Abstract

On?demand service marketplaces have become an integral part of modern urban life by enabling users to access professional services in a convenient and time?efficient manner. However, existing platforms often suffer from inefficient provider selection, limited transparency during service execution, delayed responses, and insufficient real?time coordination. These limitations reduce user satisfaction and erode trust in service platforms. This paper presents AI?Based QuickServe, a comprehensive intelligent service marketplace that integrates artificial intelligence, geolocation services, and real?time communication technologies to address these challenges. The proposed system includes AI?assisted request understanding using image analysis, a multi?criteria provider matching algorithm combining distance, availability, reliability, and timeliness, and real?time tracking using WebSockets within a secure, scalable architecture. Extensive functional evaluation with realistic workloads demonstrates improved matching quality, reduced response latency, and enhanced transparency while maintaining stable performance under concurrent usage. The system provides a robust foundation for next?generation urban service ecosystems.

Introduction

Digital platforms have transformed how users access home services such as plumbing, electrical repair, AC maintenance, cleaning, and appliance servicing. While on-demand marketplaces promise speed and convenience, many suffer from recurring issues:

• Delayed responses and last-minute cancellations

• Poor provider matching (distance-only logic)

• Limited transparency after assignment

• Weak verification and trust mechanisms

• Operational inefficiencies at scale

To address these challenges, the paper proposes AI-Based QuickServe, an intelligent, real-time service marketplace that automates the entire service lifecycle using AI, geolocation, multi-criteria matching, and secure role-based architecture.

1. Core Motivation

Existing platforms often:

• Treat requests as static text entries

• Ignore urgency, complexity, workload, and provider reliability

• Provide little real-time tracking

• Offer limited admin oversight

This leads to mismatches, dissatisfaction, and scalability issues.

QuickServe aims to create a smart, adaptive, and transparent ecosystem for customers, providers, and administrators.

2. System Overview

QuickServe is a full-stack platform with three user roles:

1) Customer

• Create service requests

• Upload images for AI analysis

• View ranked nearby providers

• Track providers live on maps

• Rate completed services

2) Provider

• Register and complete KYC verification

• Define service categories and pricing

• Go online/offline

• Accept/reject jobs

• Track earnings and navigation

3) Admin

• Approve/reject KYC documents

• Monitor requests and providers

• Adjust matching parameters

• Access analytics and system controls

The platform automates everything from request creation to feedback collection.

3. Architecture

QuickServe uses a modern layered architecture:

Frontend

• React 19 + Vite

• Tailwind CSS

• React Router v6

Backend

• FastAPI (Python)

• SQLAlchemy + PostgreSQL

Real-Time Layer

• WebSockets for live tracking and notifications

Cache

• Redis for session tokens and active provider data

External Services

• Groq Vision API (image analysis)

• Cloudinary (image/document storage)

• OpenCage (geocoding)

• MapmyIndia / Leaflet (maps and routes)

The system uses REST APIs for standard communication and WebSockets for live updates.

4. Intelligent Matching Algorithm

The core innovation is a multi-criteria provider matching algorithm designed to:

• Select verified and available providers

• Minimize travel distance

• Prioritize reliability and timeliness

• Remain scalable under high load

Matching Phases

1. Filtering

   • Match service category

   • Ensure KYC approval

   • Check availability

2. Distance Calculation

   • Use Haversine formula

   • Keep providers within configurable radius

3. Feature Computation

   • Distance score

   • Reliability score (ratings, completion rate)

   • Availability score

   • Timeliness score

4. Weighted Scoring

   Sp=wdDp+wrRp+waAp+wtTpS_p = w_dD_p + w_rR_p + w_aA_p + w_tT_pSp?=wd?Dp?+wr?Rp?+wa?Ap?+wt?Tp?

5. Ranking & Selection

   • Highest score assigned

   • Top-k backups retained

Complexity

Time complexity per request:
O(N + M log M)
Efficient due to early filtering and geospatial indexing.

5. Security and Reliability

Security is built into the system:

• JWT-based authentication (short-lived tokens)

• Role-Based Access Control (RBAC)

• HTTPS communication

• Secure password hashing (bcrypt)

• KYC document verification

• Sensitive data stored in environment variables

WebSockets enable:

• Live location updates

• Status transitions

• Event-driven notifications

6. Evaluation Results

Testing showed:

Matching Quality

• Algorithm prioritizes close and reliable providers

• Adjustable weights allow emphasis on speed or quality

• Fast assignment time

Transparency

• Real-time tracking improves trust

• Clear status updates reduce uncertainty

Scalability

• Stable response times under concurrent load

• Redis caching improves frequent queries

• WebSockets scale effectively

7. Applications

QuickServe can extend to:

• Home services (primary focus)

• Emergency maintenance

• Location-based freelancing

• Smart city coordination platforms

8. Limitations

• Dependence on third-party APIs

• GPS accuracy constraints

• Manual tuning of matching weights

• No automated learning from historical data yet

9. Future Scope

Potential improvements include:

• AI-driven personalized provider recommendations

• Predictive demand forecasting

• Dynamic pricing

• Voice-based request interface

• Multilingual support

• Automatic learning of matching weights

Conclusion

This paper presented AI?Based QuickServe, a comprehensive intelligent service marketplace for real?time urban service management. By combining AI?assisted request understanding, a multi?criteria matching algorithm, geolocation, and real?time WebSocket communication within a secure, scalable architecture, the system addresses key limitations of existing platforms, including inefficient provider selection, limited transparency, and poor coordination. Functional evaluation shows that QuickServe improves matching quality, reduces response latency, and enhances user trust. The modular architecture and configurable matching logic make it suitable as a foundation for future, city?scale intelligent service ecosystems.

Copyright

Copyright © 2026 Anushka Shimpi, Pratik Sawant, Tushar Shinde, Omkar Sola. 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.