Human Mobility Reimagined: Digital Twin Intelligence for Adaptive Campus Course Timetabling

Keshu Wu, Xinyue Ye, Suphanut Jamonnak, Xin Feng · 2025

Read Human Mobility Reimagined: Digital Twin Intelligence for Adaptive Campus Course Timetabling by Keshu Wu, Xinyue Ye, Suphanut Jamonnak, Xin Feng (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-06109.txt.

Introduction

Keshu Wu, Xinyue Ye, Suphanut Jamonnak, and Xin Feng wrote this paper in two thousand twenty-five. It recasts campus course timetabling as a mobility problem, and solves it with a digital twin of Texas A and M University.

A timetable is a movement policy: it sends thousands of students across campus in set waves. Hard rules on rooms and clashes must hold while soft aims on taste and fit are weighed. The frame joins two kinds of recommendation filters with a feedback loop that refines the picks. A score blends room use, travel span, travel time, and floor shifts. Real campus data show less travel friction and better room use than past plans.

Catalog record

CREATORKeshu Wu, Xinyue Ye, Suphanut Jamonnak, Xin Feng
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE59 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-06109

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Human Mobility Reimagined: Digital Twin Intelligence for Adaptive Campus Course Timetabling Keshu Wu, Xinyue Ye, Suphanut Jamonnak, Xin Feng Abstract Daily operations in large campuses depend on how efficiently people move through space and time. In this sense, course timetables are more than administrative schedules: they act as mobility policies that orchestrate thousands of trajectories, shaping travel burden, congestion, accessibility, and the reliability of back-to-back transitions. Designing timetables that are both feasible and mobility-friendly is challenging because hard constraints including capacity, conflicts, feasibility must be satisfied alongside soft constraints including preferences, satisfaction, coordination, all under dynamic conditions such as real-time disruptions and evolving demand. Traditional static optimization methods often struggle to capture these human mobility impacts and to adapt when campus conditions change. This paper reconceptualizes course timetabling as a recommendation-based task and leverages the Texas A&M Campus Digital Twin as a dynamic data platform to evaluate mobility consequences at scale. We propose an iterative framework that integrates collaborative and content-based filtering with feedback-driven refinement to generate ranked sets of adaptive timetable recommendations. A mobility-aware composite scoring function combining classroom occupancy, travel distance, travel time, and vertical transitions systematically balances resource efficiency with human-centered movement costs. Extensive experiments using real-world data from Texas A&M University show that the proposed approach reduces mobility friction and travel inefficiencies, improves classroom utilization, and enhances overall user satisfaction. By coupling recommendation-oriented decision-making with digital twin intelligence, this study provides a robust and scalable blueprint for mobility-centered campus planning and resource allocation, with potential extensions to broader urban systems. 1 Introduction Human mobility is a foundational layer of how complex systems function: it determines how people access services, how shared resources are utilized, and how daily activities unfold in space and time [ 1 , 2 , 3 , 4 ] . In increasingly data-rich environments, the central challenge is no longer merely to plan mobility under fixed assumptions, but to adapt mobility-centric decisions as conditions, preferences, and constraints evolve [ 5 , 6 ] . Digital twins enable this shift by coupling a physical system with a continuously updated virtual counterpart, making it possible to evaluate decisions not only by feasibility and efficiency, but also by their mobility consequences [ 7 , 8 , 9 , 10 ] . This paper adopts that perspective and studies adaptive resource allocation through a human-mobility-first lens, using a university campus as a living microcosm where dense daily movements, tight time budgets, and heterogeneous indoor-outdoor infrastructure make mobility impacts immediately tangible. Within this setting, a timetable is not simply an administrative artifact; it acts as a mobility policy that orchestrates thousands of day-to-day trajectories. Each assignment of an activity to a time and place implicitly shapes transition pressures, corridor congestion, and the uneven burdens of long walks, delayed arrivals, and vertical movements. Course timetabling in large academic institutions therefore represents a multifaceted challenge that requires the precise allocation of courses, classrooms, and time slots while meeting rigorous institutional policies and addressing the diverse needs of students and faculty [ 11 , 12 , 13 ] . The problem is compounded by the necessity to balance hard constraints —such as classroom capacities, scheduling conflicts, and feasible travel times—with soft constraints including faculty preferences, student satisfaction, and inter-departmental coordination [ 14 , 15 ] . Crucially, spatial-temporal factors such as building locations, pedestrian pathways, indoor circulation (e.g., corridors, stairs, elevators), and transition times introduce nontrivial dependencies between consecutive classes. These dependencies vary across individuals, time-of-day conditions, and infrastructure availability, making mobility inherently dynamic. Along with fluctuating enrollment, changing classroom availability, construction, and disruptions, these mobility considerations render traditional static optimization approaches insufficient [ 16 , 17 , 18 , 19 ] . Consequently, there is an urgent need for adaptive methodologies that explicitly integrate spatial, temporal, and operational considerations to reduce mobility friction while maintaining institutional feasibility. Extensive research has addressed the complexities of course timetabling using a variety of optimization techniques. Foundational approaches—such as constraint satisfaction models [ 20 ] and combinatorial optimization frameworks [ 21 ] —have laid the groundwork for automated timetabling. In addition, evolutionary algorithms, including genetic algorithms [ 22 ] , harmony search [ 23 ] , and particle swarm optimization [ 24 ] , have demonstrated the ability to navigate large solution spaces and balance competing objectives effectively. Mixed-integer programming (MIP) techniques provide exact solutions for smaller problems [ 25 , 26 , 27 ] , while multi-objective optimization methods adeptly manage trade-offs between conflicting goals, such as minimizing travel distances while maximizing classroom utilization [ 28 ] . Moreover, hyper-heuristics have emerged as a flexible means to address diverse scheduling scenarios by balancing computational efficiency and solution quality [ 21 ] . Despite these advances, many timetabling formulations still treat mobility as a coarse add-on (e.g., a single distance/time penalty) rather than modeling the full movement experience that includes inter-building flows, indoor-outdoor transitions, vertical movements, and time-varying congestion. Recent advancements have further enhanced traditional methods by integrating machine learning and hybridization strategies. For instance, fuzzy logic has enabled the dynamic incorporation of user preferences into the scheduling process [ 29 ] , and machine learning techniques have improved algorithm selection and design [ 30 , 31 ] . Clustering algorithms have been employed to simplify complex schedules [ 32 ] , and reinforcement learning has demonstrated promise in refining schedules through iterative interactions with dynamic environments [ 33 , 34 ] . Nevertheless, significant gaps remain from a human mobility perspective. Many existing approaches do not fully incorporate spatial-temporal correlations in real campus networks or account for multi-modal transportation options (walking, bicycling, e-scooters, shuttles) and indoor mobility modes (corridors, stairs, elevators)—factors that directly determine travel feasibility, comfort, and schedule reliability in large academic settings. Additionally, most models lack the capability to generate adaptive schedules that respond in real time to disruptions and evolving requirements, even though mobility conditions and infrastructure availability can change within hours. Digital twin technology has recently emerged as a transformative tool for addressing these challenges [ 35 , 36 ] . A digital twin is a dynamic, real-time virtual representation of a physical system that seamlessly integrates spatial-temporal data with operational constraints, offering a powerful platform for simulation, monitoring, and optimization. While digital twins have been widely adopted in manufacturing to enhance production scheduling and operational efficiency [ 37 , 38 , 39 ] , their application in urban systems and academic environments is still emerging. For campus operations, a digital twin provides the missing connective tissue between scheduling decisions and human mobility realities: it can encode building layouts, pathway networks, vertical circulation, and evolving conditions (e.g., closures or route changes), enabling timetables to be evaluated and refined based on realistic movement costs and constraints. Recent work in academic scheduling demonstrates the potential of digital twins to adapt dynamically to changes such as fluctuating enrollments, evolving classroom availability, and infrastructure disruptions [ 35 ] . By merging diverse datasets—including classroom attributes, transportation networks, and user feedback—digital twin technology enables a holistic approach to optimizing campus operations. Figure 1: Texas A&M Digital Twin. In this study, we leverage the Texas A&M Campus Digital Twin [ 36 , 40 ] , as shown in Figure 1 , to bridge the identified

Read the complete text