Skip to content

Stations

Service points where robots are processed.

Station Types

Type Purpose
Pick Order picking/fulfillment
Drop Item delivery
Inbound Receiving goods
Outbound Shipping goods
Charging Battery charging
Maintenance Robot repair/maintenance

Station Model

Components

Station
├── Location (node)
├── Concurrency (service slots)
├── Queue (waiting robots)
└── Service Time Model

Service Flow

graph LR
    A[Robot Arrives] --> B{Queue Full?}
    B -->|No| C[Join Queue]
    B -->|Yes| D[Wait/Reject]
    C --> E{Slot Available?}
    E -->|Yes| F[Start Service]
    E -->|No| G[Wait in Queue]
    G --> E
    F --> H[Service Time]
    H --> I[Complete]
    I --> J[Depart]

Concurrency

Number of robots serviced simultaneously.

stations:
  - id: "S1"
    concurrency: 3  # 3 robots at once

Effective Capacity

Throughput ≈ concurrency × 3600 / avg_service_time

Example:

  • Concurrency: 2
  • Avg service: 10s
  • Throughput: 720 tasks/hour max

Queue

Queue Capacity

stations:
  - id: "S1"
    queue_capacity: 15

If not specified, queue is unlimited.

Queue Behavior

When queue is full:

  • Robot must wait elsewhere
  • May reroute to another station

Service Time

Constant

Fixed time per service:

service_time_s:
  distribution: constant
  base: 5.0
  per_item: 2.0

Total = base + (per_item × items)

Lognormal

Variable, right-skewed:

service_time_s:
  distribution: lognormal
  base: 8.0
  base_stddev: 2.0
  per_item: 2.0
  per_item_stddev: 0.5

Realistic for human operators.

Exponential

Memoryless:

service_time_s:
  distribution: exponential
  base: 10.0

For queuing theory models.

Uniform

Even distribution:

service_time_s:
  distribution: uniform
  min_s: 5.0
  max_s: 15.0
  per_item: 1.5

Pick Stations

Most common station type.

Function

  • Receive picked items from robots
  • Process for order fulfillment
  • May involve human operators

Configuration Example

stations:
  - id: "pick_zone_a"
    node: "N100"
    type: pick
    concurrency: 2
    queue_capacity: 10
    service_time_s:
      distribution: lognormal
      base: 6.0
      base_stddev: 1.5
      per_item: 2.0
      per_item_stddev: 0.5

Charging Stations

Function

  • Recharge robot batteries
  • Return robots to service

Properties

Property Description
bays Concurrent charging slots
charge_rate_w Power per bay
queue_capacity Waiting capacity

Charge Time

time = (capacity × (1 - soc)) / charge_rate

Maintenance Stations

Function

  • Scheduled maintenance
  • Failure repair
  • Return robots to service

Properties

Property Description
bays Concurrent service slots
maintenance_duration_s Scheduled maintenance time
repair_time Variable repair duration

Station Assignment

Policies determine which station handles each task:

Policy Behavior
least_queue Shortest queue
fastest_service Soonest completion
due_time_priority Based on order urgency

Performance Metrics

Utilization

utilization = busy_time / (total_time × concurrency)

Average Queue

avg_queue = total_wait_time / observation_time

Throughput

throughput = tasks_completed / measurement_time