# IfcGrid spatial composition and relative placements

**URL:** <https://forums.buildingsmart.org/t/ifcgrid-spatial-composition-and-relative-placements/1471>\
**Category:** Uncategorized\
**Created:** [June 5, 2019, 6:56am UTC](https://forums.buildingsmart.org/t/ifcgrid-spatial-composition-and-relative-placements/1471 "2019-06-05T06:56:43Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Moult](https://sea2.discourse-cdn.com/flex016/user_avatar/forums.buildingsmart.org/moult/32/196_2.png) [@Moult](https://forums.buildingsmart.org/u/Moult)\
**Post date:** [June 5, 2019, 6:56am UTC](https://forums.buildingsmart.org/t/ifcgrid-spatial-composition-and-relative-placements/1471/1 "2019-06-05T06:56:43Z")

</div>

`IfcLocalPlacement` documented [here](https://standards.buildingsmart.org/IFC/DEV/IFC4_2/FINAL/HTML/schema/ifcgeometricconstraintresource/lexical/ifclocalplacement.htm) states that:

> For [IfcGrid](https://standards.buildingsmart.org/IFC/DEV/IFC4_2/FINAL/HTML/schema/ifcproductextension/lexical/ifcgrid.htm) and [IfcAnnotation](https://standards.buildingsmart.org/IFC/DEV/IFC4_2/FINAL/HTML/schema/ifcproductextension/lexical/ifcannotation.htm) the convention applies that it shall be placed relative
> 
> - to the local placement of its container, either [IfcSite](https://standards.buildingsmart.org/IFC/DEV/IFC4_2/FINAL/HTML/schema/ifcproductextension/lexical/ifcsite.htm), [IfcBuilding](https://standards.buildingsmart.org/IFC/DEV/IFC4_2/FINAL/HTML/schema/ifcproductextension/lexical/ifcbuilding.htm), or [IfcBuildingStorey](https://standards.buildingsmart.org/IFC/DEV/IFC4_2/FINAL/HTML/schema/ifcproductextension/lexical/ifcbuildingstorey.htm)
> - it should be the same container element that is referenced by the [IfcRelContainedInSpatialStructure](https://standards.buildingsmart.org/IFC/DEV/IFC4_2/FINAL/HTML/schema/ifcproductextension/lexical/ifcrelcontainedinspatialstructure.htm) containment relationship,

There is also the implementation agreement [here](http://www.buildingsmart-tech.org/implementation/ifc-implementation/ifc-impl-agreements/cv-2x3-143) which states:

> The spatial containment of IfcSpatialStructureElement forms a tree, and the local relative placement of IfcSpatialStructureElement forms another tree - both trees should be identical (i.e. if an IfcBuildingStorey is contained in #201=IFCBUILDING, then the IfcBuildingStorey.ObjectPlacement.PlacementRelTo shall point to the same instance #201=IFCBUILDING.

Here is a sample composition tree:

- `IfcProject`
  - (`IfcRelAggregates`) `IfcSite`
    - (`IfcRelContainedInSpatialStructure`) `IfcGrid`
    - (`IfcRelAggregates`) `IfcBuilding`
      - (`IfcRelAggregates`) `IfcBuildingStorey`
        - (`IfcRelContainedInSpatialStructure`) `IfcWall`

Can I confirm with others here that the `IfcWall` is allowed to have an `IfcGridPlacement`, but the `IfcBuilding` is _not_ allowed to have an `IfcGridPlacement`? (It must have an `IfcLocalPlacement` relative to the `IfcSite`’s placement as per the implementation agreement)

If so, it strikes me as odd that the origin of the building, which is often a grid intersection, cannot have an `IfcGridPlacement`.

---

<div class="post-metadata">

**Author:** ![sergej](https://avatars.discourse-cdn.com/v4/letter/s/bcef8e/32.png) [@sergej](https://forums.buildingsmart.org/u/sergej)\
**Post date:** [June 11, 2019, 8:27am UTC](https://forums.buildingsmart.org/t/ifcgrid-spatial-composition-and-relative-placements/1471/2 "2019-06-11T08:27:12Z")

</div>

> [@Moult](#):
>
> to the local placement

> [@Moult](#):
>
> and the local relative placement

Are you referring to these two? None of them specifically say that only IfcLocalPlacement is allowed. 🙂

Joke aside, you have a point. This should be clarified. I could not find any restrictions in the EXPRESS schema. The implementer quote is more specific, so, I guess the ISG has a more fleshed out view on this.

---

<div class="post-metadata">

**Author:** ![Moult](https://sea2.discourse-cdn.com/flex016/user_avatar/forums.buildingsmart.org/moult/32/196_2.png) [@Moult](https://forums.buildingsmart.org/u/Moult)\
**Post date:** [September 3, 2019, 5:51am UTC](https://forums.buildingsmart.org/t/ifcgrid-spatial-composition-and-relative-placements/1471/3 "2019-09-03T05:51:57Z")

</div>

Sorry for bumping this thread, but I am running into this issue again.

---

<div class="post-metadata">

**Author:** ![Moult](https://sea2.discourse-cdn.com/flex016/user_avatar/forums.buildingsmart.org/moult/32/196_2.png) [@Moult](https://forums.buildingsmart.org/u/Moult)\
**Post date:** [August 4, 2021, 11:32pm UTC](https://forums.buildingsmart.org/t/ifcgrid-spatial-composition-and-relative-placements/1471/4 "2021-08-04T23:32:05Z")

</div>

Re-reading this 2 years later, I think this is actually not an issue. The key phrase is:

> The following conventions shall apply as default relative positions **if** the relative placement is used.

So, **if** we choose not to use a relative placement, the implementer agreement doesn’t apply.