AI-Based College Transportation Management

Abstract

Colleges that run their own buses usually plan transport using fixed routes and estimated student counts. In practice, the number of students who actually travel changes from day to day, so some buses run with many empty seats while others are overcrowded. This paper presents the design of the AI-Based College Transportation Decision Support System, a full-stack web application that collects daily travel confirmations from students and uses rule-based software agents to analyse route-wise demand, check route compatibility, recommend bus allocation, and suggest seat optimization within capacity limits. Every recommendation is accompanied by a plain-language explanation, and no change becomes final until the administrator approves it. The previous allocation stays in force until a replacement plan is approved. The system is built with React.js, Node.js, Express.js and MongoDB, with JWT-based authentication and role-based access control. It uses predefined rules and calculations only; it does not use machine learning, GPS tracking, or external generative AI services. This paper describes the problem, the system architecture, the database design, the rule-based algorithms, and an illustrative worked example. Experimental evaluation on real college data is identified as future work.

Introduction

The AI-Based College Transportation Decision Support System is designed to improve the efficiency of college bus services by adjusting bus allocation according to students’ daily travel requirements. Traditional transportation planning relies on fixed routes and estimated student numbers, which can lead to overcrowded buses on some routes and underutilized buses on others.

The proposed system allows students to confirm each evening whether they will use the college bus the following day. It analyzes these confirmations, calculates route-wise demand, checks route compatibility and bus availability, and recommends suitable bus allocations while ensuring that seating capacity is never exceeded.

The system uses a rule-based AI approach, implemented through JavaScript calculations and predefined rules, rather than a trained machine learning model. It generates understandable explanations for its recommendations, while the administrator retains control by approving, modifying, or rejecting the proposed changes. Students are notified once the final arrangements are confirmed.

The main objectives are to collect daily travel confirmations, identify high- and low-demand routes, optimize seat utilization, verify route compatibility, maintain safe bus capacities, and support transparent transportation decisions.

The system is developed using React.js for the frontend, Node.js and Express.js for the backend, and MongoDB for database management. REST APIs enable communication between the frontend and backend, while JSON Web Token (JWT) authentication helps secure access according to user roles.

The application provides two main portals:

• Student portal: Enables registration, login, route viewing, daily travel confirmation, and access to finalized bus allocations and notifications.

• Administrator portal: Supports the management of students, buses, routes, and capacities, along with demand analysis, allocation recommendations, approval decisions, and student notifications.

The database contains collections for users, buses, routes, travel confirmations, allocations, recommendations, approvals, and notifications. Security and reliability are supported through password hashing, duplicate-confirmation prevention, valid student assignments, and mandatory administrator approval.

The system follows essential business rules: buses must not exceed their seating capacity, only valid travel confirmations are considered, routes must be compatible before redistribution, and previous allocations remain active until new arrangements are approved. If no safe optimization is possible, the system makes no changes.

The literature review identifies a research gap in affordable transportation planning solutions for small institutions with fixed routes and varying daily demand. Unlike systems that depend on GPS tracking, live traffic data, or complex optimization algorithms, this project focuses on a low-cost, transparent, and practical web-based solution.

Conclusion

This paper presented a rule-based decision-support system that helps college administrators match bus allocation to actual daily demand. By combining daily confirmations, route compatibility rules, capacity-respecting seat optimization, explainable recommendations, and mandatory administrator approval, the design aims to reduce empty seats while protecting reliability and safety. Possible future enhancements, which are not part of the current system, include SMS or email notifications, reminders for students who have not confirmed, analytics dashboards on historical occupancy, and a pilot study in a real college to measure the effect on seat utilization.

References

[1] Puneeth Y M and Mithun Dolthody Jayaprakash, “Transportation route optimization,” Int. J. Res. Appl. Sci. Eng. Technol. (IJRASET), vol. 13, no. VI, Jun. 2025, doi: 10.22214/ijraset.2025.72760. [2] G. B. Dantzig and J. H. Ramser, “The truck dispatching problem,” Manage. Sci., vol. 6, no. 1, pp. 80–91, 1959. [3] P. Toth and D. Vigo, Eds., Vehicle Routing: Problems, Methods, and Applications, 2nd ed. Philadelphia, PA, USA: SIAM, 2014. [4] J. Park and B.-I. Kim, “The school bus routing problem: A review,” Eur. J. Oper. Res., vol. 202, no. 2, pp. 311–319, 2010. [5] M. Jones, J. Bradley, and N. Sakimura, “JSON Web Token (JWT),” Internet Eng. Task Force, RFC 7519, May 2015. [6] Y. Cai and S. Sun, “Optimizing customized bus routing and maximum seat occupancy rate under the influence of epidemic outbreaks,” IEEE Access, vol. 12, 2024, doi: 10.1109/ACCESS.2024.3481427. [7] S. Parthasarathy, R. Thangasankaran, Gopinath S. B., Harshath K. S., Dinesh Kumar V. V., Jayabalajivel R., and Ranjith Kumar K. B., “TRANSPECTRA: An AI-powered centralised transport management system,” Int. J. Comput. Organ. Trends (IJCOT), vol. 16, no. 1, pp. 8–17, 2026. [8] Y. Harrath, J. AlYusuf, Z. Ghulam, and M. AlDoseri, “Optimal algorithm for managing on-campus student transportation,” in Proc. Int. Conf. Decis. Aid Sci. Appl. (DASA), IEEE, 2024, pp. 1–8, doi: 10.1109/DASA63652.2024.10836580. [9] W. Shu and Y. Li, “A novel demand-responsive customized bus based on improved ant colony optimization and clustering algorithms,” IEEE Trans. Intell. Transp. Syst., vol. 24, no. 8, pp. 8492–8506, Aug. 2023, doi: 10.1109/TITS.2022.3145655. [10] L. Qian and E. Melachrinoudis, “An integrated neural combinatorial tabu search for optimizing school bus scheduling with bell time,” Transp. Res. C, Emerg. Technol., vol. 164, Art. no. 104662, 2024, doi: 10.1016/j.trc.2024.104662. [11] W. A. Ellegood, S. Solomon, J. North, and J. F. Campbell, “School bus routing problem: Contemporary trends and research directions,” Omega, vol. 95, Art. no. 102056, 2020, doi: 10.1016/j.omega.2019.03.014. [12] M. M. Solomon, “Algorithms for the vehicle routing and scheduling problems with time window constraints,” Oper. Res., vol. 35, no. 2, pp. 254–265, 1987, doi: 10.1287/opre.35.2.254. [13] S. Røpke and D. Pisinger, “An adaptive large neighborhood search heuristic for the pickup and delivery problem with time windows,” Transp. Sci., vol. 40, no. 4, pp. 455–472, 2006, doi: 10.1287/trsc.1050.0135. [14] D. Huang, Y. Gu, S. Wang, Z. Liu, and W. Zhang, “A two-phase optimization model for the demand-responsive customized bus network design,” Transp. Res. C, Emerg. Technol., vol. 111, pp. 1–21, 2020, doi: 10.1016/j.trc.2019.12.004. [15] P. Vansteenwegen, L. Melis, D. Akta?, B. D. Galarza Montenegro, F. S. Vieira, and K. Sörensen, “A survey on demand-responsive public bus systems,” Transp. Res. C, Emerg. Technol., vol. 137, Art. no. 103573, 2022, doi: 10.1016/j.trc.2022.103573. [16] Q. Sun, S. Chien, D. Hu, G. Chen, and R. S. Jiang, “Optimizing multi-terminal customized bus service with mixed fleet,” IEEE Access, vol. 8, pp. 156456–156469, 2020, doi: 10.1109/ACCESS.2020.3018883. [17] X. Dou, Q. Meng, and K. Liu, “Customized bus service design for uncertain commuting travel demand,” Transportmetrica A, Transp. Sci., vol. 17, no. 4, pp. 1405–1430, 2021, doi: 10.1080/23249935.2020.1864509. [18] X. Liu, Y.-L. Chen, L. Y. Por, and C. S. Ku, “A systematic literature review of vehicle routing problems with time windows,” Sustainability, vol. 15, no. 15, Art. no. 12004, 2023, doi: 10.3390/su151512004.

Copyright

Copyright © 2026 S. Yoheswari, Mahadeer M, Kishore K R. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.