A routing turns a production order into a sequence of steps that can genuinely be run on the shop floor. Build the process visually, use specifications as its basis, and manage every change with version control without disrupting orders already in flight.
Real production processes do not always run in a straight line. A job can branch, several steps can run at once, and an inspection result can decide whether a unit moves on or goes back for rework. All of it can be modelled on one canvas.
| Step | How it behaves |
|---|---|
| 10 · Material preparation | Sequential: the next process starts once preparation is complete. |
| 20 · Sub-assembly A | Runs as one branch of the process. |
| 30 · Sub-assembly B | Runs in parallel with step 20, and both must finish before the process continues. |
| 40 · Final assembly | Rejoins the output of the two parallel paths. |
| 50 · Functional test | The test result determines the next path. |
| 60 · Pack & ship | Units with a PASS result continue. Units with a FAIL result return to step 40 for rework. |
Branches are decided by the conditions that actually occur during execution. The routing defines the rules and the possible paths, while the final decision follows the real result on the shop floor.
The questions that usually come up as process and IT teams start designing routings.
Yes. Operators record a result at a conditional step, and the system determines the next path from it. A PASS continues to the next process, while a FAIL sends the unit to rework.
Orders already released keep using the routing version assigned to them. The change is created as a new version and used for subsequent orders.
Not necessarily. Routings can be built on a visual canvas with drag-and-drop. AI can also draft one from a plain-language description of the process, and that draft is then reviewed and confirmed before use.