MétaCan
Menu
Back to cohort
Record W4413289654 · doi:10.1016/j.trip.2025.101575

Advancing winter road maintenance: An AI-driven web platform for real-time road condition monitoring and spatial analysis

2025· article· en· W4413289654 on OpenAlex
Michael Urbiztondo, Mingjian Wu, Tae J. Kwon

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueTransportation Research Interdisciplinary Perspectives · 2025
Typearticle
Languageen
FieldEngineering
TopicInfrastructure Maintenance and Monitoring
Canadian institutionsMcGill UniversityUniversity of Alberta
FundersIowa Department of Transportation
KeywordsWeb applicationTransport engineeringComputer scienceEngineeringWorld Wide Web

Abstract

fetched live from OpenAlex

Winter weather conditions pose significant challenges for transportation agencies, impacting road safety, traffic flow, and winter road maintenance (WRM) operations. Traditional methods for monitoring road surface conditions (RSCs) often involve time-consuming processes that require significant personnel. To address these challenges and maximize the utility of existing infrastructure, this paper presents a web-based system for real-time RSC monitoring. The system combines convolutional neural networks (CNNs) for RSC classification, a novel Nested Indicator Kriging (NIK) method for spatial interpolation, and modern web technologies to provide an intuitive interface. The system seamlessly integrates CNN models for real-time classifications using automated vehicle location (AVL) and road weather information system (RWIS) imagery. The NIK method enhances spatial coverage by classifying multiple RSC categories through two layers: the first identifies basic road conditions as bare or non-bare, while the second discriminates between more complex states, such as partially or fully snow-covered. Validated through simulations using historical data, the integrated AVL CNN model achieved a training accuracy of 99.89% and a validation accuracy of 94.62% during training, while the RWIS model reached a maximum accuracy of 98.46% and an F1 Score of 97.19%. Furthermore, the NIK method showed cross-validation accuracies averaging 73.5% for the first layer, and 86.0% for the second layer. This unified system represents an advancement in WRM decision support by automating RSC classifications and closing gaps in spatial data coverage, thus improving the efficiency and sustainability of operations and enhancing the ability of safety professionals and operators to respond to roadway hazards in real-time.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.606
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.001
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.013
GPT teacher head0.354
Teacher spread0.341 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it