MCMei Chen

Mei Chen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mei Chen has a total of 21 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Real-time crash likelihood assessment using crowdsourced Waze incident alerts

    by Eugene Boasiako Antwi, Xu Zhang, Mei Chen

    1. Traffic crashes arise from multifaceted interactions among several factors (Mannering et al., 2016) and may result in fatalities and serious injuries. Apart from the human toll, crashes signific...

  • Evaluation of Third-Party Turning-Movement-Count Data: A Kentucky Case Study

    by Siavash Taherinavid, Xu Zhang, Eugene Antwi Boasiako, Mei Chen

    Turning-movement counts (TMCs) are among the fundamental datasets in the transportation industry. This type of traffic count is regularly conducted and utilized as a core input to intersection design, signal timing, and more. However, traditional methods for collecting this data are expensive, labor-intensive, and time-consuming. Thus, the data collected from specific locations within short time periods does not necessarily capture all traffic variation. In recent years, advancements in tracking technologies have enabled third-party vendors to passively generate TMCs for different locations and time intervals. This study primarily focuses on evaluating TMCs estimated by a third-party vendor through comparison with ground-truth data. Actual TMCs are obtained from manually collected data for several planning studies across Kentucky between 2016 and 2024. Moreover, estimated counts are acquired from the StreetLight platform for these same locations. TMCs are evaluated across different functional classifications (FCs) and area types to assess their impact on the StreetLight data accuracy. The overall comparison of TMCs yields Pearson correlation coefficients greater than 0.75 and mean absolute percentage errors (MAPE) greater than 40%. When analyzed by area type and FC, urban intersections with higher functional importance show stronger agreement with observed counts than rural intersections show. The analysis suggests that variations in annual average daily traffic (AADT) had a measurable impact on estimation performance, with errors generally tending to decrease as AADT increases. Results of this study will help provide transportation agencies with a better understanding of the accuracy and utility of estimated data from third-party vendors, specifically TMCs.

  • A Graph-Based Approach to Conflating State Department of Transport and Probe Data Vendor Roadway Networks of Varying Resolution

    by Xu Zhang, Mei Chen

    Probe vehicle data has become increasingly prevalent in transportation applications over the last two decades. Despite its widespread use, many transportation agencies struggle to integrate this data with their existing data sets because of disparities between the proprietary roadway networks used by data vendors and those used by the agencies themselves. This paper presents a fully automated and efficient conflation process to address the challenge of conflating third-party probe data networks with the state Department of Transport Linear Referencing System networks. The process leverages graph data structures to effectively account for network connectivity and link directionality, which have been a challenge for existing conflation methods. In addition, it incorporates new spatial metrics to enhance matching accuracy. The process is developed in Python, which ensures full transparency, traceability, and reproducibility. It successfully conflates the Kentucky Transportation Cabinet’s all roads network with the high-resolution HERE network in under 75 min, achieving a substantial reduction in execution time compared with the previous method. The process also conflates the Traffic Message Channel network covering the National Highway System roads in Kentucky in just 3 min. A comprehensive review of conflation results demonstrates its accuracy, particularly in complex areas such as interstate junction areas, double-crossover diamond interchanges, and roundabouts. This paper provides transportation agencies with a robust tool for integrating disparate roadway networks, allowing them to better utilize third-party probe data in transportation applications.

  • Open Access

    Sustainable Safety Planning on Two-Lane Highways: A Random Forest Approach for Crash Prediction and Resource Allocation

    by Fahmida Rahman, Cidambi Srinivasan, Xu Zhang, Mei Chen

    During the safety planning stage, accurate crash prediction tools are critical for prioritizing countermeasures and allocating resources effectively. Traditional statistical approaches, while long applied in this field, often depend on distributional assumptions that may introduce bias and limit model accuracy. To address these issues, studies have started exploring Machine Learning (ML)-based techniques for crash prediction, particularly for higher functional class roads. However, the application of ML models on two-lane highways remains relatively limited. This study aims to develop an approach to integrate traffic, geometric, and critically, speed-based factors in crash prediction using Random Forest (RF) and SHapley Additive exPlanations (SHAP) techniques. Comparative analysis shows that the RF model improves crash prediction accuracy by up to 25% over the traditional Zero-Inflated Negative Binomial model. SHAP analysis identified AADT, segment length, and average speed as the three most influential predictors of crash frequency, with speed emerging as a key operational factor alongside traditional exposure measures. The strong influence of speed in the RF–SHAP results depicts its critical role in the safety performance of two-lane highways and highlights the value of incorporating detailed operating characteristics into crash prediction models. Overall, the proposed RF–SHAP framework advances roadway safety assessment by offering both predictive accuracy and interpretability, allowing agencies to identify high-impact factors, prioritize countermeasures, and direct resources more efficiently. In doing so, the approach supports sustainable safety management by enabling evidence-based investments, promoting optimal use of limited transportation funds, and contributing to safer, more resilient mobility systems.

  • Open Access

    Beta Distribution Learning for Reliable Roadway Crash Risk Assessment

    by Ahmad Elallaf, Nathan Jacobs, Xinyue Ye, Mei Chen, Gongbo Liang

    Roadway traffic accidents represent a global health crisis, responsible for over a million deaths annually and costing many countries up to 3% of their GDP. Traditional traffic safety studies often examine risk factors in isolation, overlooking the spatial complexity and contextual interactions inherent in the built environment. Furthermore, conventional Neural Network-based risk estimators typically generate point estimates without conveying model uncertainty, limiting their utility in critical decision-making. To address these shortcomings, we introduce a novel geospatial deep learning framework that leverages satellite imagery as a comprehensive spatial input. This approach enables the model to capture the nuanced spatial patterns and embedded environmental risk factors that contribute to fatal crash risks. Rather than producing a single deterministic output, our model estimates a full Beta probability distribution over fatal crash risk, yielding accurate and uncertainty-aware predictions--a critical feature for trustworthy AI in safety-critical applications. Our model outperforms baselines by achieving a 17-23% improvement in recall, a key metric for flagging potential dangers, while delivering superior calibration. By providing reliable and interpretable risk assessments from satellite imagery alone, our method enables safer autonomous navigation and offers a highly scalable tool for urban planners and policymakers to enhance roadway safety equitably and cost-effectively.

  • Open Access

    Improving Crash Data Quality with Large Language Models: Evidence from Secondary Crash Narratives in Kentucky

    by Xu Zhang, Mei Chen

    This study evaluates advanced natural language processing (NLP) techniques to enhance crash data quality by mining crash narratives, using secondary crash identification in Kentucky as a case study. Drawing from 16,656 manually reviewed narratives from 2015-2022, with 3,803 confirmed secondary crashes, we compare three model classes: zero-shot open-source large language models (LLMs) (LLaMA3:70B, DeepSeek-R1:70B, Qwen3:32B, Gemma3:27B); fine-tuned transformers (BERT, DistilBERT, RoBERTa, XLNet, Longformer); and traditional logistic regression as baseline. Models were calibrated on 2015-2021 data and tested on 1,771 narratives from 2022. Fine-tuned transformers achieved superior performance, with RoBERTa yielding the highest F1-score (0.90) and accuracy (95%). Zero-shot LLaMA3:70B reached a comparable F1 of 0.86 but required 139 minutes of inference; the logistic baseline lagged well behind (F1:0.66). LLMs excelled in recall for some variants (e.g., GEMMA3:27B at 0.94) but incurred high computational costs (up to 723 minutes for DeepSeek-R1:70B), while fine-tuned models processed the test set in seconds after brief training. Further analysis indicated that mid-sized LLMs (e.g., DeepSeek-R1:32B) can rival larger counterparts in performance while reducing runtime, suggesting opportunities for optimized deployments. Results highlight trade-offs between accuracy, efficiency, and data requirements, with fine-tuned transformer models balancing precision and recall effectively on Kentucky data. Practical deployment considerations emphasize privacy-preserving local deployment, ensemble approaches for improved accuracy, and incremental processing for scalability, providing a replicable scheme for enhancing crash-data quality with advanced NLP.

  • Open Access

    An Upper Partial Moment Framework for Pathfinding Problem Under Travel Time Uncertainty

    by Xu Zhang, Mei Chen

    Route planning under uncertain traffic conditions requires accounting for not only expected travel times but also the risk of late arrivals. This study proposes a mean-upper partial moment (MUPM) framework for pathfinding that explicitly considers travel time unreliability. The framework incorporates a benchmark travel time to measure the upper partial moment (UPM), capturing both the probability and severity of delays. By adjusting a risk parameter (θ), the model reflects different traveler risk preferences and unifies several existing reliability measures, including on-time arrival probability, late arrival penalty, and semi-variance. A bi-objective model is formulated to simultaneously minimize mean travel time and UPM. Theoretical analysis shows that the MUPM framework is consistent with the expected utility theory (EUT) and stochastic dominance theory (SDT), providing a behavioral foundation for the model. To efficiently solve the model, an SDT-based label-correcting algorithm is adapted, with a pre-screening step to reduce unnecessary pairwise path comparisons. Numerical experiments using GPS probe vehicle data from Louisville, Kentucky, USA, demonstrate that varying θ values lead to different non-dominated paths. Lower θ values emphasize frequent small delays but may overlook excessive delays, while higher θ values effectively capture the tail risk, aligning with the behavior of risk-averse travelers. The MUPM framework provides a flexible, behaviorally grounded, and computationally scalable approach to pathfinding under uncertainty. It holds strong potential for applications in traveler information systems, transportation planning, and network resilience analysis.

  • Estimating Urban Arterial Traffic Speed Distributions Using XGBoostLSS

    by Eugene Boasiako Antwi, Xu Zhang, Mei Chen

    The distribution of traffic speed is a critical input in many practical transportation planning and management applications. Accurate estimation of speed distributions facilitates the understanding of traffic patterns, the evaluation of roadway performance, and the development of effective traffic control strategies. However, given the dynamic nature of traffic conditions, particularly on arterials typified by multimodal speed distributions, modeling of the distribution of traffic speed can be challenging. In this study, we propose a novel approach using XGBoostLSS, a tree-based model, for the simultaneous prediction of percentile speeds on urban arterials for each hour of the day. The model was trained and tested on a comprehensive data set, encompassing various urban arterial characteristics, traffic volume, and historical GPS-probe speed data. Our XGBoostLSS demonstrated good accuracy on percentile speed prediction compared with other approaches in literature. Also, the study identified crucial predictor variables influencing speed predictions, including speed limit, signal and stop sign density, peak capacity, right shoulder width, segment length, and crash history. Positive relationships were observed between speed limit, peak capacity, right shoulder width, segment length, and traffic speed, while increased signal and stop sign density and directional hourly traffic volume were associated with decreased traffic speed. Also, our model interpretation suggests that a minimum signal or stop sign spacing of 0.2 mi tended to improve traffic speed. The results from this study offer valuable insights for transportation professionals in estimating urban arterial traffic speed distributions, facilitating effective traffic management, and improving roadway safety.