Team Buck • DataFest 2026
🏥

The Crisis at Stormont Vail Health

Patients from transit-poor neighborhoods miss preventative care and escalate to the ER. Discharged patients are trapped in beds because public transit can't get them home. One bi-directional shuttle fleet solves both problems on a single budget.

📊ER Ratio by Vulnerability Tier
⏱Average Stay Duration by Shuttle Zone

Transit Vulnerability Index

Each Census Block scored on transit isolation, SDoH transportation needs, financial strain, and elderly demographics. Red zones indicate the most transit-vulnerable communities. Bus routes overlaid in blue.

Key Finding: 43 of 141 Census Blocks (30.5%) have zero bus stops within 800 meters of walking distance. These areas force residents to walk over 1.6 km just to reach public transit — a near-impossible task for elderly and disabled patients.

Shuttle Zone Clusters

DBSCAN clustering groups adjacent high-vulnerability blocks into operationally efficient shuttle zones. Each zone can be served by a single shuttle route.

Inbound Morning Route

Turn-by-turn driving routes calculated via Open Source Routing Machine (OSRM) connecting SVH to designated high-vulnerability Shuttle Zones.

Outbound Discharge Manifest

Medically cleared patients living in Shuttle Zones are dynamically flagged. By prioritizing patients with extreme Length of Stay (LOS) durations, the shuttle aggressively clears high-risk bed bottlenecks.

Patient ID Census Block Shuttle Zone TOTAL STAY (HOURS) Status Action

Dead Time & Transit Correlation

Analyzing the relationship between Transit Vulnerability Scores and hospital stay durations to quantify the operational impact of transit barriers.

📉TVS vs Stay Duration (by Census Block)
📊Stay Duration by Zone Assignment
🔮Dynamically Forecasted Discharge Surges (ML Baseline)

Historical time-series forecasting predicting exactly when patients will discharge in each Shuttle Zone, enabling precision predictive dispatch.

🔬Statistical Analysis