Abstract
This study evaluates the operational effects of Dynamic Part-Time Shoulder Use (D-PTSU) through a before–after analysis of weekday morning peak conditions (06:00–09:00) using field data from January–June 2021 and January–June 2025 on the Ohio I-275 westbound corridor. A hybrid Graph Convolutional Network–Long Short-Term Memory (GCN–LSTM) model was developed to impute missing Automatic Traffic Recorder (ATR) speed-bin data for the before period. Free-flow baselines were derived from off-peak observations to compute the Buffer Index (BI), Planning Time Index (PTI), and Misery Index (MI), while paired t-tests and Wilcoxon signed-rank tests evaluated statistical significance of observed changes. A Bayesian Structural Time Series (BSTS) counterfactual model was employed to project corridor performance under the no-intervention scenario. Results show that average travel times increased across most intervals due to corridor demand growth from parallel route diversion and regional traffic growth. Despite these increases, reliability improved greatly with BI peaks previously exceeding 0.5 fell below 0.28 following implementation, with 93.3% of all BI intervals showing statistically significant changes. PTI showed peak-period reductions of up to 34.0% on the most congested weekdays. MI declined at peak intervals across 93.3% of examined time intervals, demonstrating suppression of extreme delay events at the tail of the travel time distribution. BSTS analysis attributed a 58.69% causal reduction in BI and a 52.67% causal reduction in MI below their no-intervention projections. Throughput increased by 1000–1500 veh/h, raising effective discharge capacity while stabilizing flow during peak conditions. These findings indicate that the D-PTSU improved travel-time reliability, mitigated peak-hour variability, and increased corridor discharge capacity in the observed period.
| Original language | English |
|---|---|
| Article number | 102016 |
| Journal | Transportation Research Interdisciplinary Perspectives |
| Volume | 37 |
| DOIs | |
| State | Published - May 1 2026 |
UN SDGs
This output contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 11 Sustainable Cities and Communities
Keywords
- Active traffic management
- Buffer index
- Dynamic Part-Time Shoulder Use
- Planning Time Index
- Travel time reliability
Fingerprint
Dive into the research topics of 'Evaluating the Operational Impacts of Dynamic Part-Time Shoulder Use'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver