Semantic Web Technology for Building Information Model

Muhammad Asfand-e-yar, Adam Ku

cera and Tomáš Pitner

Lab of Software Architecture and Information Systems (LaSArIS)

Facuilty of Informatics, Masaryk University, Botanická 68a, Ponava, Brno, Czech Republic

Keywords:

Building Information Model, Facility Management, Semantic Web.

Abstract:

Smart building is a trend towards the Buildings Automation paradigm. Smart building aims to autonomously

control devices and systems in given environment. These devices and systems are nevertheless supervised by

facility management. The facility management normally is aided by heterogeneous systems and applications.

Due to multifarious data of the systems, applications, and missing semantics in building automation, the data

is manually handed by facility management, for analysis and decision making. Therefore, such a system

is required to integrate such multi-form data of various systems and applications. Hence, Semantic Web

technology is proposed in this paper for the data integration. The paper explains development of Semantic Web

Model for BIM systems, used at Masaryk University. The model links various systems, which are used for

facility management, and is applicable in different environments. The model not only provide base for analysis

and decision making for facility management, but also facilitate developers to focus on front-end of application.

Hence, the aim is to structure the data for simplifying the queering mechanism used for analysis.

1 INTRODUCTION

Necessity of each organization is to ensure various

aspects of its operation that are not directly involved

in primary goal, i.e. providing service to customers

or selling products. Facility management (FM) covers

the aspects, such as space management, help desk &

service desk, maintenance or energy monitoring. In-

ternational Facility Management Association (IFMA)

deﬁnes FM as a profession that encompasses multi-

ple disciplines to ensure functionality of the built en-

vironment by integrating people, place, process and

technology.

FM distinguishes several systems and data sources

that support and simplify tasks of FM. Widely used

Computer Aided Facility Management (CAFM) sys-

tems cover most areas of FM. CAFM software serves

as repository and user interface for operational data,

for example assigning employees to rooms, log of

maintenance plans, requests & tasks, energy consump-

tion data, etc. CAFM software is used for analyzing

and evaluating a building performance in perspective

of FM (Paul et al., 2012). CAFM systems offer ad-

vanced analytical tools for evaluating efﬁciency and

performance of operation based on economic (energy

consumption), spatial (occupancy planning) and tech-

nical data (maintenance). The Building Information

Model (BIM) is a data source that contains spatial

information about building constructions (materials,

dimensions), locations (sites, buildings ﬂoors, rooms)

and devices installed in them (valves, pumps, plumb-

ing, lights, power lines, etc). Data from the BIM

database serve as an input for CAFM systems. Spa-

tial data are imported into the CAFM system as a

“background data” for space management, occupancy

planning, maintenance management and other tasks.

Finally, the task of FM is tightly connected to mod-

ern “intelligent buildings”. These facilities incorporate

wide scale of automated systems for example secu-

rity system, access control system, ﬁre alarm system

or building automation system that controls Heating,

Ventilation, Air Conditioning (HVAC) devices. The

system consists of various sensors and controllable de-

vices. Building Management System (BMS) facilitates

remote monitoring and controlling the building opera-

tions. BMS also provides additional services such as

archiving historical data and event notiﬁcation.

Currently the integration of BMS data with CAFM

and BIM is simpliﬁed to a simple structure that cannot

be effectively queried because the integration part is

completely missing. The integration is impossible be-

cause BMS data structure is determined by the network

topology, not by the semantic structure. The semantic

structure is required because the advanced analytical

109

Asfand-e-yar M., Ku

cera A. and Pitner T..

Semantic Web Technology for Building Information Model.

DOI: 10.5220/0004999201090116

In Proceedings of the 9th International Conference on Software Engineering and Applications (ICSOFT-EA-2014), pages 109-116

ISBN: 978-989-758-036-9

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

features of CAFM software are currently unavailable

for BMS data. This does not affect the small installa-

tions, where data retrieval and analysis can be easily

performed manually. However, for large sites (hun-

dreds of devices, thousands of sensors), the amount

of data prevents effective gathering of required infor-

mation. Despite of this, BMS contains large amount

of accurate, up-to-date and detailed data which are

valuable for building operation analysis that cannot be

obtained by any other way.

BMS has two types of users - one group is facility

managers who know which data they need for analysis,

economic context, but are unable to get the data from

BMS. The other group is building operators who have

capabilities to gather the data from BMS. They don’t

have enough knowledge, competence or authority to

fully evaluate the building operation to make long-term

decisions based on the results.

The work ﬂow for BMS data analysis is complex

for ﬂexible data analysis. With currently available sys-

tems, when the building operators are requested to get

the required data in provided format, they gather the ad-

dresses of respective data points according to request.

Then, the data is extracted subsequently to required

format. This process is not automated and repeated ev-

ery time, when the report is required. Therefore, a user

interface is required, that will help facility managers to

gather data from various databases accordingly. Using

the architecture and Semantic Web approach, front end

applications will hide low-end aspects of data gather-

ing and allow facility managers to query BMS directly.

Then, a facility manager should be able to query

the system, for example:

• Show me which rooms on the second ﬂoor of

“A11” building had active AC units during last 8

weekends.

• I want to receive a report after 24 hours, about

electricity consumption in 5 minute intervals of

speciﬁed 4 buildings during this night.

• I want to know which devices inﬂuence

temperature in the ofﬁce of Mr./Mrs. XY.

This paper proposes possible solutions to the problem

by deﬁning systems architecture and introducing on-

tology that facilitates BMS data queering for analysis.

Therefore, a platform is developed to integrate the data.

The paper explains related work in section 2 and gen-

eral architecture of overall FM integration platform is

explained in section 3. Overview of Semantic based

Smart Building and Ontological Model is explained in

detail in section 4. Section 5 and 6 explains use case

scenarios and provided solutions to scenarios using

Ontology model, respectively.

2 RELATED WORK

Information technology plays an important role in

intelligent building, as an increasingly sophisticated

demand (Andrew et al., 1998; Kroner, 1997), from

decades for comfort living and requirement for in-

creased occupant control. Indeed, much of the work

in regard to building automated systems was done, but

still communication is lacking, between several het-

erogeneous data for analysis. Recently, some of the

work to analyze building data is published. The work

mainly focus on the concepts and relations between

various entities of the building taxonomy. Much of

the research focus on intelligent building technology

development, performance evaluation and investment

evaluation techniques (Wong et al., 2005).

Various devices communicate and interact, with-

out direct human intervention. Coordination between

devices act as supervisors, these devices are devoted

to manage available resources to meet deﬁned require-

ments. Building management and automation systems

are still far from this vision (Michele et al., 2014).

Scenarios are deﬁned during implementation but no

dynamic changes are occurred. Currently, automatic

information management systems (i.e. performance

monitoring system and systems for interacting end-

users, devices, and services in a deﬁned environment)

are quite limited.

Ontology or taxonomy engineering is a primary

concern for deﬁning concepts and relation between

them. Therefore, main entities of building, accord-

ing to requirements, are used as concepts to design

ontology model. Hence, relations between concepts

facilitate reasoning, which ultimately contributes in

analysis. For designing self conﬁguration and self

management system, Ambient Intelligent (AmI) system

(De Paola, 2014) is an example, which uses ontology

for interacting with given environment and exploiting

knowledge for cognitive processes and autonomously

managing its own functions. Likewise Wireless Sensor

Network (WSN) is also used in Open Framework Mid-

dleware (Rob et al., 2009) for management in smart

buildings. Open Framework Middleware diagnosis

faults in sensor networks. Therefore rule base knowl-

edge management model is designed. This model

facilitates FM applications, such as in energy moni-

toring, security, water ﬂow control, etc. Additionally,

Home and Building Automation (HBA) (Michele et al.,

2014) is another ﬂexible multi-agent system. This

system applies knowledge base representation and au-

tomated reasoning for resource discovery in building

automation. Analysis, based on the reasoning, is used

to reduce cost of energy and monitor performance of

various applications.

ICSOFT-EA2014-9thInternationalConferenceonSoftwareEngineeringandApplications

110

In smart building automation, wireless pervasive

computing is introduced to enhance life comfort, and

importantly reduce maintenance and consumption cost.

The smart building automation integrates mobile tech-

nology to facilitate maintenance, which deals with

monitoring and life safety plans in case of emergency.

An ontological model is proposed (Dekdouk, 2013)

to switch-off lights when no one is in room, schedul-

ing water valves & pumps accordingly and switching

to photovoltaic installation if bright shining sun rises.

Besides this, an approach for embedded systems of

sensors is used to detect activities of visitors and occu-

pants (Paul et al., 2009), while interacting with smart

building. The focus is to support FM tasks, such as

building management, maintenance, inspection and

emergency response. Therefore, ontology model is

used for analysis to match currently received data with

the data subscribed by FM applications.

Industry Factory Classes (IFC) are extensively used

in construction of BIM in smart buildings. These

classes are extracted from IFC and imported into a

Semantic Web Model, where the requirements are ana-

lyzed by facility manger according to feasibility of a

building construction, using SPARQL. Similarly, the

Semantic Model is also used to view the 3D build-

ing models to visualize data. Therefore, an approach

(Nicolle and Cruz, 2011), based on both semantic ar-

chitecture (named as CDMF) and IFC 2x3 is used

for 3D geometries of a building. The approach, i.e.

SystemGraph, facilitates data maintenance during the

building life cycle. In project DRUM/PRE (Seppo,

2013), IFC classes are used for data maintenance &

connections, and are linked through Semantic Web

Technology to allow required queries. The IFC classes

are also used to deﬁne policies for Energy-Efﬁcient

smart buildings, i.e. in Think Home project (Mario and

Wolfgang, 2010). Therefore, Green Building XML

schema is used with IFC to construct Semantic Web

Model. Various policies are deﬁned in Think Home

project, in form of rules and restrictions. These poli-

cies facilitates facility managers in querying thermal

measured units and thermal material properties for the

construction of a building model. The explained po-

lices in the project are similarly used in (Massimiliano

and Giuseppe, 2013).

Another aspect of smart building systems is to re-

duce information load on end-users (i.e. building oper-

ators). Aim of Smart Home and Social Services (Yulia

et al., 2013) is to construct user friendly system and

ﬁlter out irrelevant data according to end-user require-

ments. Therefore, for future decision making in smart

buildings, the communication between end-users and

heterogeneous systems is important. Hence, a com-

prehensive communication among systems is provided

(Christian et al., 2008). This system enhances pos-

sibilities in making analysis and decision on stored

knowledge.

The approach explained in this paper is used for

BMS in MU, which covers several aspects of opera-

tional analysis in smart building. Main emphasis of

the BIM Ontology Model at MU is to communicate

various heterogeneous systems. BIM at MU deals with

location and device information in a building. Knowl-

edge domain of the Model facilitates in monitoring

& automatic controlling various devices, provide in-

stant response to concerned end-users, and reasoning

& analysis for future decision making. The Model will

reduce cost and time consumption during analysis and

future decision making.

3 DATA OPERATION ANALYSIS:

GENERAL ARCHITECTURE

Various data sources are related to FM and building

automation. There are three basic categories of data

sources i.e. CAFM, BIM and BMS. The proposed

architecture, as shown in ﬁgure 1, supports seamless

integration among these systems. The architecture will

support development of analytical tools. These ana-

lytical tools will be used for BMS, as used in CAFM

systems. Therefore, the architecture is essential for

efﬁcient large scale building operation analysis. The

architecture facilitates the ontology repository that pro-

vides meaning to BMS data. The ontology reposi-

tory helps analytical and monitoring developers, to

focus on front-end user interface and analytical fea-

tures. Facility managers will be able to do analysis,

using Ontology repository, without requiring data from

building operators. This will solve the gap for facility

managers between economic knowledge and technical

BMS data.

The architecture introduces additional elements

to CAFM, BMS and BIM systems. These elements

include Technology data mart, Business Intelligence

applications, Complex Event Processing Engines and

Ontology repository. Technology data mart is OLAP

data source that provides operational data for further

analysis. The operational data is received from BMS

OLTP archive server. Ontology repository enables

the applications to explore relations between elements,

which are stored in different systems. These relations

between elements will link BMS data point with the

source device in the BIM. The architecture will be used

either to develop new analytical applications or to add

new functionality to existing systems, e.g. CAFM.

Business Intelligent Applications and Complex Event

Processing Engines are the examples of new analytical

SemanticWebTechnologyforBuildingInformationModel

111

Figure 1: BMS at Masaryk University.

applications. The Ontology repository describes rela-

tions between BMS and BIM. The BIM is integrated

with CAFM in current environment. Integration be-

tween BMS and CAFM is achieved by transitive rela-

tions between the three systems.

The proposal uses the complete FM environment

of Masaryk University (MU). The MU uses all the core

systems i.e. BIM, BMS and CAFM. These three sys-

tems are distributed over 200 buildings at 40 distinct

sites in Brno, Czech Republic. The BMS connects

40 buildings; all together contain 1000 devices and

hundreds of thousands of data points. BACnet com-

munication protocol is used for integrating devices in

BMS. These devices are of various vendors and are

used for different functions. Those buildings which

are integrated in BMS also has BIM database. The

BIM database is an in-house developed database. The

BIM uses GIS based on ESRI ArcGIS server.

4 SEMANTIC BASED SMART

BUILDING

While analyzing building operation (e.g. BIM, BMS

and CAFM), several concepts are gathered from differ-

ent systems. Table 1, provides overview of semantics

concepts required for BMS, which are organized by

their source systems. For example, temperature in a

particular room is a Measured Variable, in Environ-

ment column, and is explained as;

• Meaning (physical quantity - room temperature)

• Source (data from BIM database – Location

Information and Device Information)

• Available data (BMS network addresses for

real-time data, historic data and event triggers)

• Relations ( which variable is inﬂuenced by &

what is inﬂuenced by a variable)

Detailed description of BIM is explained below. The

Ontology Model is based on practical experience and

requirements required for the MU’s BMS systems.

The Model is generalized on the abstract concepts that

are common for each of the building’s operation, mon-

itoring and FM systems. Currently, development of

Ontology is in progress, therefore concepts describing

BIM domain and its integration is discussed here.

Location Information in BIM

– Location Infor-

mation is stored in spatial database named as “building

passport”. Location is described by its location code.

These location codes serve as primary keys in spatial

database. Usually, room is represented as a location in

a building. In spatial database location code is a string

deﬁning location data as Site Code, Building Num-

ber, Floor and Room information; ﬁgure 2 elaborates

building passport.

Device Information in BIM

–Device Information

is stored in spatial database named as “technology pass-

port”. In spatial database Device Information describes

location of the device, its purpose and its connection

to a particular system in building. For example the sys-

tems could be Building automation system, security

system, CCTV, water supply, power lines, etc. In spa-

tial database, technology code is a string consisting of

System, Sub-System, Device type and Device Index;

as described in ﬁgure 2. The Device Index is used to

distinguish similar devices in a room. The “building

passport (BP)” is integrated with “technology pass-

port (TP)” to deﬁne a complete code for a device, its

connections and its location in a building.

Figure 2: BP and TP (Unique ID of BIM).

4.1 Semantic Web and Complex Data

The data is analyzed for achieving the aim, to design

Ontological Model, for complex data of BIM used at

MU. As explained before, similar Device Types could

have same or different Index Numbers, which could

be connected with same System and Sub-System. Sim-

ilar data for similar device types postulate the idea to

ICSOFT-EA2014-9thInternationalConferenceonSoftwareEngineeringandApplications

112

Table 1: Elements of Building Operation Semantics.

Environment BMS BIM

Measured Variable Device Location Information

Object (Data Point) Device Information

Object Purpose List

Physical Quantity List

select common System, Sub-System, Device Types

and Index Number. The case of complete room data,

also predict that common information from room data

should be selected, which is categorized according to

Site Code, Building Number, Floor and Room infor-

mation.

Categorizing common data facilitates to simplify

the complex data. This complex data is managed to

distinguish between various devices and rooms in dif-

ferent buildings. The common data provides a com-

plete view of various devices used in all buildings and

also contributes in grouping variety of Systems and

Sub-Systems, for analysis.

4.2 Main Concepts and Relationships

Common data is extracted from actual data of room

and devices used in BIM system. The common data

is then analyzed to construct concepts for Ontological

Model. Based on concepts of BIM system, it is de-

cided to keep building taxonomy, i.e. BP, as domain

knowledge of the Ontological Model. Therefore, build-

ing taxonomy includes Site Code, Building Number

and information of Floor and Room is kept as concepts

under the domain knowledge.

Common data of devices are named as TP. An in-

stance of TP is a self explanatory entity that informs

building operators about Device Type, its connection

with a System and Sub-System. Therefore, instance

of TP is considered as separate concept in Ontological

Model, described as taxonomy of device. The taxon-

omy of device further explains sub-concepts, which

includes System, Sub-System, Device Type and Index

Number.

The challenging step of BIM system is to deal with

repetitive index numbers of device instances in TP

and repetitive room number at each ﬂoor in a building,

used in BP. Initially, it was decided to make relation-

ship between Room and Floor information with Index

Numbers and Device Type. But after analyzing and

considering the actual data, Ontological Model pro-

vided results that don’t exist in actual data. In second

step, relationships were taken into consideration to

connect the instances of TP and BP tightly. Most of

non existing data issues were solved but the problem

come-up with repetitive BIM data. The same data

which connects TP with BP was displayed several

times. Finally, it was decided to keep the TP data

uniquely and then connect with BP data. Therefore,

a new concept was introduced named as “Identiﬁer”,

which identify each of the instances of devices that are

installed in a room, as shown in ﬁgure 3.

Analyzing common data and deﬁning concepts pro-

vides an overview of Ontological Model, as shown in

ﬁgure 3. These concepts, i.e. classes and sub-classes,

facilitates in populating the Ontological Model. Hence,

the common data extracted from actual data of BIM

system is used to populate the Ontology. The concept,

i.e. “Identiﬁer”, is populated according to total number

of devices installed in a building.

Relationships between the concepts are deﬁned

according to logical association of entities in BIM.

These relationships, i.e. predicates, are conceptual

relationships that are used by building operators at

MU. Consequently, deﬁnition of relationship is keenly

considered in reference to technical aspect of BIM.

Figure 3: BMS root Ontological Model.

4.2.1 Second Level of Ontological Model

Deﬁning concepts and relationships in Ontological

Model, facilitates in improving ﬁnal results accord-

ing to requirements. Major issues related to repetitive

results and non existing actual data are solved for a

SemanticWebTechnologyforBuildingInformationModel

113

building in a site. After populating the Ontology for

a second building in same site, again generates repet-

itive results, this is because of similar alpha-numeric

numbers assigned to ﬂoors in different buildings.

For avoiding recursive results, due to ﬂoor numbers

in different buildings, the Ontology Model is extended

to second level. The common data from building tax-

onomy and devices taxonomy is populated at root level

of Ontology, but the relations between instances are

deﬁned at second level. Therefore, for each building

at MU, an extended Ontology is used, for keeping

the uniqueness of information. Hierarchical level of

the Ontology Model is depicted in ﬁgure 4. The hi-

erarchy is used for the Ontology Model, depends on

analysis of relevant concepts in terms of entities and

linking data of BIM (i.e. building and device tax-

onomy) (Zarrad et al., 2012). In taxonomic relations,

links are established on canonical structure of concepts

and lexico-syntactic patterns (Carmen and Desislava,

2011) are used to construct BIM unique ID. These

similar relations are also used in (Asfand-e yar et al.,

2009).

Figure 4: BMS Second Level Ontology Model.

5 SCENARIO

The use cases elaborate requirements of facility man-

agers in MU’s buildings. List of devices are installed

in each building, such as various sensors, network

sockets, electrical fuse, water ﬂow meters, various elec-

trical devices and much more. These devices further

connected with CCTV, security systems, ﬁre alarm sys-

tems, time multiplexers, structured and unstructured

cables, etc.

Facility managers perform analysis on basis of

readings generated by active devices. The efﬁcient

and complicated systems connect devices and reports

administrator about various kind of readings received

from active devices. Hence, to search for a list of ac-

tive devices in room, a list of active devices should be

provided as a result by the Ontology Model to building

operators. Same as, if complete fact about a room is

queried then complete fact about a room should be

forwarded by the Ontology Model to building opera-

tor. In some cases, it is required to get information

about a speciﬁc device, which is installed in a build-

ing. Therefore, the Ontology Model should compile a

list of rooms and sent to building operator, where that

speciﬁc device is operative.

Such queries facilitate not only building operators

but also higher authorities in a FM for performing

various analysis concerning active devices in several

buildings of MU.

6 INFORMATION FILTERING

Capability of the model is to ﬁlter relevant require-

ments to facilitate user according to her queries. The

Ontology Model captures available information con-

necting certain devices with Systems and Sub-Systems

and also room information where devices are actively

functioning. A pictorial illustration of the developed

Ontology is represented in ﬁgure 3.

As discussed before, three types of cases facilitates

end-user’s requirements. First case is to ﬁlter all those

devices with connecting System, Sub-System, Device

Type and Index No. Therefore, complete room infor-

mation should be provided, which identiﬁes the room

its location in building and site. For ﬁltering required

information, following logic is used to create a query

with required information of a room at a ﬂoor in a

building and site.

Identiﬁer (?ID)

∧

hasSpeciﬁc (?ID, ?Room)

∧

hasPerticular (?ID, ? Floor)

∧

isLocatedIn (?Room,

?Building)

∧

isSitutatedAt (?Building, ?SiteCode)

∧

isDistinguishedBy (?ID, ?Index)

∧

isEquiped-

With(?Room, ?Devices)

→

isAssignedTo (?ID, ?De-

vices)

In this scenario the concept Identiﬁer ﬁlters re-

quired room in a building at a speciﬁc ﬂoor, accord-

ingly. After ﬁltering the room from complete knowl-

edge of building, then the “Identiﬁer” selects list of

those devices which are installed in that room.

In second scenario, described below, an Identiﬁer

of required device is selected by ﬁltering it according

to required System, Sub-System and Device Type. Af-

ter ﬁltering identiﬁer according to device information,

a list of rooms is selected from domain knowledge

ICSOFT-EA2014-9thInternationalConferenceonSoftwareEngineeringandApplications

114

of Ontology, in which location of installed particular

device is provided.

Identiﬁer (?ID)

∧

hasDeﬁned (?ID, ? DeviceType)

∧

isEquipedWith(?Room, ?DeviceType)

∧

hasDeﬁned

(?DeviceType, ? SystemAndSubSystem)

∧

isDistin-

guishedBy (?ID, ?Index)

→

isInstalledIn(?ID, ?Room)

Selecting list of rooms, in third scenario, has same

procedure, as in ﬁrst scenario. Only, selection of Iden-

tiﬁer is done by providing required information about

device i.e. System, Sub-System and Device Type. The

Index Number is not provided to ﬁlter the Identiﬁer,

because similar Device Types, connected with similar

System and Sub-System have different Index Number.

When appropriate Identiﬁer is identiﬁed using ﬁlter-

ing procedure, a list of room is selected from domain

knowledge of Ontology.

Identiﬁer (?ID)

∧

hasDeﬁned (?ID, ? DeviceType)

∧

isReferredTo(?ID, ?SubsystemAndSystem)

∧

isCon-

nectedWith (?DeviceType, ? SystemAndSubSystem)

∧

isInstalledIn (?DeviceType, ?Room)

→

isAssignedTo

(?ID, ?Room)

According to the third scenario; for example, a

building operator search for a list of room that has

Device Type “SK” connected with System “C” and

Sub-System “F”. Therefore, she has to provide the

device information in SPARQL query, as shown below.

Initially the query selects all those identiﬁers who has

Device Type “SK”, System “C” and Sub-System “F”,

therefore a long list of identiﬁers is selected. In second

step pattern matching process is performed. Therefore,

initially, the identiﬁers of Device Type “SK” having

Sub-System “F” are ﬁltered. Then the resultant identi-

ﬁers from Device Type “SK” and Sub-System “F” are

ﬁltered according to System “C”. The identiﬁers other

than System “C” are removed from ﬁltered list. At this

point, all those identiﬁers are listed, whose System

is “C”, Sub-System is “F” and Device Type is “SK”.

Finally, according to the ﬁltered list of identiﬁers, com-

plete information of Room is selected according to Site

Code, Building Number, Floor and Room information.

Select ?Sc ?B ?F ?R

Where { ?id1 Abstract:hasSpeciﬁc

?R. ?id1 Abstract: hasParticular ?F.

?R Abstract:isLocatedIn ?B.

?B Abstract:isSituatedAt ?Sc.

{Select ?id1 Where {

?id1 Abstract:hasConnection Abstract:SyC.

?id2 Abstract:hasReferred Abstract:SuF.

?id3 Abstract:hasAssigned Abstract:DtSK.

FILTER (?id1 = ?id2)

FILTER (?id2 = ?id3)}

} }

Figure 5: Results of SPARQL query.

Figure 5, describes results of above deﬁned query.

The query is applied on Semantic Model of one build-

ing i.e. “BBA”. The results explains that the queried

device, which is connected to System “C” and Sub-

System “F”, and is actively functioning at three rooms

of the building. The results also describes that two of

the devices are installed in one room, i.e. the room

“R001d”, shown in ﬁgure 5, is at Floor “N01”, Build-

ing “01” and Site Code “BBA”. The complete location

address of the room is “BBA01N01001d”, this address

is understandable by end-users at MU.

The Ontology Model is developed according to ex-

plained structure of BIM. SPARQL queries are applied,

subsequently to requirements of building operators and

facility managers. The Ontology Model personalizes

the information related to BIM and reduce informa-

tion load by ﬁltering irrelevant data considering users

requirements.

7 CONCLUSION

This article is about FM and explains the integration

of BIM and BMS systems. The proposed approach

addresses the missing semantic information of BMS.

Therefore, facility managers can perform operation

analysis in large-scale environments. The designed

ontology covers the concepts of BIM, used for Lo-

cation Information and Device Information. Queries

are applied on the Ontology Model for reasoning the

Location Information and Device Information based

on hierarchical structure used in BIM.

Ontology Model helps the developers in BMS to

focus on user interface and analytical methods rather

than on data integration. Therefore, facility managers

will be able to perform analysis and decision making

for future planning. This is the signiﬁcant improve-

ment in current analysis work ﬂow.

The research in several areas of large-scale BMS

data analysis is possible by introducing “Semantic

smart building ontology”. Initially, advanced analyti-

cal tools should be developed. Additional research is

also required in the ﬁeld of user interfaces, both for

the query deﬁnition and results presentation.

SemanticWebTechnologyforBuildingInformationModel

115

ACKNOWLEDEGMENTS

This work was carried out during the tenure of an

ERCIM "Alain Bensoussan" Fellowship Programme.

The research leading to these results has received fund-

ing from the European Union Seventh Framework Pro-

gramme (FP7/2007-2013) under grant Agreement No.

246016.

REFERENCES

Andrew, H., Eric, L., and James, R. (1998). Intelligent

Buildings in South East Asia. Publisher E & FN SPON,

London, Chaper 1.

Asfand-e yar, M., Anjomshoaa, A., Weippl, E., and Tjoa, A.

(2009). Blending the sketched use case scenario with

license agreements using semantics. In Karagiannis, D.

and Jin, Z., editors, Knowledge Science, Engineering

and Management, volume 5914 of Lecture Notes in

Computer Science, pages 275–284. Springer Berlin

Heidelberg.

Carmen, K. and Desislava, Z. (2011). Lexico-syntactic pat-

terns for automatic ontology building. In Proceed-

ing of the Student Research Workshop. International

Conference on Recent Advances in Natural Language

Processing. RANLP, pages 109–114.

Christian, R., Wolfgang, G., Fritz, P., and Woﬂgang, K.

(2008). Integration of heterogeneous building automa-

tion systems using ontologies. In International Con-

ference on Industrial Electronics IECON, pages 2736–

2741. IEEE.

De Paola, A. (2014). An ontology-based autonomic sys-

tem for ambient intelligence scenarios. In Gaglio, S.

and Lo Re, G., editors, Advances onto the Internet of

Things, volume 260 of Advances in Intelligent Systems

and Computing, pages 1–17. Springer International

Publishing.

Dekdouk, A. (2013). A mobile-based automation system

for maintenance inspection and lifesaving support in

a smart ict building. In Workshop AmI 2013, CCIS,

pages 320–335. Springer.

Kroner, W. M. (1997). An intelligent and responsive ar-

chitecture. ELSEVIER, Automation in Construction,

6:381–393. Issue 5-6.

Mario, J. K. and Wolfgang, K. (2010). A knowledge base for

energy-efﬁcient smart homes. In International Energy

Conference and Exhibition (EnergyCon), pages 85–90.

IEEE.

Massimiliano, G. and Giuseppe, P. (2013). Visual computer

managed security: A framework for developing ac-

cess control in enterprise applications. IEEE Software

Journal, 30:62–69. Issue: 5.

Michele, R., Floriano, S., Loseto, G., and Eugenio. Di, S.

(2014). Semantic-based resource discovery and orches-

tration in home and building automation: A multi-agent

approach. IEEE Transactions on Industrial Informat-

ics, 10:730–741.

Nicolle, C. and Cruz, C. (2011). Semantic building infor-

mation model and multimedia for facility management.

In Filipe, J. and Cordeiro, J., editors, Web Informa-

tion Systems and Technologies, volume 75 of Lecture

Notes in Business Information Processing, pages 14–

29. Springer Berlin Heidelberg.

Paul, M., Rob, B., Declan, O., Marcus, K., Kris, M., and

James, O. (2009). Pervasive knowledge-based net-

working for maintenance inspection in smart building.

In International Workshop on Managing ubiquitous

Communication MUCS, pages 59–65. ACM.

Paul, S., Karsten, M., Denis, F., Ammar, A., and Sean, M.

(2012). A cloud-based platform for energy and main-

tenance management. In International Conference on

Computing in Civil and Building Engineering, ISC-

CBE, pages 27–29.

Rob, B., Wei, T., Declan, O., Muhamamd, S. A., Susan, R.,

and Dirk, P. (2009). Open framework middle-ware

for intelligent wsn topology adaption in smart build-

ings. In International Conference on Ultra Modern

Telecommunication & Workshops, pages 1–7. IEEE.

Seppo, T. (2013). Semantic linking of building information

models. In International Conference. on Semantic

Computing (ICSC), pages 412–419. IEEE.

Wong, J., Li, H., and Wang, S. (2005). Intelligent building

research: a review. ELSEVIER Automation in Con-

struction, 14:143–159. Issue: 1.

Yulia, E., Aleksandra, D., Jose, L., and Martinez, L. (2013).

Semantic information management for user and context

aware smart home with social services. In International

Conference on Cognitive Methods in Situation Aware-

ness and Decision Support CogSIMA, pages 262–268.

IEEE.

Zarrad, R., Doggaz, N., and Zagrouba, E. (2012). Toward a

taxonomy of concepts using web documents structure.

In Proceedings of the 14th International Conference on

Information Integration and Web-based Applications

& Services, IIWAS ’12, pages 147–156, New York,

NY, USA. ACM.

ICSOFT-EA2014-9thInternationalConferenceonSoftwareEngineeringandApplications

116