Predicting real-time traffic restoration time based on the estimated traffic state
Bibliographic record
Abstract
Ensuring efficient traffic management after crash occurrence is crucial for minimizing fatalities, avoiding secondary crashes, reducing congestion, guiding traffic to alternative routes. Hence, it is important to know the time required to return the traffic state to a normal operating condition after such incidents in real time. In this paper, we present a new approach to predict the traffic restoration time after a crash occurrence based on the estimated traffic state using real-time data. The contribution of this study is threefold: first, the study developed models to predict the traffic state after a crash; second, the study predicted the traffic restoration time based on the estimated post-crash traffic state; third, the study applied three-step validation techniques to evaluate the performance of the developed approach and compare it with crash clearance time. To accomplish these tasks, we considered a 220 miles section of Interstate-75 of Florida, USA. Traffic, crash, weather, and emergency facility data from 2017 to 2019 were collected recording 24,448 events (4,939 crashes and 19,509 non-crash events) and 65 real-time features. A total of eight traffic state prediction models with high accuracy were developed using the XGBoost machine learning technique. The estimated traffic state was used to calculate the post-crash congestion level. Then pre-crash and post-crash congestion levels were compared to determine the time when traffic returned to normal operating conditions. The estimated traffic restoration time was validated by investigating the post-crash speed volume relationship, comparing it with the actual crash clearance time data, and finding the cosine similarity index. The proposed framework enables traffic management agencies to correctly predict the impact of a crash based on predicted traffic. To the best of the knowledge of the authors the developed approach to estimate traffic restoration time is a novel idea and has the potential to contribute to real-time traffic management after a crash.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".