Batcher Node
TLDR: Batcher nodes collect agents into groups before releasing them downstream as a single batch. Use for palletizing, oven loads, grouped shipping, or any process that operates on batches.
How to Batch
The first section shows:
Size: [number stepper] PCS
Set the number of agents to collect before forming a batch. The minimum is 1. For example, a batch size of 12 means the Batcher waits until 12 agents arrive, then releases them as one unit.
Release
This section (collapsed by default) controls when a batch is released. You see three toggle cards:
| Mode | Behavior | Use When |
|---|---|---|
| When Full | Wait until the batch reaches the target size | Batches must be exact (case packs, full pallets) |
| Size or Timeout | Release when full OR after a max wait time | Avoid agents waiting forever (shipping cutoffs) |
| Timeout Only | Always wait for the timeout, ignore batch size | Scheduled runs (hourly batch jobs, oven cycles) |
Max Wait Time
Only visible when "Size or Timeout" or "Timeout Only" is selected.
A number stepper with a time unit dropdown. This is how long the Batcher waits before releasing a partial batch. For example, 30 minutes means: release whatever we have after 30 minutes, even if the batch isn't full.
After batching, the group moves through the system as a single agent. Use a Separator node downstream to unbatch if needed.
Path Control
Only visible when 2 or more outgoing connections exist.
Same routing options as other nodes: Random, Custom Split, Shortest Queue, Priority, or Conditional.
Larger batches mean fewer downstream operations but longer wait times for the first agents in each batch. Smaller batches mean shorter individual waits but more frequent processing cycles. Experiment to find the sweet spot.