Machine learning-aided hybrid technique for dynamics of rail transit stations classification: a case study

Area of Science:

• Urban Planning
• Transportation Engineering
• Data Science

Background:

• Accurate rail transit station classification is vital for effective Transit-Oriented Development (TOD) and sustainable urban growth.
• Existing methods may lack the precision needed for sophisticated urban planning.
• Optimizing TOD strategies requires a refined understanding of station attributes.

Purpose of the Study:

• To develop and validate a novel classification model for rail transit stations.
• To enhance the precision of station classification using integrated methodologies.
• To support data-driven decision-making for urban planners and policymakers.

Main Methods:

• Integration of traditional methodologies with advanced machine learning algorithms.
• Application of mathematical models, clustering methods, and neural network techniques.
• Validation through a comprehensive case study on the Chengdu rail transit network.

Main Results:

• Regression models (MLR, DNN, KNN) achieved Mean Squared Error (MSE) below 0.012 for ridership forecasts.
• Neural networks achieved 100% accuracy for station classification across seven time intervals and 98.15% for the eighth.
• The model demonstrated high accuracy and reliability in station classification and ridership forecasting.

Conclusions:

• The novel classification model significantly enhances the precision of rail transit station evaluation.
• The model provides valuable insights for optimizing TOD strategies and guiding urban development.
• Accurate classification ensures reliable data-driven decisions for transit planning and sustainable urban growth.