Skip to content

Capacity Planning

Size your warehouse system for target throughput.

Goal

By the end of this tutorial, you will:

  • Calculate required fleet size
  • Determine station capacity needs
  • Plan for peak load and growth
  • Validate capacity with simulation

Time: 45 minutes

Prerequisites

  • Completed Benchmarking
  • Understanding of throughput concepts

Step 1: Define Requirements

Start with business requirements:

Requirements:
  Target throughput: 1,500 orders/hour
  Operating hours: 16 hours/day
  Peak factor: 1.3× average
  Growth buffer: 20%
  SLA: 95% orders in <60s

Calculate design capacity:

Design capacity = Target × Peak factor × Growth buffer
                = 1,500 × 1.3 × 1.2
                = 2,340 orders/hour

Step 2: Baseline Measurement

Measure current capacity:

waremax benchmark scenario.yaml --find-max-throughput

Current Configuration:
  Robots: 10
  Stations: 2 (concurrency 2 each)

Maximum Throughput: 1,050/hr
Limiting Factor: Robot fleet

Capacity gap: 2,340 - 1,050 = 1,290/hr needed

Step 3: Fleet Sizing

Theoretical Calculation

Tasks per robot = Throughput / Robot count
               = 1,050 / 10
               = 105 tasks/robot/hour

For 2,340/hr target:
  Robots needed = 2,340 / 105 = 22.3
  Round up: 23 robots
  Add margin: 25 robots

Validate with Simulation

waremax sweep scenario.yaml \
  --param "robots.count=[20,22,25,28,30]" \
  --target-throughput 2340

robots.count  Max Throughput  Meets Target?
20            1,850/hr        No
22            2,100/hr        No
25            2,380/hr        ✓ (margin: 2%)
28            2,650/hr        ✓ (margin: 13%)
30            2,580/hr        ✓ (congestion)

Optimal: 25-28 robots

Step 4: Station Capacity

Calculate Station Needs

Service rate per slot = 3600 / avg_service_time
                      = 3600 / 5s
                      = 720 tasks/hour per slot

Total slots needed = Target / Service rate
                   = 2,340 / 720
                   = 3.25 slots

Current: 2 stations × 2 slots = 4 slots ✓

Validate with Simulation

waremax sweep scenario.yaml \
  --param robots.count=25 \
  --param "stations[*].concurrency=[1,2,3]" \
  --target-throughput 2340

Total Slots  Max Throughput  Station Util
2 (1+1)      1,420/hr        98%
4 (2+2)      2,380/hr        82%
6 (3+3)      2,420/hr        55%

4 slots (2+2) is sufficient

Step 5: Traffic Capacity

Edge Capacity Analysis

waremax analyze results/ --traffic-capacity

Edge Capacity Analysis:

Critical Edges:
  E15: 92% utilized (at risk)
  E22: 85% utilized
  E8:  78% utilized

For 2,340/hr (2.2× current):
  E15 would need: 92% × 2.2 = 202% ← Over capacity!

Action: Increase E15 capacity or add parallel path

Solution

# Increase edge capacity
edges:
  - from: 14
    to: 15
    capacity: 2  # Was 1

Validate:

waremax run scenario_wide_edge.yaml \
  --param robots.count=25 \
  -o results_wide/

Step 6: Peak Load Planning

Define Peak Scenarios

# peak_scenario.yaml
orders:
  generation:
    type: variable
    schedule:
      - time: 0
        rate: 1500    # Normal
      - time: 3600
        rate: 2340    # Peak (1.5× for 1 hour)
      - time: 7200
        rate: 1500    # Return to normal

Test Peak Handling

waremax run peak_scenario.yaml \
  --param robots.count=25 \
  -o peak_results/

Peak Period Analysis (hour 2):
  Order rate: 2,340/hr
  Throughput: 2,280/hr
  Queue buildup: 60 orders
  Recovery time: 12 minutes

Result: System handles peak with minor queue buildup

Step 7: Growth Planning

Capacity Curves

waremax sweep scenario.yaml \
  --param "orders.rate=[1000,1500,2000,2500,3000]" \
  --param "robots.count=[15,20,25,30,35]" \
  -o capacity_matrix/

Capacity Matrix (Throughput):

Order Rate    15 robots  20 robots  25 robots  30 robots
1,000         980        1,000      1,000      1,000
1,500         1,450      1,500      1,500      1,500
2,000         1,850      2,000      2,000      2,000
2,500         2,100      2,420      2,500      2,500
3,000         2,250      2,680      2,920      3,000

Scaling recommendations:
  - 1,500/hr: 20 robots adequate
  - 2,500/hr: 25 robots needed
  - 3,000/hr: 30 robots + station upgrade

Growth Timeline

Current:        1,000/hr (10 robots)
Year 1 target:  1,500/hr → Add 10 robots
Year 2 target:  2,500/hr → Add 5 robots + station
Year 3 target:  3,500/hr → Major expansion

Step 8: Create Capacity Plan

# Capacity Plan

## Current State
- Throughput capacity: 1,050/hr
- Robots: 10
- Stations: 2 (concurrency 2)

## Target State (Design)
- Throughput capacity: 2,340/hr
- Design factor: 1.3 peak × 1.2 growth

## Changes Required

### Fleet
| Current | Target | Change |
|---------|--------|--------|
| 10      | 25     | +15    |

### Stations
| Current | Target | Change |
|---------|--------|--------|
| 4 slots | 4 slots| None   |

### Infrastructure
| Item | Current | Target | Change |
|------|---------|--------|--------|
| Edge E15 | Cap 1 | Cap 2 | Widen |

### Charging
| Current | Target | Change |
|---------|--------|--------|
| 4 bays  | 8 bays | +4     |

## Validation Results
- Simulated throughput: 2,380/hr ✓
- Peak handling: OK ✓
- SLA compliance: 96% ✓

## Timeline
1. Phase 1: Add 10 robots, widen E15
2. Phase 2: Add 5 robots, 4 charging bays
3. Monitor and adjust

Step 9: Validate Complete Plan

Run full validation:

waremax run capacity_plan_scenario.yaml \
  --duration 86400 \   # 24-hour test
  --runs 5 \
  -o capacity_validation/

Capacity Validation (24-hour simulation):

Throughput:
  Average: 2,320/hr
  Peak hour: 2,580/hr
  Minimum: 2,180/hr

SLA Performance:
  Orders < 60s: 96.2%
  Orders < 90s: 99.1%

Resource Utilization:
  Robots: 72%
  Stations: 78%
  Charging: 65%

Verdict: Capacity plan validated ✓

Capacity Planning Formulas

Quick Estimates

Fleet size: 

Robots ≈ (Target throughput × Avg task time) / 3600
       ≈ (2,340 × 50) / 3600
       ≈ 32.5 → 35 robots (with margin)

Station slots: 

Slots ≈ Target throughput / (3600 / Avg service time)
     ≈ 2,340 / (3600 / 5)
     ≈ 3.25 → 4 slots

Charging bays: 

Bays ≈ Fleet × Charging time / (Operating time + Charging time)
    ≈ 25 × 2 / (8 + 2)
    ≈ 5 → 6 bays (with margin)

Best Practices

Include Margins

Production capacity = Design capacity × Safety margin
                    = Theoretical × 1.1 to 1.2

Plan for Variability

Peak handling: 1.3× to 1.5× average
Daily variation: Consider high and low periods

Validate Before Implementing

Always simulate full configuration
Test peak scenarios
Verify SLA compliance

Next Steps