A Digital Twin-Driven Recommendation System for Adaptive Campus Course Timetabling

2025-03-08

AI Summary

Course timetabling in large, dynamic campuses suffers from insufficient adaptability due to tightly coupled hard/soft constraints, real-time disruptions, and evolving user preferences. Method: This paper pioneers a paradigm shift—reformulating timetabling as a personalized recommendation task. Leveraging the Texas A&M University digital twin platform, we integrate collaborative filtering with spatiotemporal-aware content-based recommendation. A multi-objective weighted scoring function explicitly models spatial–temporal constraints, and an iterative user feedback mechanism enables continuous optimization. Contribution/Results: Our approach transcends traditional static optimization, delivering real-time responsiveness and self-adaptive evolution. Empirical evaluation on real-world data demonstrates a 12.7% increase in classroom utilization, a significant reduction in inefficient faculty–student commuting, and a 23.4% improvement in user satisfaction.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Efficient and adaptive course timetabling for large, dynamic university campuses remains a significant challenge due to the complex interplay of hard and soft constraints. Traditional static optimization methods often fail to accommodate real-time disruptions, evolving user preferences, and the nuanced spatial-temporal relationships inherent in campus environments. This paper reconceptualizes the timetabling problem as a recommendation-based task and leverages the Texas A&M Campus Digital Twin as a dynamic data platform. Our proposed framework integrates collaborative and content-based filtering techniques with iterative feedback mechanisms, thereby generating a ranked set of adaptive timetable recommendations. A composite scoring function, incorporating metrics for classroom occupancy, travel distance, travel time, and vertical transitions, enables the framework to systematically balance resource utilization with user-centric factors. Extensive experiments using real-world data from Texas A&M University demonstrate that our approach effectively reduces travel inefficiencies, optimizes classroom utilization, and enhances overall user satisfaction. By coupling a recommendation-oriented paradigm with a digital twin environment, this study offers a robust and scalable blueprint for intelligent campus planning and resource allocation, with potential applications in broader urban contexts.

Problem

• Addresses adaptive course timetabling challenges in dynamic university campuses.
• Proposes a recommendation system leveraging digital twin technology for real-time adaptability.
• Optimizes resource utilization and user satisfaction through advanced filtering techniques.

Innovation

• Digital Twin for dynamic campus data platform
• Collaborative and content-based filtering techniques
• Composite scoring function for resource optimization