Overview / Usage

PeopleSense.AI, award-winning solution and crowd-analytics tool, specially developed for the rail and metro sector, integrates AI-driven analytics with multi-source data aggregation to provide real-time and predictive insights into passenger movement patterns. PeopleSense® Data Lake ingests and analyzes data from various sources to enhance crowd management, utilization efficiency and passenger safety.

Methodology / Approach

Problem: Metro and Rapid Rail agencies lack accurate real-time, data-driven insights into crowd occupancy, making it challenging to improve resource allocation and optimize train consist utilization for continuously increasing passenger satisfaction. Overcrowding during peak hours heightens safety risks and a lack of predictive capabilities limits agencies' ability to proactively manage demand fluctuations and respond effectively to emergencies.

Needs: To address these challenges, metro and rail agencies require intelligent and scalable approaches that provide real-time crowd occupancy insights, optimize capacity planning, and enable predictive and progressive demand management. Further, by leveraging AI-powered crowd analytics, these agencies can ensure efficient train consist utilization, proactive resource allocation, and improved passenger experiences.

PeopleSense® Key Offerings

PeopleSense.AI, award-winning solution and crowd-analytics tool, specially developed for the rail and metro sector, integrates AI-driven analytics with multi-source data aggregation to provide real-time and predictive insights into passenger movement patterns. PeopleSense® Data Lake ingests and analyzes data from various sources to enhance crowd management, utilization efficiency and passenger safety.

Data Lake Sources:

● AFC/Ticketing Data – Tracks entry/exit trends for station-level occupancy

● Cellular Network Data – Estimates macro-level crowd trends via telecom providers

● Bluetooth & IoT Sensors – Detects movement patterns (based on viability/requirements)

● Weight Sensors – Estimates train car occupancy (if available via rolling stock)

● RFID Tracking – Monitors real-time passenger location data

● CCTV Data – Enhances real-time station crowd assessment and congestion detection

● Historical Data & AI Models – Predicts trends and optimizes scheduling

● Existing APIs – Enhances accuracy through third-party data validation

These sources are processed using advanced Machine-Learning models to calibrate proxy headcounts against real-world conditions, ensuring accurate and reliable occupancy analytics.

Technologies Used

Intel OneAPI

Intel Geti

Intel OpenVINO