Smart Public Transportation Sensing: Enhancing Perception and Data Management for Efficient and Safety Operations

Area of Science:

• Data Science
• Transportation Systems Engineering
• Computer Science

Background:

• Public transportation generates vast, complex data from cloud computing, big data, IoT, and mobile apps.
• Large data volumes and variety hinder efficient, real-time data sensing and processing in transit management.
• Existing systems face obstacles in managing diverse data sources for effective public transportation perception.

Purpose of the Study:

• To propose a public transportation sensing platform for collecting, integrating, and organizing diverse data.
• To overcome challenges in real-time data processing and ensure data availability for transit management.
• To enhance data safety and enable comprehensive analysis of public transportation business data.

Main Methods:

• Developed a data perception platform connecting multiple data systems and edge intelligent devices.
• Integrated an efficient field-level data lineage exploration method into the FlinkSQL system.
• Implemented a row-level fine-grained permission control mechanism for flexible data management.

Main Results:

• The platform efficiently collects and integrates diverse data, including passenger travel information and smart card transactions.
• The data lineage method enables precise extraction of traveling behavior during logical plan generation.
• The system supports efficient processing of large datasets, comprehensive analysis, and high data safety.

Conclusions:

• The proposed sensing platform effectively manages large volumes of diverse public transportation data.
• Operational testing confirms high efficiency in managing operations, data assets, and the data lifecycle.
• The system enables realization of data value through secure, comprehensive analysis and backend administration.