Multi-Destination Trip Planning: AI Decision Framework

When Multi-Destination Makes Sense

Good candidates for multi-destination:

• Trips longer than 7-10 days
• Regions with clustered but distinct breaks (Indo, Pacific Islands)
• Variable forecasts where different spots will fire on different days
• Budget flexibility allowing for inter-destination transport

Better as single-destination:

• Trips under 7 days (transit costs time and money)
• Destinations with diverse breaks nearby (no need to move)
• Strong, consistent forecast at one location

The Decision Framework

Step 1: Define the Trip Window

Establish fixed dates the user can travel. This bounds all recommendations.

Step 2: Map Forecast Conditions

Query Strike Mission API for all relevant spots across the trip window. Identify peak score days for each potential destination.

Step 3: Identify Logical Groupings

Some destinations naturally cluster:

• Bali corridor: Canggu → Uluwatu → Padang in one trip with short drives
• Mentawais: Requires boat, all spots accessed from same vessel

Step 4: Calculate Transit Costs

For each potential destination change, assess:

• Time cost: Hours in transit (flights, drives, ferries)
• Financial cost: Additional transport, accommodation changes
• Opportunity cost: Missing surf during transit

Step 5: Optimize the Sequence

• Start where conditions are best earliest
• Move as conditions shift
• End where conditions are best latest
• Minimize backtracking

Transit Decision Heuristics

Move when:

• Current location score drops below 60 AND another destination shows 75+
• Transit time is less than duration of forecast improvement

Stay when:

• Conditions are "Good" or better (60+) at current location
• Transit would cause missing a peak day anywhere
• Forecast uncertainty is high

Never move when:

• Current conditions are "Firing" or "Epic" (75+)
• User is exhausted or injured