ACTIVE3D: SEMANTIC AND MULTIMEDIA MERGING FOR

FACILITY MANAGEMENT

Renaud Vanlande

Groupe Archimen, 2 rue Renée Char, BP 66606, 21066 Dijon Cedex, France

Christophe Cruz, Christophe Nicolle

LE2I, UMR CNRS 5158, Université de Bourgogne, BP 47870, 21078 Dijon Cedex, France

Keywords: Interoperability, 3D collaborative platform, IFC, BIM, Semantic Web.

Abstract: This paper presents a Semantic Web approach for facility management. This Web-based platform lets

geographically dispersed project participants—from facility managers and architects to electricians to

plumbers—directly use and exchange project documents in a centralized virtual environment using a simple

Web browser. A 3D visualization lets participants move around in the building being designed and obtain

information about the objects that compose it. This approach is based on a semantic model called CDMF

and IFC 2x3. CDMF improves data management during the lifecycle of a building. Based on graph

combinations and the contextual element SystemGraph, our proposition, addresses the problem of model

evolution, of data mapping, management, of temporal data and the problem of the data adaptation according

to the use and the user. Our framework, based on Building Information Modeling features, facilitates data

maintenance (data migration, model evolution) during the building lifecycle and reduces the volume of data.

1 INTRODUCTION

The building lifecycle management requires the

development of a specific environment solving at the

same time the problems of syntactic and semantic

heterogeneity (Keith, 2004), (Barret, 2003).

Moreover, the environment should also allow the

required extensibility and the flexibility in order to

guarantee the coherent evolution of the collaborative

processes developed in this field. The information in

an AEC project is generated during all the building

lifecycle. It is essential to structure the information

in a relevant way for a better management. The

activities in an AEC project generate a huge variety

of data and information. Consequently, the

management and the communication of these data

by various participants are complex. The process of

information sharing requires a framework in which

computer programs can exchange data automatically

regardless of the software and data location.

Moreover, in this field, the use of tools for 3D

visualization of the buildings is crucial. Towards this

goal the IAI proposed a standard called IFC (IFC,

2007) that specifies object representations for AEC

projects (IAI, 2007). Industry foundation classes

(IFCs) include object specifications, or classes, and

provide a structure for data sharing among

applications. From a collaborative point of view the

IFCs form the basis of a building description. This

basis is enriched during the building’s lifecycle with

elements related to facilities management: financial

data, maintenance rules, evacuation procedures and

so on. The quantity of information becomes

exponential and then a relevant organization of these

elements becomes very complex. Today, “Building

Information Modeling (BIM)” is promising to be the

facilitator of integration, interoperability and

collaboration in the future of building industry. The

term BIM has been recently pointed to demarcate

the next generation of Information Technologies

(IT) and Computer-Aided Design (CAD) for

buildings.

Vanlande R., Cruz C. and Nicolle C.

ACTIVE3D: SEMANTIC AND MULTIMEDIA MERGING FOR FACILITY MANAGEMENT.

DOI: 10.5220/0002792000210029

In Proceedings of the 6th International Conference on Web Information Systems and Technology (WEBIST 2010), page

ISBN: 978-989-674-025-2

2 BUILDING INFORMATION

MODELING

BIM is the process of generating, storing, managing,

exchanging and sharing building information in an

interoperable and reusable way. A BIM system is a

tool that enables users to integrate and reuse the

information of a building and the domain knowledge

throughout the lifecycle of a building (Lee, 2006). A

BIM system is a central system that manages various

types of information, such as enterprise resource

planning, resource analysis packages, technical

reports, meeting reports, etc. However, the main

feature of a BIM is the 3D modeling system with

data management, data sharing and data exchange

during the lifecycle of the building. As a matter of

fact, a building is composed of geometrical elements

which are the basis of a building’s design.

Furthermore, parametric modeling provides

powerful mechanisms that can automate the

generation of the building information. Especially

those mechanisms, in conjunction with the behavior

of building object and an object-based system,

facilitate the maintenance and the validity of the

building’s designs. Several definitions of BIM can

be found in the specialized literature. The NBIMS

(NBIMS, 2007a) divides BIM categories in three

axes which are Product, Collaborative Process and

Facility. The Product is an intelligent digital

representation of the building. The Collaborative

Process covers business drivers, automated process

capabilities and open information standards used for

information sustainability and fidelity. The Facility

concerns the well understood information

exchanges, workflows, and procedures which are

used by the different teams as a repeatable,

verifiable and sustainable information-based

environment throughout the building’s lifecycle.

According to (NBIMS, 2007b) a BIM is a

computable representation of all the physical and

functional characteristics of a building and it is

related to the project information, which is intended

to be a repository of information for the building

owner/operator to use and maintain throughout the

lifecycle of the building. According to Autodesk

(Autodesk, 2002), BIMs have three main features:

They create and operate on digital databases for

collaboration. They manage change through those

databases so that a change to any part of the

database is coordinated in all other parts. They

capture and preserve information for reuse by adding

industry-specific applications.

By analyzing the BIM definition we index a set of

features common to BIM systems (Tolman, 1999) ,

(NIST, 2007), (Eastman, 2005), (Zamanian, 1999),

(Sable, 2005), (Cruz, 2006). (1) The main feature of

BIM is the ability to store, share and exchange data.

Many methods are used to realize those processes

like files or databases. Concerning data exchange,

BIMs are developed with the aim to keep open non-

proprietary data format exchange. (2) Data managed

in BIM processes concerns building geometries

which are most of the time 3D data. 3D data is more

helpful for designers for the visualization of

complex construction conditions than 2D while it

communicates at the same time design intentions.

AEC industry visualizes the design using

stereoscopic projection tools to create an immersive

experience (Dace, 2007). Spatial relationships

between building elements are managed in a

hierarchical manner. (3) BIMS are data rich and

comprehensive as they cover all physical and

functional characteristics of a building. BIMs are

also rich semantically as they store a high amount of

semantic information about building elements.

Moreover, the data model is fully object oriented to

facilitate data management and process definition.

(4) Some of the BIMs are extensible to cover

unimplemented information domains. For instance,

the development of IFC 2.X went through a major

change in order to extend progressively the range

and the capability of IFCs by using modules. (5)

BIMs play a central role in the building lifecycle. In

order to ease data exchange, a data format has to be

widely used. By definition, BIMs enable

interoperability among diverse applications using a

shared universal information standard. (6) The

lifecycle of the project in AEC is composed of

several phases which have to be validated by the

corresponding AEC engineering designer. BIMs

cover several lifecycle phases. The state of these

phases is processed by BIMs in order to sequence

and schedule the process. BIMs support 4D analysis,

where activities from the project schedule can be

simulated and studied to optimize the sequence of

construction.

Our research aims at solving the problem linked

to the constant IFC evolution (4). The definition of a

complete framework that allows the management of

the knowledge around the building process requires

an extensible and generic formalism to represent

both specific data describing building information

and connected information defined by the user

during the building’s lifecycle. It requires also tools

to handle and query the corresponding modelling

data, and it requires also tools to manage the data

evolution during the building’s lifecycle. Moreover,

the contextual management of data that needs to

WEBIST 2010 - 6th International Conference on Web Information Systems and Technologies

correspond with the user’s view and constraints has

to be taken into account. It requires also an adaptive

graphical 2D/3D representation, dynamically

connected with data from buildings according to the

BIM features. Finally, the most important point is

the fact that the framework has to take into account

the constant evolution of specific data describing

building information and the corresponding

connected information defined by the user during the

building’s lifecycle. We have developed a method

that combines IFC and the various requirements

related to facility management.

3 PRINCIPLE OF THE

APPROACH

In our study context the requirement of the model

extensibility and the model evolution generates

others difficulties, such as mapping data between

two models. Handling information during the

building lifecycle requires a contextual and temporal

representation of knowledge. It is important to trace

each data evolution at a time and to know how to

present data according to the user context (Guha,

1995, 2004). To deal with these requirements, we

derived Names Graph (Carroll, 2005) in order to

complete our framework. Based on the context we

developed a system description and operators in an

architecture called CDMF that allows dealing with

the traceability of the data schema evolution. This

innovative approach allows knowing, at any time of

the project, the current version of the data schema

that defines the facility data.

Our approach considers all requirements at large

(temporal management, adaptive view, 2D/3D

representation) in order to propose a global solution

with a framework based on Semantic Web

technologies. To meet these requirements, we have

built a complete framework, called CDMF, derived

from Semantic Web formalism: RDF (Klyne, 2004),

Named Graph (Carroll, 2005), OWL (

McGuinness,

2004

) and SWRL (Horrocks, 2004). These

formalisms constitute the base of our approach. We

have extracted from each of them the more adapted

features to our problematic. RDF formalism allows

data modelling and can be used by operators

provided by OWL/SWRL. Finally, Named Graph

gives a contextual layer to this unit. To obtain a

complete formalism, well adapted to facilities

management, we have defined a framework called

CDMF which will be presented in the following

section.

4 OVERVIEW

In this section the architecture called CDMF is

presented. This architecture proposes to use

semantic operators in order to manage data in the

context of a facility management environment. The

objective of CDMF is to join together the semantics

of OWL and SWRL in only one formalism. For that

DMF defines a whole system of logical operators

allowing the description of classes, properties,

constraints and of rules. The principal interest of

CDMF is to offer a framework facilitating the

description of contextual data. This framework

offers a single structure that permit us to define a set

of data, all types of contexts and the actions that can

be realized on these data. CDMF aims at meeting the

various needs evoked; moreover, it achieves for the

complete system, due to its structure and its

operators, a reduction in volume of the data that

represents an information system in a collaborative

environment, as well as restricted treatments due to

the unicity of information. Thus, we used the

formalisms of the semantic Web to create an

environment meeting in a single way our various

needs.

The operators of DMF allow the modeling of

knowledge on 3 levels (Table 1): the model level,

where DMF makes it possible to define the concepts

of modeling (class, property, etc). The diagram level

allows the defining the description of knowledge.

The instance level, which makes it possible to define

the objects of the real-world according to the

structure of the diagram defined in the higher level

of abstraction. For each level a set of triplets forms

RDF graphs.

The architecture of CDMF is based on the

structure of modeling RDF. This structure RDF

makes it possible to represent knowledge with

graphs. These graphs are modeled using a set of

triplets. A triplet is composed of a subject, a

predicate, and an object. The architecture of CDMF

is composed of two layers: “DMF” and “C”. The

“DMF” layer is composed of the model construction

operators and the “C” layer is composed of the

context manager operator and the handling graph

operator.

ACTIVE3D: SEMANTIC AND MULTIMEDIA MERGING FOR FACILITY MANAGEMENT

Table 1: The 3 levels of the data modeling of DMF.

Model Schema Instance

dmf:Class type dmf:Class

dmf:Property type rdf:Property

:Building type dmf:Class

:Storey type dmf:Class

:contains type dmf:Property

:b1 type :Building

:e1 type :Floor

:e2 type :Floor

:b1 :contain :s1

:b1 :contain :s2

Fig. 1. presents the composition of this

architecture of the CDMF environment which is

made of the space system stack, the API and the

engine. This environment allows the creation of

specific applications that permit to deal with facility

management requirements.

Figure 1: This is the architecture of the CDMF

environment which is made of a space system stack, an

API and an engine. RDF is used to define the data

modeling.

1. The Space System is used to configure the

system and to allow data access. It is based on an

RDF document. This space system contains a set

of graphs called SystemGraph. From this point,

the CDMF engine checks the declared graphs

and responds to queries executed from the API.

2. The API CDMF is a set of methods used to

handle the data system. This API proposes to

access data with two main classes called

SystemSetGraph and SystemGraph. The first one

allows to access the system graph which

composes the space system. The second one is

made of methods that permit to modify system

data. For instance, the method

SystemGraph.create() provides the list of Class

elements and Property elements which can be

created.

3. The CDMF engine is the kernel of the

architecture. This engine uses a space system to

configure and to know the set of systems to use.

The engine contains processes which manage

methods of the API CDMF. The engine selects

the system graphs in the space system, and

creates data, deletes data, etc. This engine is

presented in Fig. 2.

Figure 2: The CDMF engine is composed of two parts:

The DMF stack which defines modeling operators and the

C stack which defines context definition and graph

manipulation operators.

4.1 DMF: A Reduced Set of Modelling

Operators

This section presents the DMF stack. This stack is

made of operators which allow to model information

(from simple and monovalued attributes to complex

3D objects) into semantic graphs. This section

enables us to show that the formalism that we have

defined has a restricted set of operators. We show

that these operators can be combined to meet all the

needs for semantic modeling defined in the

statements. For each operator we give its equivalent

in SWRL or OWL.

• “dmf:Class” defines a class. The equivalent

operator in OWL is “owl:Class”.

• “dmf:Property” carries out the definition of a

property of a Class.

WEBIST 2010 - 6th International Conference on Web Information Systems and Technologies

• “dmf:Equal” defines the equality between two

variables. This operator makes it possible to test

if two resources are equivalent.

• “dmf:Var” makes it possible to define variables

used in the logical formulas. Its equivalent is

defined in SWRL by the operator

“swrl:Variable”.

• “dmf:Pred1” makes it possible to define unary

predicates. Its equivalent is defined in SWRL by

“swrl:ClassAtom”.

• “dmf:Pred2” makes it possible to define binary

predicates. The equivalent operators in SWRL

are “swrl:IndividualPropertyAtom” and

“swrl:DatavaluedPropertyAtom”. A binary

predicate is a property with a subject and an

object. To make the correspondence with RDF,

the terms of subject and object are used in order

to define the first and last element of a triplet

RDF.

• “dmf:Equiv” makes it possible to define that

two classes are equivalent. The set of the

elements of the type A is equivalent to the set of

the elements of the type B. The equivalent

operator in OWL is “owl:equivalentClass”.

• etc .

4.2 Context and Mapping Operators

This section presents the operators defined in the

stack C. These operators are used to handle graphs

and to define contexts. With these operators, new

graphs can be generated by combination of existing

graphs. These operators are commonly used to

update the data model definition when a norm in

architecture is upgraded. For example, the IFC norm

has been updated six times since 2000. Moreover,

These operators allow the semantics handling of the

3D objects contained in the graphs. The results of

this handling are directly visible in the 3D scene

which is dynamically updated. The elements defined

in this part use the space of name cdmf. For each

type of graph we present its definition by using

DMF operators.

4.2.1 Union Operator

The result of the addition of G1 and G2 is the union

of the set of the triplets of G1 and the set of the

triplets of G2. The operator of a union of graphs is

defined by the class “cdmf:AddGraph”. It has a

property “cdmf:args”. This property is a list of RDF

elements (“rdf:Bag”) whose elements are graphs.

The definition of these elements allows the union of

two or several graphs.

4.2.2 Intersection Operators

The intersection operator can be defined in different

manners and can imply a different result according

to the type of intersection carried out. The

intersection operator is defined by two elements

“cdmf:InterGraph” and “cdmf:CompInterGraph”.

The first element defines a “traditional” intersection.

The second element makes it possible to specify on

which elements of a triplet the intersection is carried

out.

Traditional Intersection. The result of the

intersection between G1 and G2 is the set of the

identical triplets in G1 and G2. This operator has

two properties “cdmf:arg1” and “cdmf:arg2”. These

two properties are “CdmfGraph” types representing

the two graphs on which the intersection must be

computed.

Composed Intersection. The composed intersection

makes it possible to determine which part of the

triplet is concerned in the calculation of the

intersection. In the case of the “traditional”

intersection, one carries out the intersection on the

set of triplets of each graph. Here we can compose

the intersection with the various parts of a triplet

(subject, object). Below you will find possible

combinations of intersections.

• The intersection on the subject in the two

graphs. The result of the intersection between

G1 and G2 is the set of the triplets whose

subjects are identical in G1 and G2.

• The intersection on the object in the two graphs.

The result of the intersection between G1 and

G2 is the set of the triplets whose objects are

identical in G1 and G2.

• The intersection on the subject of the triplets of

a graph with the object of the triplets of the

other graph. The result of the intersection

between G1 and G2 is the set of the triplets

whose subjects of the graph G1 are identical to

the objects of the graph G2.

• The intersection on the subject or the object.

There is a last combination which is actually the

addition of two intersections. The result of the

intersection on the subject of G1 and the subject

or the object of G2 is equivalent to the sum of

the intersections on the subject of G1 and G2,

and on the subject of G1 and the object of G2.

The intersection operator

“cdmf:CompInterGraph” has two properties

“cdmf:arg1” and “cdmf:arg2” which are the two

ACTIVE3D: SEMANTIC AND MULTIMEDIA MERGING FOR FACILITY MANAGEMENT

graphs on which the intersection is carried out. It has

two additional properties “cdmf:on1” and

“cdmf:on2” respectively defining the two parts of

the triplets used to carry out the calculation of an

intersection.

4.2.3 Difference Operator

The difference between two graphs is indicated by

the element “cdmf:RemoveGraph”. The result of the

difference between G1 and G2 is the suppression of

the set of the triplets of G2 in The class

“cdmf:RemoveGraph” has two properties

“cdmf:src” and “cdmf:rem”. The second property

constitutes the set of the triplets to be withdrawn

from the graph indicated by the first argument.

4.2.4 Mapping Operator

The last type of operation on the graphs is the

operation of mapping described by the element

“cdmf:MapGraph”. A graph of mapping is a

transformation of a graph into another graph using

mapping rules;

The mapping operator has two properties

“cdmf:src” and “cdmf:map” indicating the source

graph and the target graph. The result of the

operation of mapping is the set of the triplets which

is defined by the rules of “Gmap”. A rule in Gmap is

described by an operator of implication.

In this part we have studied five operators which

allow carrying out various combinations of graphs.

These five operators are the union, the difference,

the intersection (traditional and composed) and the

mapping. They constitute the first part of the C

stack. The second part of the C operators is the

definition of a particular graph “SystemGraph”. This

element associates various types of information with

a graph. This element is used to represent contexts.

4.3 Context Modelling, the Element

SystemGraph

The element “cdmf:SystemGraph” uses the

mechanism of Named Graphs to define the contexts

with the help of the properties. The “SystemGraph”

element associates with the graphs presented above

all useful information which is needed to respond to

the set of problems met in facility management. This

element defines the nature of the graph, on which a

graph data model is based. The element also defines

the context of use and actions that can be realized on

this graph.

“SystemGraph” evokes the data model on which

the associated graph is built. For instance, the

definition of building X is based on a data model

that describes the composition of a building. The

data model allows to check the data coherence of the

associated graph and allows to indicate which kind

of data can be generated in the graph. This data

model is defined with the help of operators

introduced in DMF. In fact, the “SystemGraph”

defines actions that can be undertaken on the graph

such as reading, writing or deleting. This can be

done according to the actions which are authorized

on the associated graph. The description of the

context in “SystemGraph” is a list of RDF

resources. This section presents the definition of

“SystemGraph” with its properties: “cdmf:model”,

“cdmf:of”, “cdmf:action”, “cdmf:graph”. The name

space cdmf is used to present these elements. An

example is given in Fig. 3.

4.3.1 The Property “cdmf:model”

Figure 3: An initial “SpaceSystem” used at the beginning

of a facility management project to define the data model

that will support a building definition. There can be found

the properties “cdmf:model”, ”cdmf:of”, “cdmf:action”,

“cdmf:graph”.

This property defines the model on which a

“SystemGraph” element is based. The associated

model will be used to define the objects and the

properties which can be generated in the

“SystemGraph” graph. Subsequently, it is possible

to check the data coherency by comparing it to the

model. “SystemGraph” has a model which is also a

WEBIST 2010 - 6th International Conference on Web Information Systems and Technologies

“SystemGraph”. CDMF defines a class

“cdmf:SystemModelGraph” to represent a specific

“SystemGraph”. This type of “SystemGraph”

contains definitions of classes, properties and rules

defined in the syntax DMF.

4.3.2 The Property “cdmf:of”

The property “cdmf:of” defines the subject of the

“SystemGraph” element. This property defines the

context. It associates a set of RDF resources which

resume what is described by the SystemGraph. For

instance, “SystemGraph” can be the description of a

data model in the building field. “SystemGraph” can

be a data model in a certain version or

“SystemGraph” can represent data on a certain date

and in a certain language, for a certain user. It can

also define the nature of the graph and the conditions

that have to be fulfilled in order to be able to access

a graph system.

4.3.3 The Property “cdmf:action”

The property “cdmf:action” determines the actions

authorized on the graph. It defines the actions of

writing, suppression and modification. If no action is

associated to the system, this implies that only the

visualization of information is possible.

4.3.4 The Element “cdmf:Action”

The element “cdmf:Action” determines which

actions are possible, on which part of the data and

starting from which model.

An action has one or two properties. If it has only

one property add then the addition is allowed. If it

has only one property remove then the system

allows the deletion of data. If it has the two

properties, we can add and remove data in the graph

system. An element “cdmf:Add” defines which

information we can add (“cdmf:model”) and where

it has to be added (“cdmf:addIn”)

4.3.5 The Element “cdmf:Remove”

An element “cdmf:Remove” indicates the

suppressible data which have to be removed. If it

does not have this property, all the data of the graph

of the system can be deleted. The property

“cdmf:from” binds an element of the “cdmf:From”

type. According to the origin of the suppression

(“cdmf:graph”), this element defines the action to be

realized: either an addition in a

“cdmf:RemoveGraph”, or a suppression in the graph

“cdmf:graph”.

4.3.6 The Property “cdmf:graph”

The property “cdmf:graph” contains the associated

graph representing the data. The associated graph is

a “Cdmf:Graph” type. “SystemGraph” has an

attribute of the “Cdmf:Graph” type. Thus,

“SystemGraph can refer to all the types of graphs

presented in CDMF.

5 ACTIVE3D FACILITY SERVER

This section presents the Active3D Facility Server, a

web collaborative platform dedicated to the facility

management, taking into account all aspects of the

building’s lifecycle. Due to the lack of space, we

will illustrate only our proposal with two examples

of use. The first example concerns the initialization

of a space system when a facility manager needs to

configure the platform. This extension is realized by

defining a specific model. This model will be used

in the building definition process. The second point

illustrates the use of context to display specific

information to users.

5.1 Configuration of a SpaceSystem

In facility management, various versions of the

building can be managed and presented to different

actors in many countries. The representation of a

building mixes textual and graphic representations.

The first step in facility management consists in

creating the data model. A building description will

be generated starting from this model. The new data

model is created from a new applicative

environment. An initial space system is created.

Following this step, the facility manager can store

his data in the graph and has the possibility to create

data starting from “SystemGraph” ’Space System’.

The creation of “SystemGraph” includes a model, a

context (list of resources) and the graph of data. For

the model, the facility manager has to choose among

the “SystemModelGraphs” available. A

“SytemModelGraph” element is a “SystemGraph”

whose characteristic it is to contain models in its

associated graph. A SystemModelGraph represents

only a DMF model. For the context, the list of

resources contains only one resource where the

representation is ‘Building Model’.

For the associated graph, the facility manager

creates a new RDF graph which will contain the

definition of the model. For AEC projects, we have

based our model definition on the IFC 2X3. This

ACTIVE3D: SEMANTIC AND MULTIMEDIA MERGING FOR FACILITY MANAGEMENT

model contains approximately 600 classes (IAI,

2007), (IFC, 2007). The following snapshot presents

a part of the IFC model in the application. This

model is created with the tools proposed by the

application (creation of classes and properties).

Fig. 4 presents a snapshot of an IFC building.

Each IFC object is represented in the Building

Model by an operator “dmf:Class”. IFC links are

represented by an operator “dmf:Property”. From

this “SystemGraph”, we can declare classes,

properties, rules, etc. In this example, we have

created simply three classes which are “Building,

Floor” and “Space” and a property “contain”.

Figure 4: An example of a 3D view of a building object in

a Facility Management view using DMF operators.

5.2 Context Representation

From the ”SystemGraph” element, the facility

manager defines a view on data according to a

specific context. This context can be linked to a

specific step in the building’s lifecycle or it can be

linked to a specific type of user (for example

plumbers, architects or structure engineers).

In the Active3D collaborative platform, this context

is used to build user models that define data,

operators and interface for a specific user. Thus,

during the identification protocol, when a user tries

to connect himself to the platform, a specific graph

is built and a view of a building is built according to

its context. Figures 5 and 6 show two different views

of the same building according to the Architect view

and the Structure Engineering view respectively.

6 CONCLUSIONS

In this paper we have presented a Semantic Web

approach for facility management. This approach

Figure 5: Architectural view of a building. In this snapshot

it is required to display the walls for the architect.

Figure 6: A structural view of a building. This view is

required for structure engineers. The corresponding graph

provides all elements needed to make structure

calculations.

allows facility managers to support the building’s

lifecycle management from the design to the

destruction of the building in a collaborative context.

Several actors provide and handle building

information. This approach is based on a semantic

model called CDMF and the IFC 2x3 standard

which defines the 3D geometries of the objects of

building. CDMF improves data management during

the lifecycle of a building. Our proposition, based on

graph combinations and the contextual element

“SystemGraph”, addresses the problem of model

evolution, of data mapping, of the management of

temporal data, and of the adaptation of data

according to the use and the user. Our framework

facilitates data maintenance (data migration, model

evolution) during the building lifecycle and reduces

the volume of data.

A collaborative Internet platform was developed

to support the building’s lifecycle. This platform is

mainly used to federate all the actions realized on a

building during its lifecycle, to merge all

information related to these actions in an adaptive

hypermedia graph, to extract some trade views of the

building by combining information collected during

WEBIST 2010 - 6th International Conference on Web Information Systems and Technologies

the lifecycle from heterogeneous sources and to

handle all these views through a dynamic and

adaptive 3D interface. Currently, the Active3D

platform supports more than 100 specific building

information systems where more than 400 actors

from all civil engineering domains collaborate at

each step of the building’s lifecycle.

REFERENCES

Autodesk White Paper: Building Information Modelling,

(Available online at: http://images.autodesk.com/

apac_sapac_main/files/4525081_BIM_WP_Rev5.pdf),

2002

Barrett P., Baldry D., 2003. Facilities Management,

Towards Best Practice, Blackwell Publishing, ISBN

0632064455

Bizer C., Cyganiak R., 2005, NG4J Named Graph API for

Jena, http://sites.wiwiss.fu-berlin.de/suhl/bizer/ng4j/

Carroll J.J., Bizer C., Hayes P., Stickler P., 2005. Named

Graph, Provenance and Trust, http://www2005.org/

cdrom/docs/p613.pdf

Cruz C., Nicolle C., 2006. Active3D: Vector of

Collaboration, Between Sharing and Data Exchange,

INFOCOMP, Jounal of Computer Science, 5 (3), pp.

1-8, January, 2006

Dace A. Campbell., 2007. Building Information Modeling:

The Web3D Application for AEC, ACM Web3D,

Perugia, Italy, (2007)

Eastman C., Wang F., You S. F., Yang D., 2005.

Deployment of an AEC industry sector product model,

Computer Aided Design, Vol.37, No.12, pp. 1214-

1228, 2005

Guha R.V. and Fikes R., 2004, Context for the semantic

web. In Proceedings of ISWC’2004

Guha R.V.: Contexts, 1995, A Formalization and Some

Applications. PhD thesis

Horrocks I., Patel-Schneider P.F., Boley H., Tabet S.,

Grosof B., Dean M., 2004. SWRL: A Semantic Web

rule Language: Combining OWL and RuleML,

http://www.w3.org/Submission/SWRL/

IAI, http://www.iai-international.org, 2007

IFC 2x3 Workshop, Boston, USA, March 19, 2007,

http://127.0.0.1:4664/redir?url=http%3A%2F%2Fbuil

dingsmart%2Ecom%2Eau%2F&src=1&schema=2&s=

suabHoX9yEfSmitJzdEaD6NuyKQ

Keith A., Atkin B., Bröchnet J., Haugen T., 2004.

Facilities Management, Innovation and Performance,

Taylor and Francis Edition, ISBN 0415321468, 2004

Klein M., 2002. Interpreting XML via an RDF schema, In

ECAI workshop on Semantic Authoring, Annotation

& Knowledge Markup, Lyon, France

Klyne G., Carroll J.J., 2004. Resource description

Framework (RDF): Concepts and Abstract Syntax,

http://www.w3.org/TR/rdf-concepts/

Lee G., Sacks R. and Eastman Charles M., 2006.

Specifying parametric building object behavior (BOB)

for a building information modeling system,

Automation in Construction, Elsevier, Volume 15,

Issue 6, November, Pages 758-776, Knowledge

Enabled Information System Applications in

Construction, 2006

McGuinness D.L., Van Harmelen F., 2004. OWL

Ontology Language, Overview http://www.w3.org/

TR/owl-features/

NBIMS: National BIM Standard Purpose, US National

Institute of Building Sciences Facilities Information

Council, BIM Committee, (Available online at:

http://www.nibs.org/BIM/ NBIMS_Purpose.pdf),

52006), 2007b

NBIMS: National Building Information Modeling Stan-

dard Part-1: Overview, Principles and Methodologies,

US National Institute of Building Sciences Facilities

Information Council, BIM Committee, (Available

online at:

http://www.facilityinformationcouncil.org/bim/publi-

cations.php), 2007a

NIST CIS2 Web Site, Web Site, (Available online at:

http://cic.nist.gov/vrml/ cis2.html), 2007

SABLE Web Site, Web Site, (Available online at:

http://www.blis-project.org/~sable/), 2005

Tolman F., 1999. Product modelling standards for the

building and construction industry: past, present and

future, Automation in Construction, Vol.8, No.3, pp.

227-235, 1999

Zamanian K. M., Pittman J. H., 1999, A software industry

perspective on AEC information models for distributed

collaboration, Automation in Construction, Vol.8,

No.3, pp. 237-248, 1999

ACTIVE3D: SEMANTIC AND MULTIMEDIA MERGING FOR FACILITY MANAGEMENT