Object Tree 1.1

Introduction

The COINS Building information model (CBIM) is primarily structured around an object tree. function fulfillers are organized according the decomposition principle, i.e. part-of relationships between an assembly object (parent) and its constituting part objects (children). A part object in its turn may act as an assembly object for even smaller part objects, etcetera until atomic part objects (the leaves of the decomposition tree) are reached. Leaf objects are the basic building stones of this structure: we know either how to produce them or how to obtain from a third party (supply company).

Object tree classes

Figure 1 shows the UML representation of the object tree data structure. Since Function Fulfiller is an abstract class (no instances) the actual object tree contains physical objects or spaces.

Figure 1: Object tree classes.
Figure 1: Object tree classes.
These function fulfiller types cannot be mixed, i.e. a physical object cannot decompose into spaces and vice versa. This is explicitly specified in the supproperties physicalChild/physicalParent and spatialChild/spatialParent of the child/parent relations.

The interconnection between a physical object tree and a space tree can be specified with the isSituatedIn/situates relation.

Figure 2: A parent object and its immediate child objects should reside in adjacent layers. The parent layer index and the child layer index differ exactly one
Figure 2: A parent object and its immediate child objects should reside in adjacent layers. The parent layer index and the child layer index differ exactly one
Afkomstig van CoinsWiki NL, de Vrije Encyclopedie. "http://www.coinsweb.nl/wiki/index.php/Object_Tree_1.1"
Personal tools