Alexandra Kondyli
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Alexandra Kondyli has a total of 56 co-authors. Here are the top co-authors by number of shared publications:
- Vishal Chandra Kummetha
Center for Urban Transportation Research, University of South Florida, Tampa, FL, USA
5 shared publications - Mohamadamin Asgharzadeh
Univ. of Kansas
3 shared publications - Steven D. Schrock
Department of Civil, Environmental, and Architectural Engineering, University of Kansas, Lawrence, KS
3 shared publications Kirti Mahajan
Department of Civil, Environmental, and Architectural Engineering, The University of Kansas, Lawrence, KS, USA
2 shared publicationsA. Barmpoutis
Digital Worlds Institute, University of Florida, Gainesville, Florida, USA
1 shared publicationAbiodun E. Akinwuntan
Department of Physical Therapy, Rehabilitation Science, and Athletic Training, University of Kansas Medical Center, Kansas City, Kansas, USA
1 shared publication
Latest Publications
- Open Access
- 30 December 2025
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2680 (9) , 50-74
Evaluating the Effects of Secondary Tasks on Driver Gaze Duration: A Duration Modeling Approach
In simulating drivers’ engagement in secondary tasks in a driving simulator, the objective of this study was to investigate drivers’ gaze duration when performing the tasks, reflecting the time away from the driving scene (commonly referred to as total eyes off road time [TEORT]). Eighty-five participants engaged in the secondary tasks on a driving simulator at University of Kansas, KS. TEORT values were extracted using an eye-tracker for secondary tasks, which were then modeled using a duration modeling approach. Specifically, a correlated grouped random parameters accelerated failure time Weibull duration model with heterogeneity in the mean was developed, capturing the repeated nature of observations and allowing drivers’ gaze durations to be heterogeneous. The random parameter for Age Group 1 suggested that most young drivers (54%) took a shorter time to complete the secondary task, whereas the random parameter for Age Group 3 indicated that about 58% of older drivers took a shorter time to complete the secondary task. These results showed significant heterogeneity in TEORT and mixed effects on safety, whereby drivers’ longer gaze duration indicated risky behavior. The findings of this study confirmed that not only was the driving behavior of one specific age group (Age Group 1) riskier than the others but that the risk within each group was not uniform—individual drivers within an age group exhibited significantly different TEORT, suggesting the need for more nuanced, intragroup analysis. The study highlights how analyzing gaze behavior can aid in designing advanced driving assistance systems to minimize distraction.
- 30 December 2025
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2680 (9) , 50-74
Open AccessEvaluating the Effects of Secondary Tasks on Driver Gaze Duration: A Duration Modeling Approach
In simulating drivers’ engagement in secondary tasks in a driving simulator, the objective of this study was to investigate drivers’ gaze duration when performing the tasks, reflecting the time away from the driving scene (commonly referred to as total eyes off road time [TEORT]). Eighty-five participants engaged in the secondary tasks on a driving simulator at University of Kansas, KS. TEORT values were extracted using an eye-tracker for secondary tasks, which were then modeled using a duration modeling approach. Specifically, a correlated grouped random parameters accelerated failure time Weibull duration model with heterogeneity in the mean was developed, capturing the repeated nature of observations and allowing drivers’ gaze durations to be heterogeneous. The random parameter for Age Group 1 suggested that most young drivers (54%) took a shorter time to complete the secondary task, whereas the random parameter for Age Group 3 indicated that about 58% of older drivers took a shorter time to complete the secondary task. These results showed significant heterogeneity in TEORT and mixed effects on safety, whereby drivers’ longer gaze duration indicated risky behavior. The findings of this study confirmed that not only was the driving behavior of one specific age group (Age Group 1) riskier than the others but that the risk within each group was not uniform—individual drivers within an age group exhibited significantly different TEORT, suggesting the need for more nuanced, intragroup analysis. The study highlights how analyzing gaze behavior can aid in designing advanced driving assistance systems to minimize distraction.
- 17 July 2025
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2679 (10) , 399-410
Deriving Capacity and Critical Density of Basic Freeway Segments
The variability of freeway capacity has been extensively analyzed in numerous studies. Recent research has also focused on the critical density at capacity and its relationship to breakdown occurrences across different segment types to better quantify the quality of service for freeways. Despite recent progress, identifying a representative value for the capacity and critical density of basic freeway segments remains challenging. These segments seldom reach capacity, because breakdowns often result from downstream bottlenecks. This paper aims to estimate the capacity and critical density of basic freeway segments by implementing a methodology that quantifies how variations in the number of merging ramp vehicles affect the total downstream capacity and critical density. The assumption was made that low onramp volumes would cause merge segments to operate in a similar manner to basic segments. The results demonstrated that decreasing the number of merging ramp vehicles reduced the total critical density and increased the total capacity, implying that basic segments may have a lower critical density and higher capacity compared with merge, diverge, and weaving segments.
- 17 July 2025
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2679 (10) , 399-410
Deriving Capacity and Critical Density of Basic Freeway Segments
The variability of freeway capacity has been extensively analyzed in numerous studies. Recent research has also focused on the critical density at capacity and its relationship to breakdown occurrences across different segment types to better quantify the quality of service for freeways. Despite recent progress, identifying a representative value for the capacity and critical density of basic freeway segments remains challenging. These segments seldom reach capacity, because breakdowns often result from downstream bottlenecks. This paper aims to estimate the capacity and critical density of basic freeway segments by implementing a methodology that quantifies how variations in the number of merging ramp vehicles affect the total downstream capacity and critical density. The assumption was made that low onramp volumes would cause merge segments to operate in a similar manner to basic segments. The results demonstrated that decreasing the number of merging ramp vehicles reduced the total critical density and increased the total capacity, implying that basic segments may have a lower critical density and higher capacity compared with merge, diverge, and weaving segments.
- 18 April 2025
- Published by Springer Nature in Transportation
- Vol. 53 (6) , 1-25
Leveraging machine learning algorithms to estimate associated factors of students’ mode choice preferences for out-of-school-time activities
- 18 April 2025
- Published by Springer Nature in Transportation
- Vol. 53 (6) , 1-25
Leveraging machine learning algorithms to estimate associated factors of students’ mode choice preferences for out-of-school-time activities
- 7 June 2024
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2678 (12) , 1329-1345
Leveraging Machine Learning Algorithms to Predict and Analyze Single-Vehicle and Multi-Vehicle Crash Occurrences on Motorways
Road crashes are a common occurrence in many parts of the world, causing significant loss of life, injury, and economic damage. Crashes can be broadly classified into single-vehicle (SV) crashes and multi-vehicle (MV) crashes. Various statistical approaches have been implemented to identify the key factors behind these two types of crashes and it has been concluded that these factors need to be analyzed separately. The dataset for this research included various types of roadway design parameters and traffic conditions. Combinations of three feature-selection techniques, namely ANOVA, correlation matrix, and ExtraTreesClassifier algorithm, were utilized to separately select the appropriate variables for SV and MV crash analysis. Various machine learning (ML) models (e.g., LightGBM, XGBoost, etc.) along with a statistical method (binary logistic regression) have been adopted to predict SV and MV crash occurrences. The results show that gradient boosting-type ML algorithms outperform the remaining prediction models, and the LightGBM was found to be the most powerful in prediction. The LightGBM classifier produced accuracy, ROC_AUC, and avg. F-1 score of 0.75, 0.83, and 0.76, respectively, for MV crashes and 0.76, 0.82, and 0.76, respectively, for SV crashes. The SHapley Additive exPlanations (SHAP) analysis was used to explain how each variable affected the models’ output. The results confirmed that the crash factors associated with SV and MV crashes are different and that some variables have inverse impact. Artificial intelligence and ML can assist transportation professionals in better understanding the causes of SV and MV crashes and advance the process toward Vision Zero.
- 7 June 2024
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2678 (12) , 1329-1345
Leveraging Machine Learning Algorithms to Predict and Analyze Single-Vehicle and Multi-Vehicle Crash Occurrences on Motorways
Road crashes are a common occurrence in many parts of the world, causing significant loss of life, injury, and economic damage. Crashes can be broadly classified into single-vehicle (SV) crashes and multi-vehicle (MV) crashes. Various statistical approaches have been implemented to identify the key factors behind these two types of crashes and it has been concluded that these factors need to be analyzed separately. The dataset for this research included various types of roadway design parameters and traffic conditions. Combinations of three feature-selection techniques, namely ANOVA, correlation matrix, and ExtraTreesClassifier algorithm, were utilized to separately select the appropriate variables for SV and MV crash analysis. Various machine learning (ML) models (e.g., LightGBM, XGBoost, etc.) along with a statistical method (binary logistic regression) have been adopted to predict SV and MV crash occurrences. The results show that gradient boosting-type ML algorithms outperform the remaining prediction models, and the LightGBM was found to be the most powerful in prediction. The LightGBM classifier produced accuracy, ROC_AUC, and avg. F-1 score of 0.75, 0.83, and 0.76, respectively, for MV crashes and 0.76, 0.82, and 0.76, respectively, for SV crashes. The SHapley Additive exPlanations (SHAP) analysis was used to explain how each variable affected the models’ output. The results confirmed that the crash factors associated with SV and MV crashes are different and that some variables have inverse impact. Artificial intelligence and ML can assist transportation professionals in better understanding the causes of SV and MV crashes and advance the process toward Vision Zero.
- 5 February 2024
- Published by Taylor & Francis in Transportation Letters
- Vol. 16 (10) , 1378-1388
Effectiveness of unconventional anti-texting messages displayed on dynamic message signs
- 5 February 2024
- Published by Taylor & Francis in Transportation Letters
- Vol. 16 (10) , 1378-1388
Effectiveness of unconventional anti-texting messages displayed on dynamic message signs
- 9 August 2023
- Published by S. Karger AG in Gerontology
- Vol. 69 (11) , 1307-1314
Driving Automation for Older Adults with Preclinical Alzheimer’s Disease
Introduction: Older adults with preclinical Alzheimer’s disease (AD) show changes in on-road driving performance. The impact of preclinical AD on using automated vehicle (AV) technology is unknown. The aim was to evaluate safety and cognitive workload while operating AV technology in drivers with preclinical AD. Introduction: This cross-sectional study included 40 participants: 19 older adults (age 74.16 ± 4.78; MOCA scores 26.42 ± 2.52) with preclinical AD, evidenced by elevated cortical beta-amyloid; and 21 controls (age 73.81 ± 5.62; MOCA scores 28.24 ± 1.67). All participants completed two scenarios in a driving simulator. Scenario 1 included conditional automation with an emergency event that required a manual take-over maneuver. Scenario 2 was identical but with a cognitive distractor task. Emergency response time was the main safety outcome measure. Cognitive workload was calculated using moment-to-moment changes in pupillary size and converted into an Index of Cognitive Activity (ICA). Mann-Whitney U and independent t tests were used to compare group differences. Results: Emergency response times were similar between drivers with preclinical AD and controls in scenario 1 (20.85 s ± 1.08 vs. 20.52 s ± 3.18; p = 0.83) and scenario 2 (14.83 s ± 7.37 vs. 13.45 s ± 10.43; p = 0.92). Likewise, no differences were found in ICA between drivers with preclinical AD and controls in scenario 1 (0.34 ± 0.08 vs. 0.33 ± 0.17; p = 0.74) or scenario 2 (0.30 ± 0.07 vs. 0.29 ± 0.17; p = 0.93). Conclusions: Older drivers with preclinical AD may safely operate AV technology, without increased response times or cognitive workload. Future on-road studies with AV technology should confirm these preliminary results in drivers with preclinical AD.
- 9 August 2023
- Published by S. Karger AG in Gerontology
- Vol. 69 (11) , 1307-1314
Driving Automation for Older Adults with Preclinical Alzheimer’s Disease
Introduction: Older adults with preclinical Alzheimer’s disease (AD) show changes in on-road driving performance. The impact of preclinical AD on using automated vehicle (AV) technology is unknown. The aim was to evaluate safety and cognitive workload while operating AV technology in drivers with preclinical AD. Introduction: This cross-sectional study included 40 participants: 19 older adults (age 74.16 ± 4.78; MOCA scores 26.42 ± 2.52) with preclinical AD, evidenced by elevated cortical beta-amyloid; and 21 controls (age 73.81 ± 5.62; MOCA scores 28.24 ± 1.67). All participants completed two scenarios in a driving simulator. Scenario 1 included conditional automation with an emergency event that required a manual take-over maneuver. Scenario 2 was identical but with a cognitive distractor task. Emergency response time was the main safety outcome measure. Cognitive workload was calculated using moment-to-moment changes in pupillary size and converted into an Index of Cognitive Activity (ICA). Mann-Whitney U and independent t tests were used to compare group differences. Results: Emergency response times were similar between drivers with preclinical AD and controls in scenario 1 (20.85 s ± 1.08 vs. 20.52 s ± 3.18; p = 0.83) and scenario 2 (14.83 s ± 7.37 vs. 13.45 s ± 10.43; p = 0.92). Likewise, no differences were found in ICA between drivers with preclinical AD and controls in scenario 1 (0.34 ± 0.08 vs. 0.33 ± 0.17; p = 0.74) or scenario 2 (0.30 ± 0.07 vs. 0.29 ± 0.17; p = 0.93). Conclusions: Older drivers with preclinical AD may safely operate AV technology, without increased response times or cognitive workload. Future on-road studies with AV technology should confirm these preliminary results in drivers with preclinical AD.
- 12 April 2023
- Published by Springer Nature in Data Science for Transportation
- Vol. 5 (2) , 1-15
Driver Classification Using Self-reported, Psychophysiological, and Performance Metrics Within a Simulated Environment
Driver classification provides an efficient approach to isolating unique traits associated with specific driver types under various driving conditions. Several past studies use classification to identify behavior and driving styles; however, very few studies employ both measurable physiological changes and environmental factors. This study looked to address the shortcomings in driver classification research using a data-driven approach to assess driving tasks performed under varying mental workloads. Psychophysiological and driving performance changes experienced by drivers when engaged in simulated tasks of varying difficulty were coupled with machine-learning techniques to provide a more accurate estimate of the ground truth for behavioral classification. A driving simulator study consisting of six tasks was carefully designed to incrementally vary complexity between individual tasks. Ninety drivers were recruited to participate in both the subjective and driving components of this research. Positive and Negative Affect Schedule (PANAS), Cognitive Reflection Task (CRT), Interpersonal Reactivity Index (IRI), Empathy Assessment Index (EAI), Psychological Entitlement Scale (PES), 18-point Need for Cognition (NFC), and a basic demographic survey were administered. Time-series clustering using the Dynamic Time Warping (DTW) score within hierarchical clustering was applied to determine difference in driving styles. The use of pre-driving psychometric questionnaires to determine the most suitable metrics for predicting driving style outside of the automobile suggest promising results. The results of the binary logistic regression indicate that an individual’s annual driving mileage, CRT score, age, IRI fantasy score, and EAI affective response score, contribute significantly towards predicting their driving style.
- 12 April 2023
- Published by Springer Nature in Data Science for Transportation
- Vol. 5 (2) , 1-15
Driver Classification Using Self-reported, Psychophysiological, and Performance Metrics Within a Simulated Environment
Driver classification provides an efficient approach to isolating unique traits associated with specific driver types under various driving conditions. Several past studies use classification to identify behavior and driving styles; however, very few studies employ both measurable physiological changes and environmental factors. This study looked to address the shortcomings in driver classification research using a data-driven approach to assess driving tasks performed under varying mental workloads. Psychophysiological and driving performance changes experienced by drivers when engaged in simulated tasks of varying difficulty were coupled with machine-learning techniques to provide a more accurate estimate of the ground truth for behavioral classification. A driving simulator study consisting of six tasks was carefully designed to incrementally vary complexity between individual tasks. Ninety drivers were recruited to participate in both the subjective and driving components of this research. Positive and Negative Affect Schedule (PANAS), Cognitive Reflection Task (CRT), Interpersonal Reactivity Index (IRI), Empathy Assessment Index (EAI), Psychological Entitlement Scale (PES), 18-point Need for Cognition (NFC), and a basic demographic survey were administered. Time-series clustering using the Dynamic Time Warping (DTW) score within hierarchical clustering was applied to determine difference in driving styles. The use of pre-driving psychometric questionnaires to determine the most suitable metrics for predicting driving style outside of the automobile suggest promising results. The results of the binary logistic regression indicate that an individual’s annual driving mileage, CRT score, age, IRI fantasy score, and EAI affective response score, contribute significantly towards predicting their driving style.
- 30 November 2022
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2677 (5) , 590-610
Simulator-Based Framework to Incorporate Driving Heterogeneity via a Biobehavioral Extension to the Intelligent Driver Model
Mathematical models of car-following, lane changing, and gap acceptance are extensively used to describe behavioral variability when driving. Car-following, in relation to the intelligent driver model (IDM), was the primary component of this research. This research was targeted toward developing a framework to incorporate driver behavioral variability using methodologies adapted from the cognitive and physiological sciences. The main goal was to fuse driving variables such as preferred gap, speed, jerk, and acceleration, together with continuous biobehavioral variables of engagement level, mental workload, situation awareness, and other static driver properties (i.e., age, experience, and driving history), to introduce biobehavioral heterogeneity into the IDM. Incorporating non-conventional tasks during car-following, such as distracted driving, was also a key component of this research. Additionally, assessing the effectiveness of incorporating group-based biobehavioral and driving performance traits rather than individual-level traits into the IDM was also explored. Ninety drivers were recruited to validate the framework using a fixed-base simulator. The scenarios were designed to capture the performance and cognitive parameters when subject to varying task complexities. A biobehavioral extension to the IDM was developed by grouping individual driving performance and behavioral traits. The extended model was successfully validated and found to introduce driving heterogeneity while enhancing the IDM’s car-following modeling accuracy for the sample population. The established methods also serve as a key step toward the inclusion of individual/group-level traits that not only consider driving performance but also harvest cognitive and biological processes that directly affect car-following.
- 30 November 2022
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2677 (5) , 590-610
Simulator-Based Framework to Incorporate Driving Heterogeneity via a Biobehavioral Extension to the Intelligent Driver Model
Mathematical models of car-following, lane changing, and gap acceptance are extensively used to describe behavioral variability when driving. Car-following, in relation to the intelligent driver model (IDM), was the primary component of this research. This research was targeted toward developing a framework to incorporate driver behavioral variability using methodologies adapted from the cognitive and physiological sciences. The main goal was to fuse driving variables such as preferred gap, speed, jerk, and acceleration, together with continuous biobehavioral variables of engagement level, mental workload, situation awareness, and other static driver properties (i.e., age, experience, and driving history), to introduce biobehavioral heterogeneity into the IDM. Incorporating non-conventional tasks during car-following, such as distracted driving, was also a key component of this research. Additionally, assessing the effectiveness of incorporating group-based biobehavioral and driving performance traits rather than individual-level traits into the IDM was also explored. Ninety drivers were recruited to validate the framework using a fixed-base simulator. The scenarios were designed to capture the performance and cognitive parameters when subject to varying task complexities. A biobehavioral extension to the IDM was developed by grouping individual driving performance and behavioral traits. The extended model was successfully validated and found to introduce driving heterogeneity while enhancing the IDM’s car-following modeling accuracy for the sample population. The established methods also serve as a key step toward the inclusion of individual/group-level traits that not only consider driving performance but also harvest cognitive and biological processes that directly affect car-following.