Abstract
The Indian Railway Catering and Tourism Corporation (IRCTC) operates one of the most heavily utilized railway reservation systems globally, reflecting the central role of Indian Railways (IR) as an affordable and essential mode of transportation across the country. However, selecting an appropriate train remains a complex decision-making task for passengers, primarily due to the uncertainty surrounding ticket availability on preferred travel dates. To address this challenge, the present study proposes a hybrid decision support system designed to aid passengers in selecting optimal train options, particularly for tourism-related travel. This research employs a Dominance-Based Rough Set Approach (DRSA) within a Multi-Criteria Decision-Making (MCDM) framework to analyze preference-based data and extract interpretable decision rules in the form of “if...then” statements. These rules assist decision makers in evaluating multiple train-related criteria simultaneously. For comparative purposes, the Classical Rough Set Approach (CRSA) is also implemented to identify the relative advantages and limitations of both rough set methodologies in addressing train selection complexity. In addition, the study integrates machine learning techniques by utilizing two predictive models – Extreme Gradient Boosting (XGBoost) and Support Vector Machine Classifier (SVMC) – to estimate overall train ratings based on user preferences and historical data. Model performance is evaluated using standard classification metrics, including accuracy and precision. By combining MCDM techniques with machine learning algorithms, the proposed hybrid framework enhances the train reservation experience, enabling passengers to make informed, preference-aligned travel decisions through the Indian Railways reservation system.