协作头脑风暴
brainstorming
在实现前澄清需求、探索设计选择,并形成清晰的开发方向。
wave-planning-optimizer【开发与工程】Automated wave planning and pick path optimization skill to maximize warehouse throughput and order accuracy
独立 GitHub 快照 · 自动分类 · 以原作者文档为准
name: wave-planning-optimizer description: Automated wave planning and pick path optimization skill to maximize warehouse throughput and order accuracy allowed-tools:
The Wave Planning Optimizer is an automated skill that optimizes wave planning and pick path sequencing to maximize warehouse throughput and order accuracy. It intelligently groups orders into waves, balances workloads, and coordinates with carrier cutoff times to ensure efficient fulfillment operations.
skill: wave-planning-optimizer
inputs:
orders:
- order_id: "ORD001"
lines: 3
priority: "standard"
carrier_cutoff: "14:00"
zone_requirements: ["ZONE_A", "ZONE_B"]
- order_id: "ORD002"
lines: 5
priority: "expedited"
carrier_cutoff: "12:00"
zone_requirements: ["ZONE_A"]
resources:
available_pickers: 10
picker_capacity_lines_per_hour: 60
constraints:
max_wave_size: 200
batch_size_target: 12
planning_horizon_hours: 4
outputs:
waves:
- wave_id: "WAVE001"
release_time: "08:00"
orders: ["ORD002", "ORD003", "ORD004"]
total_lines: 45
estimated_completion: "09:30"
assigned_pickers: 3
batches:
- batch_id: "BATCH001"
orders: ["ORD002"]
pick_sequence: ["A-01-02", "A-03-05", "A-04-01"]
- wave_id: "WAVE002"
release_time: "09:30"
orders: ["ORD001", "ORD005"]
total_lines: 38
estimated_completion: "11:00"
assigned_pickers: 3
metrics:
total_waves: 2
average_batch_size: 10.5
estimated_throughput_lines_per_hour: 85