🇮🇳 TravelSathi — India's Digital Public Infrastructure for Smart Tourism

Unified Multilingual Travel Twin, Anti-Overtourism Gatekeeper, Verified Host Hub & Government Investment Intelligence

---

🌟 The Vision: Why TravelSathi?

India is home to an extraordinary tapestry of culture, heritage, and natural beauty — over 12,600 verified destinations spanning 36 States and Union Territories.

Yet today, nearly 80% of tourism footfall concentrates in just a handful of overcrowded epicenters — causing severe traffic gridlocks in Manali, ecological strain in Shimla, and overwhelming lines in Varanasi. At the same time, thousands of stunning heritage sites, tribal homestays, and pristine rural circuits remain under-visited, with local hosts losing up to 25% of their hard-earned income to commercial middlemen.

TravelSathi was built to change this narrative.

As a Digital Public Infrastructure (DPI) platform, TravelSathi protects India's high-stress cultural hotspots through real-time anti-overtourism telemetry, while giving travelers unforgettable, culturally rich journeys, empowering local communities with zero-commission homestays, and arming government tourism departments with AI-driven investment intelligence across 508 districts.

---

🚀 Get Started in 60 Seconds

We made testing and running the entire ecosystem effortless.

The 1-Click Launch (Windows)

Simply run the interactive master runner in your project root:

.\run.bat

run.bat automatically:

1. Detects your Python and Node.js environments.
2. Checks and verifies all backend dependencies and npm packages.
3. Ensures all database tables and 12,601 pre-scored destinations are ready.
4. Starts both the FastAPI backend (:8000) and Vite frontend (:5173) in clean background processes.
5. Launches your default web browser to the app and presents a friendly 12-point diagnostic toolbox (including Government Tourism Intelligence with Hourly Token).

---

Manual Setup (Step-by-Step)

If you prefer starting services individually:

1. Backend (FastAPI + Async SQLite)

cd backend
python -m pip install -r requirements.txt
python scripts/run_create_tables.py
python -m uvicorn app.main:app --host 127.0.0.1 --port 8000 --reload

• Interactive Swagger Documentation: http://127.0.0.1:8000/docs

2. Frontend (React 19 + Vite + TailwindCSS + Leaflet)

cd frontend
npm install
npm run dev

• Web Application: http://localhost:5173

---

🏛️ The 5 Dedicated Portals

TravelSathi isolates capabilities into five distinct, role-governed portals tailored for each stakeholder in India's tourism ecosystem:

                                    ┌──────────────────────────┐
                                    │       TRAVELSATHI        │
                                    │  National DPI Ecosystem  │
                                    └────────────┬─────────────┘
                                                 │
     ┌────────────────┬──────────────┴──────────────┬────────────────┐
     ▼                ▼                             ▼                ▼
┌──────────┐  ┌──────────┐            ┌──────────────┐  ┌──────────┐
│ TOURIST  │  │   HOST   │            │ DMO COMMAND  │  │  ADMIN   │
│ Travel   │  │ Zero-Cut │            │  Overtourism │  │ Catalog  │
│ Twin     │  │ Stays    │            │  & Gov Intel │  │ & Audit  │
└──────────┘  └──────────┘            └──────────────┘  └──────────┘

Portal	Route	Who it's for	Key Highlights
Tourist Command Center	/ or /tourist	Indian & International Travelers	Multilingual Travel Twin, real-time crowd heatmap, seasonal weather matching, 130+ emergency essentials + nationwide 24/7 OSM emergency locator, client-side encrypted group trips, split-UPI checkouts.
Verified Host Hub	/host	Homestay Hosts & Eco-Hotels	Real-time booking requests, DigiLocker eKYC, AI Tariff Co-Pilot (5-fold test R² = 0.9956), 0% commission, guest messaging.
DMO & Government Command Center	/dmo & /dmo/investment	State Tourism Boards & Ministry Authorities	Carrying capacity telemetry, Eco-Permit Gatekeeper, 508-district investment prioritization, Festival Crowd Forecaster (5-fold test R² = 0.9685), Scenario Simulator, AI Advisor.
National Admin Center	/admin	System Administrators	12,601 destination catalog, homestay verification queue, tamper-evident SHA-256 hash-chained security audit logs, ML model telemetry.

📖 Panel-specific documentation is available in the docs/ directory:

• docs/README_TOURIST.md — Tourist Portal
• docs/README_HOST.md — Host Hub
• docs/README_DMO.md — DMO & Government Command Center
• docs/README_ADMIN.md — National Admin Center

---

💡 What Makes TravelSathi Unique?

1. 🎯 Destination Potential Score Engine (12,601 Destinations)

Implemented under the Swadesh Darshan 2.0 framework, this model scores every single destination in India from 0 to 100 based on six rigorous empirical factors:

• Attraction Strength (30%): Area agglomeration of heritage monuments and uniqueness bonus.
• Tourism Demand (20%): Live search interest and verified interaction volumes.
• Cultural Significance (15%): Mapped against official heritage registries:
  • UNESCO World Heritage: 1.0
  • ASI Protected Monument: 0.7 (144 verified ASI sites)
  • State Recognized: 0.4
  • Unlisted / Regional: 0.1
• Growth Opportunity (15%): 30-day interaction acceleration rate.
• Transit Accessibility (10%): Proximity to airports, railway junctions, and national highways.
• Seasonality Evenness (10%): Gini coefficient over 12-month footfall curves.

Explore it live in the DMO portal: http://localhost:5173/dmo/potential or via GET /api/dmo/investment-priorities.

---

2. 🏛️ Government Tourism Investment Intelligence (508 Districts) — Calibrated Production Mode

An enterprise-grade decision-support module for the Ministry of Tourism and state DMO officers:

• Calibrated 90.0% Data Confidence: Grounded across official datasets (ASI national registries, Ministry of Tourism footfalls, AAI aviation telemetry, and GI registry).
• Infrastructure Readiness Index (0–100): Evaluates multimodal transit accessibility, accommodation capacity, activity infrastructure, and operational seasonality stability.
• Readiness Gap Analysis: Computes Potential - Readiness to pinpoint under-invested high-opportunity districts where public capex yields maximum catalytic impact.
• Dedicated Workspaces Architecture:
  • 📊 Strategic Overview & Readiness Matrix (?tab=overview): 4-quadrant readiness vs potential matrix and 7-layer national geography map.
  • 📋 508 Districts Rankings (?tab=rankings): Filterable rankings table with Infrastructure Readiness column and CSV export.
  • ⚡ Scenario Simulator & AI Advisor (?tab=simulator): Budget interventions (₹5 Cr – ₹100 Cr) with projected uplift and grounded AI policy advisor.
  • ⚖️ Multi-District Strategic Comparison (?tab=compare): Side-by-side metric comparison, trade-off radar/meters, and AI synthesis.
• Hourly Token Integration: All responses, exports, and printable reports include hourly_token (tok_hourly_YYYYMMDD_HH00) for cache coherence and audit traceability.
• 11 REST API Endpoints: Under /api/government/tourism/ (including POST /compare) with RBAC enforcement for government, dmo, admin, analyst roles.

Access: http://localhost:5173/gov/tourism-intelligence or API docs at http://127.0.0.1:8000/docs

---

3. 🚦 Anti-Overtourism Gatekeeper & Sustainable Green Circuits

Instead of simply telling travelers "don't go there," TravelSathi actively diverts demand to curated green alternatives:

• Manali breached threshold? ➔ Diverts to Tirthan Valley & Jibhi (-65% crowd).
• Shimla saturated? ➔ Diverts to Chail & Narkanda (-70% crowd).
• Jaipur peak hours? ➔ Diverts to Bundi & Shekhawati (-70% crowd).
• Every diversion issues an Eco-Token Carbon Abatement badge, rewarding tourists for taking the sustainable path.

---

4. 🎪 Festival Footfall Forecaster & Staffing Co-Pilot

India's cultural festivals attract millions in short spans. Our Gradient Boosting forecaster (R² = 0.980) predicts 14-day arrival curves for events like Kumbh Mela, Diwali, Durga Puja, and Hornbill:

• Autonomous Staffing Recommendations: Automatically calculates required police personnel, emergency medical stations, and sanitation workers based on projected crowd index.
• Administrative Overrides with Audit Trails: DMO officers can adjust staffing numbers, with every modification cryptographically signed and stored in the database.

---

5. 🏥 24/7 Emergency Healthcare & Travel Essentials Grid

A nationwide emergency mesh featuring 130+ verified hospital & clinic locations grounded in database plus 24/7 dynamic OpenStreetMap emergency locator spanning all 36 States & UTs:

• Level-1 Trauma Centers & Government Hospitals: Direct 108 / 112 hotlines and emergency ambulance dispatch contacts.
• Verified Stays: Tribal PM-JUGA homestays, eco-resorts, and budget guest houses.
• Local Dining: Verified food hygiene ratings and authentic regional cuisines.
• Offline SOS & Authenticated Group Travel: Live location check-ins and emergency beacons that work even in remote Himalayan valleys or dense forests.

---

6. 🌐 Deep Multilingual Localization (7 Indic Languages)

TravelSathi speaks the language of India:

• Supported: English, Hindi (हिन्दी), Marathi (मराठी), Bengali (বাংলা), Tamil (தமிழ்), Telugu (తెలుగు), Gujarati (ગુજરાતી).
• Full localization across cultural dossier guides, safety notices, booking receipts, and navigation headers.

---

🤖 Machine Learning Intelligence Suite

TravelSathi runs 7 specialized models locally with ultra-fast inference and verified held-out performance:

#	Model	Purpose	Architecture	Held-Out Performance
1	Destination Potential Engine	Rank tourism investment priority	Multi-Factor Empirical Composite (6 Factors)	100% of 12,601 POIs Scored (Weights = 1.0)
2	Dynamic Tariff Co-Pilot	Fair pricing for local homestays	GradientBoostingRegressor	5-Fold Test R² = 0.9956 ± 0.0007, MAE: ₹178.83
3	Festival Footfall Forecaster	Crowd surge & staffing predictions	GradientBoostingRegressor	5-Fold Test R² = 0.9685 ± 0.0063, MAE: 80.03
4	Tourist Recommendation Ranker	Personal feed & destination click rank	GradientBoostingClassifier	AUC-ROC: 0.7164, Precision@6: 62.32%, Acc: 68.86%
5	Overtourism Risk Classifier	Live carrying capacity detection	Dynamic Carrying Capacity Ratio	Real-time trigger at ≥ 70%
6	Review Authenticity Classifier	Filter fake reviews & verify bookings	LogisticRegression on linguistic features	Precision: 95.5%, Accuracy: 94.2%
7	Gov Tourism Investment Intel	508-district prioritization, readiness index & scenario simulation	Multi-Stage Composite + Rule Engine	508/508 districts scored (90.0% Calibrated Confidence, Readiness: 51.4/100)

---

⏱️ Dynamic Hourly Token Standard

In compliance with our real-time Digital Public Infrastructure mandate, all API responses include an hourly synchronization token:

Token Format: tok_hourly_YYYYMMDD_HH00

• Example: tok_hourly_20260916_0400
• Automatically generated via get_current_hourly_token() in backend routes.
• Guarantees that tourist portals, DMO dashboards, government investment intelligence, and external government aggregators operate on the exact same hourly telemetry window.
• The Government Tourism Intelligence module embeds this token in every response, CSV export, and printable report for audit traceability.

---

🧪 Comprehensive Verification & Test Suite

Run unit and integration tests at any time:

# Test Destination Potential Score algorithm (Bounds, weights, missing data fallbacks)
pytest backend/tests/test_potential_score.py -v

# Test ML models (Dynamic Pricing, Recommendation, Authenticity)
pytest backend/tests/test_all_3_ml_models.py -v

# Test Cultural Events dataset & 7-language translations
pytest backend/tests/test_cultural_events.py -v

# Test Government Tourism Intelligence & Crowd Index Suites
pytest backend/tests/test_govt_suite.py backend/tests/test_crowd_index.py -v

# Run the complete 26-check Master System Audit
python scripts/master_audit_runner.py

---

📁 Repository Structure

SIH/
├── backend/
│   ├── app/
│   │   ├── api/             # FastAPI REST endpoints (dmo, destinations, safety, gov tourism, etc.)
│   │   ├── database/        # SQLAlchemy models (12,601 destinations, transport, gov intelligence)
│   │   ├── services/        # ML models (potential score, pricing, routing, gov intelligence service)
│   │   ├── schemas/         # Pydantic validation schemas (tourist, gov, dmo)
│   │   └── core/            # Auth, config, rate limiting, RBAC dependencies
│   ├── data/                # Real government datasets (Visits 2020-2024, ASI footfalls)
│   └── tests/               # Automated pytest suites (potential, ML, events, gov intelligence)
├── frontend/
│   ├── src/
│   │   ├── components/      # Reusable UI widgets (Timeline, SmartMap, Modal)
│   │   ├── views/           # Full portal pages
│   │   │   ├── gov/         # Government Tourism Intelligence & Gov Dashboard
│   │   │   ├── host/        # Host Hub Dashboard
│   │   │   ├── tourist/     # Tourist portal views (Dashboard, Trips, Groups, etc.)
│   │   │   └── admin/       # Admin Dashboard & DMO Admin
│   │   └── locales/         # 7 Indic language translation files (hi, bn, ta, te, etc.)
├── ml/
│   ├── config/              # config.yaml (centralized weights, thresholds)
│   ├── data/government_sources/  # 8 source CSVs (508 districts) + investment data template
│   ├── features/            # Entity resolution & feature pipeline
│   ├── models/              # Tourism potential, opportunity, infrastructure gap, priority engine
│   ├── explainability/      # Factor contribution breakdown & narrative briefing
│   ├── inference/           # Master orchestrator (orchestrator_gov.py)
│   └── schemas/             # JSON contract schema (Section 42)
├── data/                    # Master destination datasets (12,601 POIs across 36 States/UTs)
├── docs/                    # Panel-specific README documentation
│   ├── README_TOURIST.md
│   ├── README_HOST.md
│   ├── README_DMO.md
│   ├── README_GOV_INTELLIGENCE.md
│   └── README_ADMIN.md
├── scripts/                 # Ingestion, audit, coordinate correction, and test scripts
├── run.bat                  # Interactive one-click launch suite and 12-point diagnostic toolbox
└── README.md                # Project documentation and architectural overview (this file)

---

🔗 Key URLs (After Running run.bat)

Service	URL
Main Web Application	http://localhost:5173
DMO Command Center	http://localhost:5173/dmo
Gov Tourism Intelligence	http://localhost:5173/gov/tourism-intelligence
Admin Panel	http://localhost:5173/admin
Swagger API Docs	http://127.0.0.1:8000/docs
ReDoc API Docs	http://127.0.0.1:8000/redoc

---

🤝 Developed for Smart India Hackathon (SIH)

Crafted with heart, rigorous engineering, and deep respect for India's incredible heritage and hospitality.