center focused on freight, energy, resilience, safety, and connected infrastructure, and co-leads the UT-ORNL Transportation Convergent Research Initiative. Under his leadership, CTR has grown to $15M in annual expenditures with over 150 personnel across four research thrusts: electrified connected automated transport, transportation operations, cybersecurity and resilience, and safety. His research spans transportation cybersecurity, intelligent transportation systems, infrastructure resilience, and automated and electric vehicle systems. He has secured over $40M in funded research across three institutions and published 73 journal articles. He is a licensed Professional Engineer and serves on TRB committees for connected/automated vehicles.

ABSTRACT

Transportation agencies face increasing challenges in maintaining aging infrastructure under growing climate uncertainty, more frequent extreme weather events, and constrained investment resources. While Transportation Asset Management (TAM) provides a systematic framework for infrastructure investment and lifecycle planning, existing approaches often prioritize assets primarily based on condition and maintenance needs, with limited consideration of multi-hazard exposure and transportation network criticality.

This research presents a GeoAI-enabled framework for resilience-oriented transportation asset management that integrates multi-hazard exposure assessment, transportation network criticality, and unsupervised machine learning to prioritize roadway resilience investments across Oklahoma. Statewide datasets representing flood, wildfire, drought, extreme heat, and extreme precipitation hazards were integrated to develop a Multi-Hazard Exposure Index (MHEI) using the Analytical Hierarchy Process (AHP). Transportation network criticality was evaluated by combining transportation importance, accessibility to population and emergency services, and infrastructure interdependency into a Network Criticality Index (NCI). These complementary indices were subsequently integrated to produce a Transportation Resilience Priority Index (TRPI) for ranking roadway segments according to resilience investment priority.

To complement the continuous prioritization index, a GeoAI-based K-means clustering approach was implemented using four resilience indicators (MHEI, transportation importance, accessibility, and infrastructure interdependency) to identify roadway segments exhibiting similar resilience characteristics. The resulting five roadway resilience typologies provide additional decision support by identifying groups of roadway segments requiring different resilience planning and management strategies.

The proposed framework generated statewide maps of multi-hazard exposure, transportation network criticality, roadway resilience typologies, and resilience investment priorities. Results identified transportation corridors exhibiting both elevated hazard exposure and high network importance, while overlay analysis with Oklahoma Department of Transportation (ODOT) maintenance districts demonstrated how the framework can support agency-specific resilience planning and investment prioritization. By integrating geospatial analytics, multi-criteria decision analysis, and GeoAI within a unified transportation asset management framework, this study provides a scalable and transferable methodology for supporting risk-informed transportation infrastructure investment under increasingly complex multi-hazard conditions.

PRESENTER

Richa Bhattarai

Dr. Richa Bhattarai is a Research Scientist in the School of Civil Engineering and Environmental Science at the University of Oklahoma. She holds a Ph.D. in Environmental Remote Sensing from Chiba University, Japan. Her research integrates geospatial science, artificial intelligence, and data-driven analytics to address complex challenges in infrastructure resilience, hazard mitigation, and environmental sustainability. She develops advanced analytical frameworks that combine spatial data, machine learning, and network analysis to support risk-informed decision-making for transportation systems, climate adaptation, and emergency management. In addition to conducting interdisciplinary research, Dr. Bhattarai is committed to translating research into operational tools through technology commercialization and collaborations with public agencies and industry, helping bridge the gap between scientific innovation and real-world decision support to enhance infrastructure resilience and public safety.

Samiha Karim Subah, University of Oklahoma 

More Information Coming Soon!

ABSTRACT

Freight transportation networks are increasingly exposed to disruptions arising from extreme weather events, infrastructure failures, and other sources of uncertainty. Inland waterways are particularly vulnerable to flooding, drought conditions, and operational disruptions, yet they remain critical components of the U.S. freight transportation system. Disruptions affecting inland waterways can generate cascading impacts across rail and truck networks, reducing system performance and increasing logistics costs throughout connected supply chains. These challenges highlight the need for decision-support tools that help transportation agencies and infrastructure operators improve system resilience before disruptions occur.

This research develops a Resilience with Preparedness Options framework that integrates preparedness investments, recovery actions, and freight routing decisions within a unified two-stage stochastic optimization model. The framework captures both proactive and reactive resilience strategies by jointly determining pre-disruption investments and post-disruption operational responses. A key feature of the model is the explicit representation of preparedness–recovery interactions, allowing preparedness actions to improve the effectiveness and speed of recovery efforts following a disruption.

The framework is applied to a multimodal freight transportation network based on the Mississippi River corridor and its connected rail and highway systems. Multiple disruption scenarios are considered, including flooding, drought, winter ice conditions, rail infrastructure failures, and hurricane impacts. Results demonstrate that the network maintains nearly complete demand satisfaction across disruption scenarios, with resilience levels exceeding 98% even under severe disruption conditions. The analysis further reveals that resilience is achieved primarily through multimodal adaptability and flow reallocation rather than protection of individual transportation assets. Rail emerges as a critical substitute mode during major inland waterway disruptions, while targeted preparedness investments improve both system capacity and recovery effectiveness.

The findings provide practical insights for freight transportation planning and infrastructure investment by highlighting the importance of multimodal connectivity, adaptive recovery strategies, and coordinated preparedness planning. While demonstrated in an inland freight context, the proposed framework is sufficiently general to support resilience planning in other transportation systems, including maritime freight networks facing emerging operational and geopolitical uncertainties.

PRESENTER

Juana Jaramillo-Rios

Juana Jaramillo Rios is a Ph.D. student in Supply Chain and Analytics at the University of Missouri–St. Louis. Her research applies operations research, econometrics, and behavioral modeling to address challenges in humanitarian logistics, transportation systems, and supply chain resilience. Her work spans disaster preparedness and recovery, multimodal freight transportation, consumer behavior, and public policy, with applications including relief distribution, inland waterway transportation, and last-mile delivery. Juana is passionate about using analytical methods to generate insights that support more resilient, efficient, and socially impactful systems.

LinkedIn: www.linkedin.com/in/juana-jaramillo-rios

Keshu Wu, Texas A&M University 

More Information Coming Soon!

Lokesh Das, Wichita State University 

More Information Coming Soon!

ABSTRACT

Parking is a challenge for last-mile commercial delivery drivers in urban environments. This work develops an analytical basis for understanding optimal parking decisions for delivery drivers. We introduce the Stochastic Parking Problem (SPP), in which a driver delivering a package to a nearby customer must decide whether to park at an available parking spot or double park at an unavailable parking spot at the risk of receiving a fine. If the delivery driver does not park, the driver may continue to search for alternative parking. We formulate the SPP as a Markov Decision Process. The optimal parking policy for the driver minimizes the expected cost of completing the delivery. We characterize the structure of the optimal policy for the SPP based on the probability that each parking spot is available and the expected cost of double parking. In particular, we show that the optimal policy for the SPP takes on one of four forms. For each form, we provide analytical conditions that determine its optimality for a given problem instance of the SPP. Through simulation experiments, we benchmark the optimal policy against common parking search strategies to highlight where the optimal policy for the SPP yields different parking decisions. Insights from this work benefit both companies aiming to improve delivery efficiency as well as cities aiming to control congestion related to double parking and cruising of commercial vehicles. For logistics providers, we provide guidance to drivers on the parking search process to enhance productivity. From the policy perspective, we provide sufficient enforcement levels and fine amounts to eliminate double parking as well as improve utilization of limited parking infrastructure in urban environments. Overall, this work reveals the importance of understanding parking choices at the level of route implementation for the delivery driver to improve productivity and design effective enforcement policies.

PRESENTER

Sara Reed

Sara Reed is an Assistant Professor of Business Analytics in the School of Business at the University of Kansas. She received her Ph.D. in Applied Mathematical and Computational Sciences from the University of Iowa, where she also earned an M.S. in Mathematics. Her research interests are transportation logistics, particularly focusing on challenges and opportunities in last-mile delivery. Her research has been published in Management Science, Transportation Science, and Transportation Research Part E: Logistics and Transportation Review. She was awarded first place in the 2020 Bayer Women in Operations Research Scholarship Prize by the Institute for Operations Research and Management Sciences (INFORMS) and honorable mentions for the INFORMS Transportation Science and Logistics Society 2023 Best Paper Award and 2022 Dissertation Prize. She currently serves as an editorial advisory board member for Transportation Research Part E: Logistics and Transportation Review.

ABSTRACT

In-vehicle voice assistants may reduce visual-manual interaction but can also create workload, misunderstanding, and poorly calibrated trust. This study examined age-group differences in voice-assistant perceptions and the associations of usability, trust, workload, and perceived safety with intention to use. A cross-sectional online survey was conducted in 2026 with 374 U.S. licensed drivers recruited through Prolific: 188 younger drivers aged 18-25 years and 186 older drivers aged 65-85 years. Mann-Whitney U tests compared group perceptions, while hierarchical regressions with HC3 robust standard errors tested associations and age-group interactions. Gaussian-mixture profiling and random-forest SHAP analyses described within-group heterogeneity and predictive patterns. Younger drivers reported greater in-vehicle voice-assistant familiarity (p=.010), whereas older drivers reported slightly greater trust (p=.037); no significant mean differences emerged for usefulness, ease of use, workload, perceived safety, or intention to use. Usefulness and trust were positively associated with perceived safety, while usefulness, ease of use, and perceived safety were associated with intention to use. Ease of use was more strongly associated with intention among younger drivers, whereas perceived safety was more strongly associated with intention among older drivers; these focused interactions should be interpreted cautiously because the overall interaction block was narrowly non-significant. The study integrates technology-acceptance and human-factors constructs to examine in-vehicle voice-assistant perceptions across age groups while identifying distinct within-group perceptions.

PRESENTER

Efthymia Kostopoulou

Dr. Kostopoulou is a Postdoctoral Scholar at Michigan State University. She received the B.S. and M.Eng. degrees in civil engineering from the National Technical University of Athens, Athens, Greece, in 2020, and the M.S. and Ph.D. degrees in civil engineering from the University of Massachusetts Amherst, Amherst, MA, USA, in 2022 and 2026, respectively. From 2021 to 2026, she was a Graduate Research Assistant with the Department of Civil and Environmental Engineering, University of Massachusetts Amherst. She was a College of Engineering Teaching Fellow in 2024 and 2025. Her research interests include human factors, traffic safety, traffic signal optimization, sustainable and multimodal transportation systems, transportation accessibility, and data-driven transportation analysis.

ABSTRACT

Commercial motor vehicle (CMV) drivers face increased risks of obesity, metabolic disorders, and cardiovascular disease due to long driving hours, prolonged sedentary work conditions, poor dietary habits, and irregular sleep patterns compared to non-commercial vehicle (NCMV). Driving studies on simulators have demonstrated that metabolic disorders negatively affect vehicle control, hazard perception, and decision-making. However, research on direct comparisons of cognitive and driving performance between CMV and NCMV drivers is sparse. 

Objectives

1. To compare cognitive and simulated driving performance between CMV and NCMV drivers.

2. To identify factors that predict simulated driving performance in both CMV and NCMV drivers.

Materials and methods

This prospective cross-sectional study consisted of CMV drivers (n=31, Males=23, median age=54.0 IQR 46.0-63.0) and NCMV drivers (n=30, Males=13, median age=41.5 IQR 24.0-66.3) with an active driver’s license. Both groups underwent cognitive tests on attention, flexibility, and speed, along with a simulated driving test assessing vehicle control, collisions, and brake reaction times.

Results

Both groups performed similarly on cognitive tests except the Montreal Cognitive Assessment (MoCA), where CMV drivers performed significantly worse (26.03 ± 2.59) than the NCMV drivers (28.30 ± 1.78), p=0.0006. Similarly, in the simulated driving performance, CMV drivers showed significantly prolonged complex reaction time (3.90 ± 0.57) compared to the NCMV drivers (3.38 ± 0.80), p=0.01. CMV drivers had fewer off-road accidents (p<0.001) and fewer road edge excursions (p=0.002) compared to the NCMV drivers. A forward multivariable logistic regression showed that history of at-fault crashes in the past 5 years (OR 0.15, p=0.02) predicted a positive outcome in simulated driving performance with area under the curve (AUC) = 0.6840. Health factors such as BMI and comorbidities, though different between the groups, did not predict the simulated driving outcome.

Conclusion 

CMV drivers had lower MoCA scores and slower brake response times than NCMV drivers, indicating minor but significant deficits in overall cognitive function and reaction time. However, CMV drivers demonstrated safer driving and better vehicle control with significantly fewer off-road accidents and fewer road edge excursions compared to the NCMV drivers. Collision history in personal vehicles was found to be an influential predictor of driving safety over health parameters and cognitive performance. Implications of this study include consideration of adding history of motor vehicle crash during recruitment, annual performance reviews, and DOT tests.

PRESENTERS

Dr. Shelley Bhattacharya, DO, MPH, FAAFP, AGSF is a practicing physician and has been involved in transportation safety for over 20 years. She was the KUMC Principal Investigator for the Department of Transportation UTC grant working on improving the on-the-road safety of commercial CDL drivers. 

Dr. Abiodun Akinwuntan, PhD, MPH, MBA, FASAHP, FACRM, is a Professor and the Dean of the School of Health Professions at the University of Kansas Medical Center (KUMC), Kansas City Dr. Akinwuntan is a driving rehabilitation specialist and a leading authority on the use of virtual-reality technologies to improve daily living activities in neurologically impaired persons. He has collaborated extensively on funded grants totaling more than $15 million and authored more than 110 peer-reviewed publications and abstracts including 8 book chapters. 

Dr. Hannes Devos, PhD is a driving rehabilitation specialist. He has published over 60 driving-related manuscripts in peer-reviewed journals with high impact factors. He also has over 80 abstracts presented at national and international conferences, 6 book chapters on fitness to drive, and awarded more than $14 million, as PI or Co-I, on driving-related studies. 

ABSTRACT

Perceived safety is an important factor influencing transportation choice and ridership of public transit systems (Friman et al., 2020). Patterns of self-reported victimization differ from patterns of perceived safety on transportation systems (Ceccato et al., 2024). Underreporting of incidents on-board is also widespread (Marteache et al., 2015; Wretstrand, 2008). However, limited research has studied the key contributors to poor perceived safety (e.g. disorderly behaviour, crime, or drug use) through on-board observation. This study aims to establish a methodology for live in-person documentation, or coding, of code of conduct (CoC) violations, anti-social behaviour, and general conditions on board Metro Transit light rail and bus rapid transit vehicles and at stations in the Twin Cities. Formative interviews with Metro Transit light rail transit (LRT) and bus rapid transit (BRT) users were conducted to aid in development of metrics for field data collection in conjunction with ongoing development of a discreet smartphone-based coding tool. This enabled researchers to collect data discreetly, and allows for an empirical understanding of frequency of CoC violations and other indicators of transit system health. Observations were collected by pairs of researchers over a 3.5-month period from October 2025 to January 2026 and a 1 month period from July 2027 to August 2027 and included observations from 2 LRT lines, 5 BRT routes, and 27 different stations and platforms, totalling nearly 300 unique trips. 

A subset of the in-person data collection sessions was also reviewed and coded via security footage obtained from the Metro Transit Police Department for the same CoC violations, anti-social behaviour, and general conditions to assess the viability of remote assessment. An accompanying public survey was also distributed among Metro Transit LRT and BRT users during the data collection period. Respondents were asked to recount their most recent LRT or BRT trip, including whether they had observed any of the same behaviours or conditions documented by study researchers, enabling another avenue of cross-comparison and agreement assessment. 

Findings will be shared regarding the development of the mobile coding tool, data collection protocols, and on-board observations of CoC violations collected from Metro Transit lines in the Twin Cities area. Preliminary findings on in-person vs. video coding feasibility and agreement will be discussed. Findings from this study should serve as guidance for future assessments of transit system safety and to help direct interventions designed to improve perceived safety and on-board and at-station conditions in public transit environments. 

PRESENTER

Marshall L. Mabry

Marshall is an Assistant Scientist in the Human Factors Safety Laboratory in the Mechanical Engineering Department at the University of Minnesota. He received his Bachelor's degree in Urban Studies, with a minor in Environmental Science, Policy, and Management from the University of Minnesota - Twin Cities, and is a second year master's student in the Urban and Regional Planning program at the Humphrey School of Public Affairs. Marshall’s current research interests include transit system safety and rider experience, school zone pedestrian safety, and driving simulation. Further research projects have investigated visual perception of e-scooter riders as well as gender bias in medical training & pre-hospital trauma care. 

ABSTRACT

This research introduces a multimodal framework for automated pavement condition assessment that provides pavement condition index (PCI) predictions and qualitative descriptions using a single-shot PCI estimation network and a fine-grained dense captioning network. The PCI estimation network uses YOLOv8, the segment anything model, and a four-layer convolutional neural network for PCI prediction. The dense captioning network uses a YOLOv8 backbone, a transformer architecture, and a convolutional feed-forward module to generate textual descriptions. To train and evaluate these networks, we developed a pavement dataset containing bounding box annotations, textual descriptions, and PCI values. The PCI estimation network recorded a mean absolute error of 16.21 when tested on the validation dataset. The dense captioning network, on the other hand, generated accurate descriptions with bilingual evaluation understudy-1 (0.3799), Google’s BLEU (0.3234), and metric for evaluation of translation with explicit ordering (0.4253) scores, handling complex scenarios well. The proposed framework can improve infrastructure management and decision-making in pavement maintenance.

PRESENTER

Blessing Agyei Kyem

Blessing Agyei Kyem received the B.Sc. degree in civil engineering from Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana, in 2023, and the M.S. degree in civil engineering from North Dakota State University (NDSU), Fargo, ND, USA, in 2026. He is currently pursuing the Ph.D. degree in civil, construction and environmental engineering at NDSU, where he is a Graduate Research Assistant in the SMART Lab under Dr. Armstrong Aboah. Prior to graduate study, he worked as a Junior Data Scientist at Bismuth Technologies, Kumasi, Ghana, from 2019 to 2022, and as a machine learning instructor at BITLabs.

His research interests include deep learning, computer vision, and vision-language models, with applications to pavement asset management, intelligent transportation systems, cooperative perception, and connected and autonomous vehicles. He has authored 15 peer-reviewed publications in venues including the IEEE/CVF International Conference on Computer Vision (ICCV), Automation in Construction, Construction and Building Materials, IEEE Transactions on Intelligent Transportation Systems, and Expert Systems with Applications, and has reviewed over 70 manuscripts for IEEE, ASCE, and Elsevier journals. He received the 2026 Graduate Research Assistant of the Year Award from the NDSU College of Engineering. He is a member of IEEE, ASCE, ITE, and Tau Beta Pi.

ABSTRACT

Luke Attard1, William N. Collins1, Caroline R. Bennett1, and Jian Li

1. Department of Civil, Environmental, and Architectural Engineering, University of Kansas

2. Department of Electrical Engineering and Computer Science, University of Kansas

This presentation introduces a human-centered framework for infrastructure inspection that integrates Augmented Reality (AR) and Artificial Intelligence (AI) to support more efficient, consistent, and data-driven assessment of civil infrastructure systems. Using the Magic Leap 2 headset, we are developing an AR-based inspection platform that enhances inspectors’ ability to detect, visualize, document, and revisit structural defects in the field.

The system uses AI models to identify visible damage, such as cracks, corrosion, and surface deterioration, across different types of structural components and materials. Detected defects are overlaid directly onto the physical structure in real time through spatial anchoring, allowing inspectors to see areas of concern in their actual geometric context. Magic Leap’s spatial meshing capability generates a real-time 3D representation of the inspected structure, which serves as the basis for aligning AI-detected defects with the physical environment.

The application is developed in Unity, where detected damage is projected onto the spatial mesh to create a spatially coherent visualization that conforms to the target geometry. Each AI inference is associated with a spatial anchor and stored using Magic Leap’s spatial anchor storage system, enabling persistent documentation and retrieval of inspection results. This allows inspectors to compare current observations with prior records and supports long-term monitoring of defect progression.

By combining holographic visualization, AI-driven damage detection, and persistent spatial documentation, this work demonstrates how immersive technologies can assist infrastructure inspection and maintenance. The proposed framework has the potential to improve inspection efficiency, support more frequent condition assessment, and contribute to proactive infrastructure management and sustainability.

PRESENTER

Jian Li

Dr. Jian Li is the Deane E. Ackers Professor and Dean R. and Florence W. Frisbie Associate Chair of Graduate Studies in the Department of Civil, Environmental, and Architectural Engineering at the University of Kansas, with a courtesy appointment in Electrical Engineering and Computer Science. His research bridges structural engineering, sensing, artificial intelligence, and digital twins to advance data-driven monitoring, diagnostics, and predictive modeling for civil infrastructure systems. His work integrates machine learning, computer vision, Bayesian inference, data assimilation, and cyber-physical sensing technologies to improve the reliability, resilience, and safety of buildings, bridges, dams, and other large-scale structural systems.

Dr. Li is a Fellow of the American Society of Civil Engineers (ASCE) and the recipient of several honors, including the ASCE Walter L. Huber Civil Engineering Research Prize, the IASCM Takuji Kobori Prize, the Rising Stars in Structural Engineering Award, and the Miller Professional Service Award for Distinguished Research. He currently chairs the ASCE Structural Health Monitoring and Control Committee, serves on the Board of Directors and as Secretary of the U.S.-China Earthquake Engineering Foundation, and holds editorial board appointments with several international journals.

ABSTRACT

Sarper Demirdogen and Jie Han

1 Graduate Research Assistant, Department of Civil, Environmental, and Architectural Engineering, University of Kansas, Lawrence, KS 66045

2 Roy A. Roberts Distinguished Professor, Department of Civil, Environmental, and Architectural Engineering, University of Kansas, Lawrence, KS 66045

The modulus of subgrade reaction is a key parameter in the design of concrete pavements and is commonly estimated from soil properties or determined directly from plate load tests. This study evaluates numerical and analytical approaches for assessing non-stabilized and geosynthetic-stabilized pavement systems, with particular emphasis on the modulus of subgrade reaction and the representation of geogrid stabilization in numerical models. Two-dimensional numerical models developed for evaluating pavement systems were validated against Burmister’s two-layer elastic solutions using settlement coefficient influence curves as reference solutions. The validation confirmed that the numerical models reasonably captured the layered system response and showed that plate-size correction factors commonly used to estimate the modulus of subgrade reaction from smaller plate load tests may require modification for layered pavement systems. Burmister-based analyses also indicated that the modulus of subgrade reaction increases with increasing subbase thickness and increasing modulus contrast between the subbase and underlying subgrade. Also, various approaches have been used in three-dimensional numerical analyses to model geogrids, including structural geogrid elements, cable elements, and equivalent improvements in soil properties. Among these, structural geogrid elements are commonly adopted in finite difference modeling and can reasonably capture moderate settlement reductions in stabilized systems compared with non-stabilized conditions. However, the predicted improvement in the modulus of subgrade reaction may vary depending on how the geogrid is represented in the numerical model. This limitation becomes more pronounced for multiaxial geogrids, for which conventional structural geogrid elements may not fully reproduce the observed improvement because soil-geogrid interlocking and aggregate confinement are not explicitly represented. Overall, the findings highlight the importance of properly accounting for layered system effects, plate size dependency, and soil-geogrid interaction mechanisms when evaluating the modulus of subgrade reaction in non-stabilized and geosynthetic-stabilized pavement systems.

PRESENTER

Sarper Demirdogen

Dr. Sarper Demirdogen is a Graduate Research Assistant pursuing a second PhD in the Civil, Environmental, and Architectural Engineering Department at the University of Kansas under the supervision of Dr. Jie Han. His research focuses on geosynthetics, pavement systems, reinforced soil structures, and dam engineering. Before joining the University of Kansas, he worked as a geotechnical engineer at Turkiye’s General Directorate of State Hydraulic Works, contributing to the design, assessment, and post-earthquake inspection of numerous dams. He earned his PhD in Geotechnical Engineering from Gazi University and his MSc from the University of South Florida. Dr. Demirdogen has published more than 20 peer-reviewed journal and conference papers. His honors include the GSI Fellowship, Gazi University’s Best PhD Thesis Award, and fully funded scholarship for his graduate studies in the US. He is a member of ASCE Geo-Institute, the International Geosynthetic Society, and Deep Foundation Institute.

Sherbaz Khan, University of Louisiana at Lafayette 

More Information Coming Soon!