Real-time Parking Data Hub and Analytics Platform
แพลตฟอร์มข้อมูลที่จอดรถและอุปกรณ์หน้างานแบบเรียลไทม์
Streams events from five groups of parking field devices into a Kafka-based data hub for instant alerts, a live floor map of 5,908 bays, and customer behaviour analytics.
About the project
Collects device status and events from every point in a parking facility into a central data hub, serving both instant alerts and historical analysis.
Five event groups
- Barrier gate — gate status, entries and exits, and counts at each gate
- Intercom — voice and video recordings stored in media storage
- Driver face — driver face capture
- Dispenser — low-supply warnings for thermal paper and ribbon in ticket dispensers
- Occupancy — real-time availability, overall and by zone
Data flow
Each device's producer publishes events through Apache Kafka brokers into the data hub, and a real-time messaging service forwards them to apps and notifications. The platform runs on a Kubernetes cluster with separate production and UAT environments, with an API proxy connecting back-end services to front-end apps.
Dashboards and analytics
- A per-floor bay map covering all 5,908 bays, with zone summaries
- Bay entries and exits by building, plus outdoor traffic by speed and lane
- Customer behaviour from licence plate recognition (LPR): vehicle mix, average stay, visitor versus member revenue, and frequent users
- Network and availability monitoring at the 99.98% level
Technologies
Apache Kafka · KSQL · MongoDB · MSSQL · Kubernetes · Realtime Messaging · Push Notification · Barrier Gate Integration
Gallery · 5 images
