Infor SyteLineGlossary

What Is Phantom BOM?

Also known as: phantom assembly, blow-through, transient subassembly

Definition

A phantom BOM is a subassembly that exists in the bill of material for engineering clarity but is never stocked or built as its own order. Planning blows straight through it, passing demand directly to its components on the parent job.

Phantom BOM Explained

Phantoms solve a structural problem. Engineering wants to group thirty components into a logical subassembly - a wiring harness, a hardware kit, a modular option package - so the bill is readable and maintainable. Manufacturing never builds that group as a separate stocked item; it consumes the components directly on the parent. Marking the subassembly as a phantom gives engineering their structure while telling planning to ignore the level entirely.

When MRP or APS explodes a bill and encounters a phantom, it does not create a planned order for it. It passes the demand through, multiplying quantities by the phantom's usage, and plans the phantom's components against the parent's due date. When a job is created for the parent, the job material list shows the phantom's components rather than the phantom itself, so the shop picks real parts. Lead time offset for a phantom is conventionally zero, since no build time is consumed.

Phantoms are especially valuable in modular and configured product structures. A make-to-order manufacturer can define option groups as phantoms, letting a configurator or an engineer select modules while the resulting job materialises as a flat list of real components. Maintaining that structure through phantoms means an engineering change to a shared module updates every product that references it, rather than requiring edits to dozens of flat bills.

The classic misconception is that a phantom cannot have on-hand inventory. In practice partial builds happen - a harness gets assembled ahead, or a kit is returned from the floor - and most ERP systems, SyteLine included, will consume existing phantom stock before blowing through to components. That behavior is usually desirable but surprises planners who assumed the level did not exist at all. Verify how your configuration handles phantom on-hand before relying on it either way.

Why It Matters

  • Phantoms let engineering maintain readable modular bills without creating stocked items and orders manufacturing does not want.
  • They collapse planning levels, which shortens cumulative lead time and reduces the number of orders planners must manage.
  • A shared phantom module means one engineering change propagates to every product using it, cutting change administration dramatically.
  • Misusing phantoms for assemblies that are genuinely stocked destroys inventory visibility for those parts.

In Practice

A gotcha to watch: converting an existing stocked subassembly to a phantom while inventory is still on hand. Planning stops creating supply for it, but the existing balance sits in stock and may or may not be consumed depending on configuration. Run the balance to zero first, then change the flag, then verify the next MRP run explodes as expected.

Frequently Asked Questions

What is the difference between a phantom and a normal subassembly?

A normal subassembly is planned, ordered, built, and stocked as its own item with its own job. A phantom exists only as a level in the bill; planning blows through it and passes demand straight to its components, and no order is created for it. Use a phantom when the group is a logical structure rather than a physical stocked part.

Can a phantom item have on-hand inventory?

Yes, and it commonly does when partial builds or returns occur. Most systems, including SyteLine, will consume any existing phantom stock before exploding through to components, which is usually the desired behavior. It does surprise planners who assumed the level was purely notional, so confirm how your configuration handles phantom on-hand before designing processes around it.

Working with Phantom BOM in a live environment? Our engineers do this every day - and our AI agents automate most of it.