Optimal Fleet Configuration: Why 15×150t Beats Both Extremes

We built a discrete event simulator to answer a fundamental fleet sizing question for Phase 0: Should we deploy fewer large vehicles or more small ones? The answer lies in the middle, and it could save Project Dyson significant operational costs.

The Question

The consensus specification for Transport Vehicles calls for a $2B fleet budget with flexibility on configuration. The divergent views range from Gemini's preference for 150,000 kg payloads (faster transit) to Claude's 200,000 kg baseline to GPT's variable configuration. We built a Monte Carlo simulator to find the optimal trade-off.

The Key Finding

15 vehicles with 150,000 kg payload capacity achieves the lowest cost per kg delivered at ~$0.2k/kg.

This mid-range configuration outperforms both extremes:

Configuration	Throughput	Cost/kg	Failure Impact
5×300t	Lower	Higher	20% per vehicle
8×250t	Moderate	Moderate	12.5% per vehicle
10×200t	Moderate	Moderate	10% per vehicle
15×150t	High	$0.2k/kg	6.7% per vehicle
20×100t	Moderate	Higher	5% per vehicle
25×80t	Lower	Higher	4% per vehicle

Why the Middle Wins

Transit Time Efficiency

The simulation models transit time as scaling with payload mass. A 150t vehicle achieves better thrust-to-mass ratios than a 250t vehicle, completing more round trips per year. This compounding effect over the 15-year mission duration creates substantial throughput differences.

Redundancy Without Overhead

With 15 vehicles, losing one unit costs only 6.7% of fleet capacity. The 8×250t configuration loses 12.5%—nearly double the impact. Yet going to 25×80t adds operational complexity without proportionate throughput gains.

Queue Smoothing

More frequent, smaller deliveries reduce queue depth variability at the Manufacturing Hub. The 15×150t configuration hits the sweet spot where delivery frequency is high enough to smooth material flow without creating scheduling chaos.

The Physics Behind It

The fundamental constraint is propulsion system scaling. Hall-effect thrusters scale non-linearly—doubling payload doesn't require doubling propulsion mass. This creates economies of scale that favor larger vehicles. However, three factors push back:

1. Solar array constraints (300-500 m² limit) restrict power available for propulsion
2. Larger payloads require longer loading/unloading cycles, reducing effective throughput
3. Failure risk compounds—each large vehicle lost hurts more

The 150t payload class threads this needle optimally.

Sensitivity Analysis

We tested the configuration across various failure rates and mission durations:

Failure Rate Impact (15-year mission):

• 0% failure: 15×150t optimal
• 3% failure: 15×150t optimal
• 10% failure: 15×150t still optimal (higher redundancy value)

Mission Duration Impact (3% failure):

• 5 years: 15×150t optimal
• 15 years: 15×150t optimal
• 25 years: 15×150t optimal, with larger margin over alternatives

The configuration is robust across parameter ranges.

Implications for Project Dyson

1. Adopt 15×150t as Baseline Configuration

This provides optimal throughput-to-cost ratio while maintaining acceptable redundancy.

2. Standardize Cargo Containers for 150t Vehicle Bays

Early container standardization enables logistics optimization across the entire supply chain.

3. Maintain Design Margin for Human Rating

The 150t vehicle size can accommodate structural upgrades for future crew transport capability without redesigning the entire fleet.

4. Optimize Hall-Effect Thruster Clusters for 150t Payload Class

Propulsion system development should target this payload range, not the original 200-250t estimates.

Cost Implications

At $200/kg delivered, the optimized fleet can transport substantial material mass over its 15-year lifetime:

• Per vehicle: ~10+ mission cycles
• Fleet throughput: Sufficient for multi-gigawatt solar collector production
• Total investment: $2B fleet + operations

The $200/kg figure compares favorably to Earth launch costs while providing the mass throughput needed for Phase 1 manufacturing.

Try It Yourself

We've published the interactive simulator so you can explore these trade-offs. Adjust vehicle count, payload capacity, failure rates, and mission duration to see how cost per kg and throughput change.

Methodology

The simulation uses:

• Discrete event simulation with priority queue (binary heap)
• Full logistics cycle: loading (2-5 hours) → transit → unloading (2-5 hours) → return
• Transit time scaling with payload mass via thrust-to-mass ratio
• Exponential failure distribution based on annual failure rate
• 50+ Monte Carlo runs per configuration for statistical validity

Results should be interpreted as relative comparisons between configurations.

What's Next

This research answers RQ-0-19 and provides the second validated fleet optimization for Phase 0. Combined with the constellation sizing results (RQ-0-3) and spectral processing findings (RQ-0-2), we're building a rigorous, simulation-validated foundation for the construction plan.

Remaining research priorities include:

• Detailed propulsion system sizing for 150t payload class
• Cargo container standardization study
• Human-rating upgrade path analysis

---

Research Question: RQ-0-19: Fleet size vs vehicle capacity tradeoff

Interactive Tool: Fleet Logistics Simulator