38980-vm/app-9w9pd00g5j41/ANALYZE_TRIP_ENHANCEMENT.md

Analyze-Trip Edge Function Enhancement

Overview

Enhanced the analyze-trip edge function with advanced metrics calculation, density-based scoring, and comprehensive debug information to provide intelligent tour recommendations.

Key Enhancements

1. Distance & Duration Calculations

Each place now includes:

• Distance from previous place: Calculated using Haversine formula (km)
• Travel time from previous place: Estimated based on distance (assuming 40 km/h average speed)
• Visit duration: Parsed from duration string (e.g., "2 hours" → 120 minutes)

Example Output:

{
  "name": "Göreme Open Air Museum",
  "type": "museum",
  "lat": 38.6425,
  "lng": 34.8317,
  "distanceFromPreviousKm": 3.2,
  "travelTimeFromPreviousMinutes": 5,
  "visitDurationMinutes": 120
}

2. Daily Density Score

Each day is analyzed with a comprehensive density score:

Formula:

density_score = (total_distance_km * 5 + total_time_hours * 10) / number_of_places

Density Levels:

• Low (<20): Self-planning is manageable
• Moderate (20-35): Tour is optional but could add value
• High (35-50): Tour is recommended
• Very High (≥50): Tour is highly recommended

Metrics Calculated:

• Total places per day
• Total distance traveled (km)
• Total travel time (minutes)
• Total visit time (minutes)
• Total time commitment (hours)
• Density score and level

Example Daily Metrics:

{
  "dayNumber": 1,
  "date": "2024-06-15",
  "totalPlaces": 5,
  "totalDistanceKm": 45.3,
  "totalTravelTimeMinutes": 68,
  "totalVisitTimeMinutes": 360,
  "totalTimeMinutes": 428,
  "densityScore": 42.8,
  "densityLevel": "high"
}

3. AI Decision Logic Based on Density

The AI now makes recommendations based on density scores:

Decision Thresholds:

• Density ≥50: Highly recommend tour (confidence ≥0.85)
• Density 35-50: Recommend tour (confidence 0.70-0.85)
• Density 20-35: Optional tour (confidence 0.50-0.70)
• Density <20: Don't recommend tour (confidence <0.50)

Additional Factors:

1. Total Distance: >100km strongly suggests organized transportation
2. Time Commitment: >8 hours/day suggests professional guidance
3. Group Size: ≥4 travelers benefit from private guide
4. Place Count: ≥5 places/day requires efficient routing

4. Debug Information

Comprehensive debug info explaining every recommendation:

Structure:

{
  "debug_info": {
    "dailyMetrics": [...],
    "overallMetrics": {
      "totalDays": 3,
      "totalPlaces": 12,
      "totalDistanceKm": 125.7,
      "totalTimeHours": 18.5,
      "averageDensityScore": 38.2,
      "maxDensityScore": 48.5
    },
    "decisionFactors": [
      {
        "factor": "High Density Day",
        "value": 48.5,
        "impact": "positive",
        "reasoning": "At least one day has high density (35-50), suggesting tour guidance would improve experience."
      },
      {
        "factor": "Long Distance Travel",
        "value": "126 km",
        "impact": "positive",
        "reasoning": "Total distance exceeds 100km, organized transportation would save time and reduce stress."
      }
    ],
    "recommendation_reasoning": "AI Analysis: Your plan has high density with 126km total distance. A guided tour would optimize routing and save approximately 2 hours. Confidence: 82%. Max density score: 48.5."
  }
}

Technical Implementation

Helper Functions

1. calculateDistance(): Haversine formula for accurate distance calculation
2. parseDurationToMinutes(): Converts duration strings to minutes
3. estimateTravelTime(): Calculates travel time based on distance
4. calculateDensityScore(): Computes daily density score
5. getDensityLevel(): Categorizes density into levels
6. analyzeTripMetrics(): Main analysis function for all days

Enhanced Interfaces

interface Place {
  name: string;
  type: string;
  lat?: number;
  lng?: number;
  duration?: string;
  // Calculated metrics
  distanceFromPreviousKm?: number;
  travelTimeFromPreviousMinutes?: number;
  visitDurationMinutes?: number;
}

interface DayMetrics {
  dayNumber: number;
  date: string;
  totalPlaces: number;
  totalDistanceKm: number;
  totalTravelTimeMinutes: number;
  totalVisitTimeMinutes: number;
  totalTimeMinutes: number;
  densityScore: number;
  densityLevel: 'low' | 'moderate' | 'high' | 'very_high';
  places: Place[];
}

interface DebugInfo {
  dailyMetrics: DayMetrics[];
  overallMetrics: {
    totalDays: number;
    totalPlaces: number;
    totalDistanceKm: number;
    totalTimeHours: number;
    averageDensityScore: number;
    maxDensityScore: number;
  };
  decisionFactors: {
    factor: string;
    value: string | number;
    impact: 'positive' | 'negative' | 'neutral';
    reasoning: string;
  }[];
  recommendation_reasoning: string;
}

AI Prompt Enhancement

The AI prompt now includes:

• Detailed density analysis for each day
• Per-place distance and travel time information
• Decision factors with impact assessment
• Clear density-based decision guidelines
• Confidence calculation formula

Response Examples

High Density Trip (Recommend Tour)

{
  "recommend": true,
  "reason": "Your itinerary has very high density (score: 52.3) with 135km total distance. A guided tour would optimize routing and save approximately 3 hours.",
  "recommended_type": "daily_tour",
  "daily_tour_slug": "red_tour",
  "confidence": 0.87,
  "comparison_metrics": {
    "distance_saved_km": 40,
    "time_saved_hours": 3.2,
    "logistics_removed": ["Ticket purchasing", "Transfer arrangement", "Guide finding", "Route planning"],
    "expert_value": ["Local expert knowledge", "Historical information", "Hidden spots", "Local recommendations"]
  },
  "debug_info": {
    "overallMetrics": {
      "maxDensityScore": 52.3,
      "totalDistanceKm": 135.0,
      "totalTimeHours": 22.5
    },
    "decisionFactors": [
      {
        "factor": "Very High Density Day",
        "value": 52.3,
        "impact": "positive",
        "reasoning": "At least one day has very high density (≥50), indicating complex logistics that would benefit from professional tour organization."
      }
    ]
  }
}

Low Density Trip (No Recommendation)

{
  "recommend": false,
  "reason": "Your plan has low density (score: 15.2) with manageable distances. Self-planning is feasible.",
  "confidence": 0.38,
  "debug_info": {
    "overallMetrics": {
      "maxDensityScore": 15.2,
      "totalDistanceKm": 28.0,
      "totalTimeHours": 8.5
    },
    "decisionFactors": [
      {
        "factor": "Low Density",
        "value": 15.2,
        "impact": "negative",
        "reasoning": "Days have low density (<20), self-planning is manageable."
      }
    ]
  }
}

Benefits

1. Data-Driven Decisions: Recommendations based on quantifiable metrics, not just place count
2. Transparency: Users can see exactly why a tour was recommended
3. Accurate Calculations: Real distance and time estimates using geographic coordinates
4. Flexible Thresholds: Density scoring adapts to different trip complexities
5. Debugging Support: Comprehensive debug info helps developers understand AI decisions

Usage

The function is automatically called when analyzing trip itineraries. No changes needed to the API call:

const response = await supabase.functions.invoke('analyze-trip', {
  body: {
    destination: 'Cappadocia',
    days: [...],
    travelers: 2,
    interests: ['history', 'nature']
  }
});

// Response now includes debug_info
const { recommend, confidence, debug_info } = response.data;

Future Enhancements

Potential improvements:

• Real-time traffic data integration
• Weather-based adjustments
• Seasonal crowd density factors
• User feedback loop for confidence calibration
• Machine learning model for pattern recognition