4.1.  Material Classification

Instances of IfcMaterial can be identified via the class’ Name string-based attribute, which may be required to be unique in a project. This attribute is appropriate to capture material names at the level needed by Reporting use-cases, such as Bill of Materials.

The IfcMaterial class also includes a Category string-based attribute in order to capture a very general classification of materials. Examples include concrete, steel, aluminum, etc. Values in the Category attribute are not standardized by IFC. Figure 5 shows an instance diagram with sample instances of the IfcMaterial class that capture various steel grades.

 

Figure 5 — Sample IfcMaterial instances

IFC can optionally capture Material Classifications defined by International Standards as instances of IfcClassificationReference, which can then be associated to IfcMaterial instances via IfcExternalReferenceRelationships. In that case, instances of IfcClassificationReference are expected to carry sufficient information to interpret material classification according to the names or classification keys defined by a specific Classification System.

Figure 6 shows a class diagram summarizing the three approaches available in IFC for material classification:

 

Figure 6 — Material Classification concepts in IFC

Figure 7 shows an instance diagram with a sample of `IfcMaterial`s and their associations to `IfcClassificationReference`s.

 

Figure 7 — Sample of Material Classification with IfcClassificationReferences

4.2.  Material Attribution

Specialized attribution associated to an instance of IfcMaterial are handled via sets of material properties captured by an instance of the IfcMaterialProperties class. IFC standardized several Property Sets applicable to `IfcMaterial`s that are normally used in various kinds of Simulations. The following table lists a sample such Property Sets and their associated kinds of Simulations.

Table 1 — Sample of Property Sets that capture specialized material attributes with applicable Kinds of Simulations

Property Sets	Example of Applicable Simulation
Pset_MaterialMechanical	Structural Analysis
Pset_MaterialConcrete	Structural Analysis
Pset_MaterialWood	Structural Analysis
Pset_MaterialSteel	Structural Analysis
Pset_MaterialEnergy	Energy Analysis
Pset_MaterialThermal	Thermal Analysis
Pset_MaterialWater	Water-Quality Analysis

5.1.  Overview

3D Tiles is an open standard, published by OGC, designed for streaming and rendering massive 3D geospatial content such as Photogrammetry, 3D Buildings, BIM/CAD, Instanced Features, and Point Clouds. It defines a hierarchical data structure and a set of tile formats which are used to efficiently manage and display 3D content, optimizing performance by dynamically loading appropriate levels of detail.

3D Tiles allows for the encoding of information associated to any level of its hierarchical data structure — including tile-set, tile, individual geometry and a pixel within it. 3D Tiles refers to information associated to any of the levels of such hierarchy as metadata. Figure 8 shows an example of metadata assigned at various granularities in a hierarchical tile structure.

 

Figure 8 — Example metadata at various granularities in 3D Tiles

5.2.  Materials in 3D Tiles

Referring specifically to materials, 3D Tiles can deliver photorealistic visuals by using Physically-Based Rendering (PBR)⁶. Moreover, 3D Tiles can capture semantics behind any attribute, as part of its metadata. The 3D Tiles specification includes the standardization of some common semantics used by metadata properties. However, at the time of this writing, such standardization of semantics has not included any classification or attribution of material information yet.

Nevertheless, a specific implementation of 3D Tiles can still capture material-related information by introducing it as application-specific attributes. The following class declaration snippet, encoded in JSON according to the 3D Tiles specification, shows the definition of a two string properties named Material_Name and Material_Grade, onto a class named Column.

 

{
  "entity": {
    "class": "Column",
    "properties": {
      "stringProperty": "Material_Name",
      "stringProperty": "Material_Grade"
    }
  }
}

Listing 1 — Sample material properties added to a class for 3D Tiles, encoded in JSON

6.1.  Material Classification

A Physical Material in BIS is modeled via a class-hierarchy with a common base-class — PhysicalMaterial — defined under the DefinitionElement branch of the ecosystem. `DefinitionElement`s in BIS model information meant to be referenced or shared. Figure 9 depicts some of the Physical Material classes that BIS has standardized so far in its ecosystem.

 

Figure 9 — Material Class Hierarchy in BIS

BIS uses the PhysicalMaterial class-hierarchy to encode a general classification of materials that can be understood by machines. More specific classifications can be captured in BIS via one of the following two approaches:

Any instance in the BIS ecosystem, called Element, can optionally carry a human-friendly business key that is unique in a certain context, and thus, it can be used as a form of identification. This special property of any BIS Element is referred to as Code. The first approach listed above relies on the Code property of PhysicalMaterial instances to capture the desired classification key. Figure 10 shows a few sample instances of the Steel class, with their Code properties assigned to their classification key according to ASTM International.

 

Figure 10 — Sample Steel Instances with classification keys and descriptions

The BIS ecosystem contains a set of classes meant to capture Classification Systems that can be used to provide additional classifications in parallel to BIS instances. They can be used to capture a more complete representation of a Classification System, or to classify any BIS instance according to one or more Classification Systems in parallel. Figure 11 shows the same Steel instances from Figure 10 but this time associated to more complete classification tables according to ASTM International, by using BIS ClassificationSystems_ classes.

 

Figure 11 — Sample Steel Instances classified based on BIS ClassificationSystems classes

6.2.  Material Attribution

Being modeled as instances, PhysicalMaterial BIS elements can carry specialized attribution. In general, BIS offers two approaches to capture attribution information for any element:

The second approach is the one applicable to material attribution. A BIS Element-Aspect class captures sets of properties that can be optionally attached to an Element instance. In the case of Material attribution, it enables the definition of Element-Aspect classes that capture sets of properties per specialized Kind of Simulation. These Element-Aspects can then be associated to particular PhysicalMaterial instances. Figure 12 shows instances of the Steel class, attributed with mechanical properties, defined in the SteelMechPropertiesAspect Element-Aspect class, attached to those instances.

 

Figure 12 — Sample Steel Instances with Attribution via Element-Aspects

The following table lists a sample of Element-Aspect classes in the BIS ecosystem that capture specialized material attribution and their associated kinds of Simulations.

Table 2 — Sample of Element-Aspect classes that capture specialized material attributes with applicable Kinds of Simulations

Element-Aspect	Example of Applicable Simulation
GenericMechPropertiesAspect	Structural Analysis
ConcreteMechPropertiesAspect	Structural Analysis
SteelMechPropertiesAspect	Structural Analysis
HydraulicMaterialAspect	Hydraulic Analysis

7.1.  Overview

Granta MI is a commercial Materials Data Management (MDM) system by Ansys and has been developed over the last 20 years to manage materials data for many industries and applications including design & simulation, test data management, sustainability, and computational materials design. It has a flexible schema, which allows it to be configured to many different use cases, but to aid implementation and standardization a number of templated configurations have been developed incorporating best practice for MDM. The use of schema element building blocks allow new configurations to be quickly developed, while retaining the best practice standards and interoperability with downstream systems such as simulation.

7.2.  Material Classification

As Granta MI is a database system, each material is represented as a record in the system. There are two methods to classify materials.

 

Figure 13 — A typical hierarchical materials classification in Granta MI

The use of these attributes means that the materials can easily be filtered and analyzed by the user and programmatically. A tree hierarchy can also be built from these attributes giving flexibility in user experience. However, when classifying many different material classes, the number of attributes required will be large, leading to redundancy. Granta MI has a number of standard templates to classify materials, from a simple set of attributes which can be used across all material classes, to more complex sets for a detailed classification of a particular class of materials.

 

Figure 14 — A simple attribute-based classification system in Granta MI

7.3.  Material Attribution

The properties of materials are captured in Attributes defined in the schema. Like most database systems, these attributes can be a number of different data-types, e.g. numerical, text, lists, files and images. As the system was developed specifically for engineering data, it has a number of complex data types that can be utilized such as tables, grids, data series, and mathematical expressions. This allows for a very flexible data structure, but for key use cases, preconfigured templates are available to promote best practice, standardization and interoperability.