CAB5: A Unified Corporate Cab Booking Management System with Offline AI Assistance

Abstract

This paper presents CAB5, a web-based platform that combines cab booking and package delivery services into a single unified system. The platform addresses critical frag- mentation problems in current urban mobility solutions, where users depend on multiple separate applications. CAB5 features a multi-role architecture with Customer, Driver, and Admin modules, enabling users to book rides, track packages, and manage logistics seamlessly. The system incorporates transpar- ent pricing, real-time vehicle tracking, and a novel offline AI assistant called POKO that functions without continuous internet connectivity. Built using modern technologies including Next.js, TypeScript, and Tailwind CSS, CAB5 is designed for scalability, maintainability, and high performance. The results demonstrate that CAB5 provides a reliable, cost-efficient, and scalable solution for smart urban mobility and logistics, effectively eliminating the need for multiple disjointed platforms.

Introduction

This work addresses the growing inefficiencies in urban transportation and logistics caused by the fragmentation of ride-hailing and delivery services across multiple platforms. With rapidly expanding global markets for both ride-hailing and last-mile delivery, users and drivers face operational challenges due to separate systems, leading to higher costs, reduced convenience, and poor coordination. To solve this, the paper proposes CAB5, a unified web-based platform that integrates cab booking, ride services, and package delivery into a single ecosystem.

CAB5 introduces a multi-role system consisting of customers, drivers, and administrators. Customers can book rides and deliveries with real-time tracking and transparent pricing, drivers can manage availability and earnings, and administrators can monitor operations through centralized dashboards. A key feature of the system is POKO, an offline AI assistant designed to provide user support even without internet connectivity, improving accessibility in low-network areas.

The system is supported by mathematical models that govern its core operations. Fare calculation is based on a linear formula considering distance, time, base fare, service type, and waiting charges. A dynamic pricing model adjusts fares during peak demand using supply-demand ratios. Driver assignment is optimized using a cost-based matching algorithm incorporating geographic distance (via the Haversine formula). ETA prediction is handled through a regression model, while queueing theory is used to analyze request processing efficiency. An additional confidence scoring mechanism supports offline AI responses.

From a technical perspective, CAB5 uses a modular architecture with a React/Next.js frontend, Tailwind CSS for styling, WebSockets for real-time communication, and LocalStorage for prototype data handling. The system workflow follows sequential states from request initiation to completion, ensuring structured ride and delivery management.

Conclusion

This paper presented CAB5, a unified corporate cab booking management system that successfully integrates ride-hailing and package delivery services into a single ecosystem. The system addresses critical fragmentation problems in current urban mobility solutions. Future work will focus on: 1) Integration with cloud-based databases (MongoDB, Fire- base). 2) Secure payment gateways (UPI, cards). 3) Machine learning-based demand prediction.

References

[1] GeeksforGeeks, ”Cab Booking System - Low-Level Design,” [Online]. Available: https://www.geeksforgeeks.org/system-design/cab-booking- system-low-level-design/ [2] V. D. Nalawade, ”Online Cab Booking System,” IJARSC, vol. 8, no. 3, pp. 45-52, 2021. [3] ”Cab Booking System Project Payment Method,” Scribd, Technical Report, 2022. [4] S. S. Waghare, D. H. Pethkar, and A. J. Kumavat, ”Online Cab Booking System,” IJSRCEIT, vol. 9, no. 7, pp. 234-241, 2023. [5] S. R. Mallreddy and V. Santhosh, ”Car Booking Systems Through Online,” JETIR, vol. 6, no. 2, pp. 112-118, Feb. 2019. [6] I. E. Mani, ”A Study on Online Cab Services,” JETIR, vol. 5, no. 9, pp. 78-84, Sep. 2018. [7] A. Aggarwal and S. Garg, ”Real-time Driver Matching Algorithm,” IEEE ITSC, 2020, pp. 156-162. [8] M. H. Rahman et al., ”Deep Learning Approach for ETA Prediction,” IEEE T-ITS, vol. 22, no. 8, pp. 4892-4901, 2021. [9] K. Zhang et al., ”Dynamic Pricing Strategies,” IEEE TNSE, vol. 9, no. 2, pp. 567-578, 2022. [10] L. Wang and X. Li, ”Queueing Models for Ride-Hailing,” IEEE T-IV, vol. 7, no. 3, pp. 612-623, 2022. [11] P. Kumar and R. Singh, ”Offline AI Assistants,” IEEE EDGE, 2021, pp. 89-96. [12] S. Banerjee et al., ”Last-mile Delivery Optimization,” IEEE Access, vol. 10, pp. 23456-23470, 2022. [13] N. Sharma and V. K. Jain, ”Comparative Analysis of Ride-hailing,” IEEE TCSS, vol. 8, no. 4, pp. 901-912, 2021. [14] R. Gupta et al., ”Secure Authentication Framework,” IEEE CNS, 2022, pp. 45-52. [15] J. Lee et al., ”Real-time Vehicle Tracking,” IEEE TVT, vol. 70, no. 11, pp. 11234-11245, 2021. [16] T. Das and P. Sen, ”Chatbot Implementation for Customer Support in Mobility Services,” IEEE TAI, vol. 3, no. 5, pp. 789-801, 2022. [17] M. A. Khan et al., ”Smart City Mobility Framework,” IEEE IoTJ, vol. 9, no. 15, pp. 13456-13470, 2022. [18] V. Patel et al., ”Performance Analysis of Web-based Cab Systems,” IEEE ICWS, 2023, pp. 212-219.

Copyright

Copyright © 2026 Aditya Mishra, Shubham Tiwari, Ashutosh Shrivastava, Anjali Awasthi. 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.