About

Lily Elefteriadou is a professor at the Engineering School of Sustainable Infrastructure & Environment. The provided page text does not include specific details about her research focus, background, or key contributions. Therefore, a detailed biography cannot be generated from the available information.

Research topics

• Computer Science
• Engineering
• Transport engineering
• Real-time computing
• Simulation
• Telecommunications
• Operations research

Selected publications

• Traffic Impacts of Autonomous Shuttles and Their Inclusion in HCM Urban Streets Analysis

  SSRN Electronic Journal · 2026-01-01

  preprintOpen access
  PublisherDOI

• Development of a New Methodology for Evaluating Arterial Weaving Sections Including All Traffic Movements

  Transportation Research Record Journal of the Transportation Research Board · 2026-02-01

  article

  Arterial weaving segments present unique operational challenges because of frequent lane changes between origin–destination (OD) pairs. Existing methodologies, such as those in the Highway Capacity Manual (HCM7) and NCHRP 15-66, are formulated exclusively for through movements and therefore cannot accurately predict running speeds for vehicles performing weaving or turning maneuvers, which may represent more than half of the total traffic in these facilities. This study develops and validates a new methodology that estimates running speeds by OD and by segment, explicitly capturing the effects of lane utilization, turbulence index, and weaving ratio. Using the NCHRP 15-66 dataset, which includes drone observations from fifteen field sites and more than 1,000 calibrated microsimulation runs, models were estimated using feasible generalized least squares and evaluated through 10-fold cross-validation. The analysis confirmed that vehicles associated with different ODs exhibit statistically distinct running speeds, reinforcing the need for OD-specific modeling. The proposed formulation achieved an average RMSE of 5.27 mph across all OD movements and produced 29.6% lower RMSE than NCHRP 15-66 for through movements under high-demand conditions (> 450 vehicles per hour per lane). These findings demonstrate that OD-specific and regime-based modeling provides a more accurate and transferable representation of arterial weaving operations than existing analytical approaches.

  PublisherDOI

• Calibration and Evaluation of Car-Following Models for Autonomous Shuttles Using a Novel Multi-Criteria Framework

  ArXiv.org · 2026-02-12

  articleOpen accessSenior author

  Autonomous shuttles (AS) are fully autonomous transit vehicles with operating characteristics distinct from conventional autonomous vehicles (AV). Developing dedicated car-following models for AS is critical to understanding their traffic impacts; however, few studies have calibrated such models with field data. More advanced machine learning (ML) techniques have not yet been applied to AS trajectories, leaving the potential of ML for capturing AS dynamics unexplored and constraining the development of dedicated AS models. Furthermore, there is a lack of a unified framework for systematically evaluating and comparing the performance of car-following models to replicate real trajectories. Existing car-following studies often rely on disparate metrics, which limit reproducibility and performance comparability. This study addresses these gaps through two main contributions: (1) the calibration of a diverse set of car-following models using real-world AS trajectory data, including eight machine learning algorithms and two physics-based models; and (2) the introduction of a multi-criteria evaluation framework that integrates measures of prediction accuracy, trajectory stability, and statistical similarity, which provides a generalizable methodology for a systematic assessment of car-following models. Results indicated that the proposed calibrated XGBoost model achieved the best overall performance. Sequential model type, such as LSTM and CNN, captured long-term positional stability but were less responsive to short-term dynamics. LSTM and CNN captured long-term positional stability but were less responsive to short-term dynamics. Traditional models (IDM, ACC) and kernel methods showed lower accuracy and stability than most ML models tested.

  PublisherOA PDF

• Calibration and Evaluation of Car-Following Models for Autonomous Shuttles Using a Novel Multi-Criteria Framework

  Open MIND · 2026-02-12

  preprintSenior author

  Autonomous shuttles (AS) are fully autonomous transit vehicles with operating characteristics distinct from conventional autonomous vehicles (AV). Developing dedicated car-following models for AS is critical to understanding their traffic impacts; however, few studies have calibrated such models with field data. More advanced machine learning (ML) techniques have not yet been applied to AS trajectories, leaving the potential of ML for capturing AS dynamics unexplored and constraining the development of dedicated AS models. Furthermore, there is a lack of a unified framework for systematically evaluating and comparing the performance of car-following models to replicate real trajectories. Existing car-following studies often rely on disparate metrics, which limit reproducibility and performance comparability. This study addresses these gaps through two main contributions: (1) the calibration of a diverse set of car-following models using real-world AS trajectory data, including eight machine learning algorithms and two physics-based models; and (2) the introduction of a multi-criteria evaluation framework that integrates measures of prediction accuracy, trajectory stability, and statistical similarity, which provides a generalizable methodology for a systematic assessment of car-following models. Results indicated that the proposed calibrated XGBoost model achieved the best overall performance. Sequential model type, such as LSTM and CNN, captured long-term positional stability but were less responsive to short-term dynamics. LSTM and CNN captured long-term positional stability but were less responsive to short-term dynamics. Traditional models (IDM, ACC) and kernel methods showed lower accuracy and stability than most ML models tested.

  DOI

• Estimation of Passenger Car Equivalents on Basic Freeway Segments from Field-Observed Traffic Data

  Transportation Research Record Journal of the Transportation Research Board · 2025-07-10

  articleSenior author

  Passenger car equivalents (PCEs) represent the effects of heavy vehicles on traffic operations. PCE estimation is based on equating the passenger-car-only flow rate to the mix-fleet flow rate such that it results in the same performance for the selected measure. Most existing methods rely on simulation to generate PCE values. However, this approach generates PCEs using the truck characteristics of the simulation rather than local field conditions and heavy vehicle types. Most existing methods estimate the marginal effect of adding one specific truck in the traffic stream assuming relatively high volumes, which results in higher PCE values. This study proposes a new method for estimating PCEs using field-observed traffic data considering the impact of each truck type for a broad set of flows, not just their marginal effect at high flows. Based on density equivalency, the maximum likelihood estimation is used to estimate the means and the variances of the PCE values from traffic data collected at 10 sites on the national motorways of Thailand. When aggregated across all observed percentages of heavy vehicles and flow rates, the PCE values are 1.46 (trucks with length between 5.2 and 13.0 m) and 2.08 (trucks with length of 13.1 m or longer). For low to medium percentage of trucks, the PCEs become lower after the onset of oversaturation. However, for high percentages of trucks the PCEs tend to be higher after the onset of oversaturation. The proposed methodology can be applied using data from other locations to estimate the corresponding PCEs from field-observed data.

  PublisherDOI

• Exploratory Driving Performance and Car-Following Modeling for Autonomous Shuttles Based on Field Data

  IEEE Transactions on Intelligent Transportation Systems · 2025-04-09 · 3 citations

  articleSenior author

  Autonomous shuttles (AS) operate in several cities and have shown potential to improve the public transport network. However, there is no car-following model that is based on field data and allows decision-makers (planners, and traffic engineers) to assess and plan for AS operations. To fill this gap, this study collected field data from AS operations, analyzed their driving performance, and suggested changes in the AS trajectory model to improve passenger comfort. A sample was collected with over 4,000 seconds of data of AS following a conventional car (human driver). The sample contained GPS positions from both AS and conventional vehicles. Latitude and longitude coordinates were used to calculate the speed, acceleration, and jerk of the leader and follower. The data analyses indicated that AS has higher jerk values that may impact the passengers’ comfort. Several existing models were evaluated, and the researchers concluded that the calibrated ACC model resulted in lower errors for AS spacing and speed. The results of the calibration indicate that the AS exhibits lower peak acceleration and higher deceleration than those found in calibrated parameters of autonomous vehicle models from other studies.

  PublisherDOI

• Simulation-Based Traffic Impacts of Autonomous Shuttle Deployment on Urban Signalized Arterials

  SSRN Electronic Journal · 2025-01-01

  preprintOpen accessSenior author
  PublisherDOI

• IEEE INTELLIGENT TRANSPORTATION SYSTEMS SOCIETY

  IEEE Transactions on Intelligent Transportation Systems · 2024-02-01

  articleOpen access
  PublisherOA PDFDOI

• IEEE INTELLIGENT TRANSPORTATION SYSTEMS SOCIETY

  IEEE Transactions on Intelligent Transportation Systems · 2024-04-01

  articleOpen access
  PublisherOA PDFDOI

• Measuring Mobility: Quality, Quantity, Utilization, and Accessibility

  Springer optimization and its applications · 2024-01-01

  book-chapter1st authorCorresponding
  PublisherDOI

Recent grants

• CPS: TTP Option: Synergy: Traffic Signal Control with Connected and Autonomous Vehicles in the Traffic Stream

  NSF · $1.3M · 2015–2021

Frequent coauthors

• Cristina Olaverri-Monreal

  Virginia Tech

  123 shared

• Antonio Bucchiarone

  78 shared

• Kyoungho Ahn

  Virginia Tech

  78 shared

• Hesham Rakha

  Virginia Tech

  78 shared

• Bart De Schutter

  78 shared

• Simona Sacone

  University of Genoa

  78 shared

• Hussein Dia

  78 shared

• Javier Barria

  Virginia Tech

  78 shared

Awards & honors

• 2023 AAAS Lifetime Fellow
• 2021 ARTBA S.S. Steinberg Award
• 2020 Distinguished Alumni Award, Center for Urban Science an…
• 2019 ASCE Harland Bartholomew Award
• 2018 University of Florida Term Professorship Award