ARCHIVES

Efficient public transportation systems are essential for ensuring equitable mobility across urban and rural regions. However, conventional scheduling approaches often prioritize profitability, resulting in uneven service distribution and reduced accessibility in low-demand areas. This paper proposes a fairness-aware machine learning framework for intelligent transport scheduling using a hybrid dataset approach. Due to limited availability of granular transport datasets, real- world statistical data from Tamil Nadu is combined with synthetically generated route-level data to simulate realistic scenarios. The methodology integrates clustering techniques for region classification and regression models for demand prediction. To ensure equitable allocation, fairness metrics such as Equal Service Ratio (ESR) and Gini Index are incorporated into the decision-making process. Experimental results demonstrate that the proposed model achieves high predictive accuracy (approximately 95–96%) while maintaining balanced service distribution. The framework highlights a scalable and practical solution for sustainable and inclusive public transport planning.