TransDec

Description

TransDec (Transportation Decision-Making) is an end-to-end, data-driven system that enables spatiotemporal decision-making queries on transportation systems with real-time and historical data. TransDec can process massive amounts of GPS trajectories from public transportation vehicles to analyze a variety of performance metrics of public transportation systems, such as on-time performance, bus bunching, and bus delay prediction as well as performing traffic forecasting and intelligent routing.

Within TransDec, we built an Archived Data Management System (ADMS) for collecting, archiving, and fusing very large-scale and high-resolution (both spatial and temporal) traffic sensor data. ADMS started collecting data from the Los Angeles County transportation network since 2011. This dataset includes:

  • Inductive-Loop Traffic Detectors: 9,300 sensors located on the highways and arterial streets of Los Angeles County (covering 5,400 miles, cumulatively) sensing occupancy, volume, and speed at the rate of 1 reading/sensor/min.
  • Bus: 2,036 buses operating on 145 different routes in Los Angeles County, with sensors recording the geospatial location of each bus every 1 minutes, next-stop information relative to current location, and delay information relative to predefined time-tables.
  • Rail: 35 trains operating on 4 different lines in Los Angeles County, with sensors recording geospatial location information of each train at every 2 minutes.
  • Ramp meter: 1,851 ramp meters that regulate the flow of traffic entering into highways according to current traffic conditions.
  • Event: event data includes detail free-text format information (e.g., severity, street information, and the number of casualties) about events such as collisions and traffic hazards.

Demo

Video Traffic Analytics
Real-time Video Traffic Analytics
Accident Monitoring & Analysis
Time-Dependent Fleet Routing

Dashboards

TransitDashboard
AccidentDashboard
CongestionDashboard

Interviews

IMSC is a research center that focuses on data-driven solutions for real-world applications by applying multidisciplinary research in the area of data science.