turning big data into the big picture...turning big data into the big picture pete comps conduent,...
TRANSCRIPT
Turning Big Data into the Big Picture
Pete CompsConduent, Director, Business Development
Chicago, IL
Xavier DefrenneConduent, Director, Fare Collection
Atlanta, GA
Traditional Reports
• Tabular− Excel
− Business Intelligence tools
− Tables, rows and columns
− Occasional graph
• Geographic Information Systems (GIS)− Require geo-located data (e.g., stop location)
− Typically produce static images
− Limited use
Typical Sources of Transit “Big Data”
• Fleet Management (CAD/AVL) systems
• Automatic Passenger Counters
• Fare collection systems
• GTFS
• Maintenance Management Systems
• Mobility Enablers− TNC (Uber, Lyft)
− Scooters (Lime, etc.)
− Bikeshare
Challenge:
Inconsistent content and format
Data Consolidation
• Use common data elements (e.g., timestamps) to rationalize data from multiple sources
• Build a “data warehouse” that provides a ready source of consolidated data
• Automate data mining process so consolidated data is always current
MAP
Fare Collection
Passenger Counter
GTFS
Maintenance Management
TNC (Uber, Lyft)Scooters
Bikeshare
CAD / AVL
Mobility Analytics Platform (MAP)
MAP helps with fact-based decision making and planning
• “Big Data” mining and consolidation
• Easily defined query parameters
• Analysis using AI
• Visualization− Geographic
− Interactive
− 4th Dimension with scenario playing (am-pm, hour per hour, vehicle progression)
• Simulation
MAP Views
MAP Views
MAP Views
MAP Views
Video
• Example with Houston
• 4 Views− Validations across the network
− Vehicle load
− Schedule adherence (early/late)
− Origin/Destination
Vehicle Load
4
Vehicle Load
Observation of real stop times
(Tap on validations)
Alignment of real / theoretical journey
Passenger alighting forecasting
Inference of vehicle load
3
2
4
1
4
Key Presentation Take-Aways
• “Big Data” mining and consolidation allows comprehensive analytics
• Interactive visualization enhances analysis− Geographic
− 4th dimension animation