Automation in Construction 39 (2014) 126–133

Contents lists available at ScienceDirect

Automation in Construction

j ourna l homepage: www.e lsev ie r .com/ locate /autcon

Intensified construction process control using information integration

Esa Viljamaa a,⁎, Irina Peltomaa b

a VTT Technical Research Centre of Finland, P.O. Box 1100, FI-90571 Oulu, Finlandb VTT Technical Research Centre of Finland, P.O. Box 3, FI-92101 Raahe, Finland

⁎ Corresponding author. Tel.: +358 20 722 2355; fax: +E-mail address: esa.viljamaa@vtt.fi (E. Viljamaa).

0926-5805/$ – see front matter © 2013 Elsevier B.V. All rihttp://dx.doi.org/10.1016/j.autcon.2013.08.015

a b s t r a c t

a r t i c l e i n f o

Article history:Accepted 22 August 2013Available online 17 September 2013

Keywords:Infrastructure buildingProcess controlInformation integrationSemantic methodsInformation managementOntology

In this research, novel information technology and advances in communication and machine control systemswere combined into a method intensifying construction control. Changes in the operation environment of con-structors have revealed a clear need for more efficient process management. The objective of this research wasto improve process management through more effective information integration, processing, and exploitation,leading to an intensified infrastructure building process with more effective process control and reaction to pro-cess status changes. Based on requirements gathered through interviews with domain professionals, the devel-oped method exploits ontology-based information formulation to integrate design and as-build data with thehelp of advanced communication and machine control applications. In addition to actual construction processcontrol, the use of the developed method potentially intensifies operations before and after the project, enhanc-ing process efficiency, cost-effectiveness, and eco-efficiency. The results of the presented development phasewillbe verified in the following practical implementation phase.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

The infrastructure building companies among other industries arefacing huge pressure to strengthen and intensify all the processes dueto hard competition, new regulations and public demand formore envi-ronmental friendly processes. Talking about the larger scale infrastruc-ture building projects they are less organized and formalized due tomore badly measured process parameters, low technical status of par-ticipants in the process and sometimes quite complicated subcontractorand system chains. The technology variety and technology knowledgelevel of different subcontractors in the same construction site maycause difficulties to project management to integrate and exploit rele-vant process data. The information needed may be scattered in severalinformation systems which are at different technology stages. The pro-ject management requires combined information for efficient decision-making. Usually an infrastructure building project consists several par-allel processeswheremachine break down or other kind of interruptionmay cause delays and unplanned costs. A random interruption in theprocess containing several parallel sub-processes may cause large-scale problems. These interruptions should be handled immediatelyand preferably automatically as fast as possible. However, in manycases in the infrastructure building field, process data is handled slowlyand manually by error-prone human operators without knowledgefrom every affecting process parameter. The implication of the afore-mentioned is that in infrastructure building process, the process control

358 20 722 2320.

ghts reserved.

is very challenging. Difficulties in control lead to the situations wherethe status of the sub-processes is worse known, whichmakes the over-all process control and dynamic reaction to the changes impossible andtherefore processes are inefficient, hazardous, expensive and slow. Tosummarize, in order to be efficient in process management, one has tohave themost recent and refined process information that is not nowa-days possible due to heterogeneous and incompatible information sys-tems of different counterparts. In a way of development of improvingsystems are at least mentioned highly varying technology awarenessof different counterparts, closed systems of different machine and soft-ware provider systems and lack of standards in data formats etc.

According to the description above, there is a clear need in the infra-structure building industry for the deployment of more intelligentmethods to manage projects. These methods should concern both theinformation of the original plan and the dynamically changing construc-tion time process data. The research problem was to develop suitablemethod to enhance the process management by providing more exten-sive information accessible from one place. The developed methodshould enable an integration of themost important subcontractor processinformation to intensify the infrastructure building process by enablingmore efficient process control and reaction to the changes in the field.The method should take advantage of the current innovations to be ad-vanced. The utilized integration method should not only be conventionaldata format based point-to-point integration but also take into accountnovel methods enabling advanced utilization of the integrated data. Theimplemented integration method should be general purpose in natureto ensure maximal applicability. The main benefits of the method shouldbemore efficient, cost-effective, more environmentally friendly and saferinfrastructure building projects through the improved project data

127E. Viljamaa, I. Peltomaa / Automation in Construction 39 (2014) 126–133

exploitability. The most important facilitators of the developed methodare novel technologies in the areas of the machine control systems, dataformats, location systems, information management technologies andfinally general web technologies. Fig. 1 illustrates the basement of the de-velopedmethod, the utilization of common ontology format that enablesintelligent access to the integrated project data. The method is based onmaterial gathered from the companies operating mostly in Scandinavianarea and results are therefore biased to follow in Scandinavian culture ofproject management and work in road construction.

This paper presents an advanced method for intensified infrastruc-ture building process. By improving processmanagement the infrastruc-ture building process is accelerated and rationalized. The intensificationhere refers to more cost-effective, quicker, and safer process manage-ment when the current and forecasted process status is better knownto the process management personnel. The significance and the level ofthe intensification of themethodwill bemeasured in the sub-sequent re-search where two similar projects will be managed and compared withand without the utilization of the developed method. The novelty ofthe developed process control method lies on combining recently devel-oped technologies with new ways and offering an effective and moderntool for infrastructure building site management. In the method, themanagement of the design data, automatic dynamically changing andmanually added process data is formalized in away that it could be easilyintegrated as a whole and also exploited efficiently by using automateddata fetching and inferring.

2. Novel technologies and research methods

2.1. Infrastructure construction process management in academia

In the academia, the infrastructure building topics as well as systemintegration are vastly researched areas. Boddy et al. [1] havemade an in-teresting review from the point of view of computer integrated con-struction, and Shen et al. [2] have made an excellent review on systemintegration in the field of AEC/FM (architecture, engineering, construc-tion and facilities management). In the following some examples of in-tegration research in the area of infrastructure building are gathered.

In themid-1990s Navon et al. [3] begun to research construction in-dustrymanagement. Industry related state-of-the-art research has beenconducted in Israel [3]. Navon and his colleagues focused in their re-searches on automated productivity measurements using GPS-dataand the concept of work envelopes [3,4]. In the same time Navon statedthat in the beginning of the 2000 century the background for the

Fig. 1. Introduced process management method brings advances through easier d

construction process control was thin and all the actions were done bymanually made calculations [3].

Later on twoGerman research projects have started to study thewaysto improve construction processes and management. Researched focusareas were e.g. improving the total construction process and optimizingthe workflowwith the help of machine control, logistics planning, virtualphase visualization, documentation, and data management. State-of-the-art technology like PDM-systems and close range identification was usedto help in reaching the targets [5–9].

Several academic researchers have also researched lately the poten-tials of ICT-related constructionmanagement [10–12]. The common fac-tor of this ICT-related researched mentioned above is the use of novelICT technology for process measurement and follow-up.

Kosovac (2007) [13] proposes a framework for information use andmanagement in AEC/FM. The prerequisite for meeting requirements ofinformation management of complex domain is the efficient communi-cation between parties, both humans and machines. The frameworkidentifies three basic types of assertions (senses, relationships, informa-tion) and their two properties (category, scope)which are used to relateall kinds of semantic resources and informationmanagement approaches.In the pilot implementation numerous and diverse components sharetheir content via Web Services using the proposed framework.

El-Diraby et al. (2005) [14] present a domain taxonomy for construc-tion. The taxonomy is based on IFC (Industry Foundation Classes) andseveral other classification systems. It uses sevenmajor domains to clas-sify construction concepts: Process, Product, Project, Actor, Resource,Technical Topics, and Systems. The major ontological model is process-oriented and can be summarized as follows: construction knowledge isencapsulated in several overlapping systems, where a set of Actors usea set of Resources to produce a set of Products following certain Process-es that are part of a Project according to boundary conditions andwithinthe confines of the work environment (Technical Topics) [14]. The oper-ation of developed domain ontology was evaluated during e-COGNOSproject [15] as a part of web based knowledge management software,which connected various systems usingWeb Service technology. The de-velopment of ontology architecture was continued by adding moreknowledge levels (application knowledge, user knowledge) to domainontology [16].

The study presented in this paper has the same kind of ideas ofusing ontologies for system integration and information managementduring infrastructure construction process as in previously presented re-searches. However, the focus of the study is on how to use developed in-formation integration method and other mentioned novel technologies

ata integration and more versatile and efficient utilization of integrated data.

128 E. Viljamaa, I. Peltomaa / Automation in Construction 39 (2014) 126–133

emphasizing the construction time process follow-up and to improvingprocess management.

2.2. Semantic Technologies state-of-the-art

As amain technology for realizing themethod for improved processcontrol, Semantic Web related technologies were utilized. SemanticWeb is a concept for the next generation WWW in which informationis given well defined meaning [17]. The underlying technological infra-structure of Semantic Web is referred to as Semantic Technologies [18].

Semantic Technologies are rather new technologies and still in theearly adoption phase in industry, and the use of these technologiesis more familiar in e-commerce and Semantic Web applications. Se-mantic Technologies provide wide and scalable methods and tools forthe machine accessible representation and manipulation of knowledge[19]. Semantic Technologies contain for example ontology tools, dy-namic mediation tools, inference tools and thesaurus tools [20]. Ontol-ogies and ontology tools are in central role in information integrationand in enabling semantic interoperability. Semantic interoperability isthe ability of participating system domains to understand the meaningand use of terminology from different domains, and to map betweenagreed concepts in order to make a semantically compatible informa-tion environment. Semantic interoperability can be ensured by provid-ing contextual knowledge of domain applications [21]. In informationintegration the semantics of information is preserved while the contextis transformed. Semantic Technologies, especially the usage of ontol-ogies, offer away towards solving the information integration challenge[22].

2.3. Ontology state-of-the-art

Concept ontology has various definitions, but most widespread is“An ontology is an explicit specification of a conceptualization [23].”.As expressed simply ontology is a conceptual datamodel used to link to-gether other,more granular datamodels in order to reach agreement ondatameaning [20]. Ontology defines a common vocabulary for informa-tion sharing in a domain to be reused and shared across software appli-cations and by groups of people [23–27].

According to Zimmermann et al. (2005) [28], ontologies provide away to define semantics, provide support for handling disparate datasources and provide mechanism to define complex knowledge models.Ontologies act as a vehicle that carries information semantics and can beutilized to discover the semantic mappings between systems [27].

Ontologies are usually developed independently of each other fordifferent purposes [29], and it will rarely be the case that a single ontol-ogy fulfills the needs of a particular application [30]. When workingwith several ontologies, methods for their coordinated concurrent usebecome necessary. For extending the captured specialist knowledge ofa certain domain or relating it to other domains, ontologies have tobe reconciled. The techniques for doing so include matching, mappingand alignment [29].

In general three approaches to utilize ontologies in information inte-gration from heterogeneous information sources can be separated: thesingle ontology approach, the multiple ontology approach, and the hy-brid approach [31,32]. Respectively mapping of ontologies can be cate-gorized in different methodologies depending of the approaches to useglobal ontology together with local ontology; either local ontologiesare used alone (one-to-one paradigm), or a global ontology exists eitherwithout (single-shared ontology) orwith local ontologies (mix of single-shared and one-to-onemapping) [33,34]. Single ontology approach usesone global ontology for providing a shared vocabulary for the specifica-tion of the semantics; all information sources are related to the one glob-al ontology (single-shared ontology). In multiple ontology approach,each information source is described by its own ontology and source on-tologies are mapped to each other using inter-ontology mappings (one-to-one). In hybrid approach the semantics of each source is described by

its own ontology, which is built upon one global shared vocabulary orontology (mixed) [31,34].

In ontology basedmodel the semantics of information is modeled asclasses and relations between those classes. The W3C's Semantic Webactivity has developed many related technologies such as Resource De-scription Framework (RDF) [35], RDF Schema (RDFS) [35],Web Ontolo-gy Language (OWL) [36], and SPARQL [37] to realize the ontology basedinteroperability in the WWW.

2.4. Open design data formats

New open design data formats for the building processes makeit possible to reuse design data in different CAD and other applications.These new data formats are in many cases well documented and XML-formatted making it even easier to be exploited in general. Thedeveloped method exploits one of the lately developed formats calledInfraModel2 that is based on LandXML specification and targets i.e.road, street and water route design [38,39]. The sub-ontology used inthis research for design data was developed to follow the InfraModel2specification by extracting the concepts and relations and modelingthem into OWL Lite format. XSLT transformation was used to convertthe design file to RDF format and parsed to comply with ontology.

2.5. Data integration

The semantic interoperability provided by e.g. BIM formats is impor-tant, but not the only problemwhen data integration is realized in prac-tice. As it is known, used common BIM formats do not solve problems ofdata integration. There must also be a plan on how to provide useful in-terfaces for practical software queries to integrated data. Basically thereare two main possibilities to do it: data warehousing or mediated sche-ma. Both have its pros and cons concerning their updateability, datanewness and applicability [39,40].

In construction research, the usage of data warehousing technologyis quite common [41], but the mediated schema approach is gainingmore attention due to its more flexible nature and growing demandsfor freshness of the used information in systems [42].

In this research, the data warehousing was chosen and XML basedBIM formatswere chosen to be used to ease the data transformation pro-cess. In addition, the data warehouse used triple store as a database typeproviding possibilities for intelligent queries and computer inferring.

2.6. Commercial automated machine control systems

The recent commercial machine control systems have opened acompletely new era in some of the sub-processes in the infrastructurebuilding. With the help of the wireless communication technologies,machines can now receive location, work instructions and send up-to-date process data on-demand enabling better real-time process controlon their behalf. Novel machine control systems may also provide quiteaccurate measurements of realized work phases that may be later onused for process follow-up when the realization data is compared tothe target or design data. At least Novatron [43], Trimble [44] andTopcon [45] provide mentioned control systems.

2.7. Commercial digital site management systems

There are few bigger players like Topcon [46], Trimble [47] andCAT [48] providing site management systems and software enablingcontrol of multi-vehicle worksites. These systems providemap basedviews of vehicle status and even some process related informationlike transported mass amounts etc. However, these systems almostalways only work within themachines provided by the samemanufac-turer. They are not functioning in theworksites having systems andma-chines frommultiple vendors and that is causing problems to integrateproject data as a whole.

129E. Viljamaa, I. Peltomaa / Automation in Construction 39 (2014) 126–133

2.8. Wireless data transfer technologies

Realization of themethod described in this studywould not be possi-blewithout efficient data transfer. Novelwirelessmobile communicationtechnologies like wireless operator networks provide efficient tool formobile unit data updates. The requirement to transfer larger amountsof data wirelessly more frequently is clear if the greater level of real-time behavior is wanted. E.g. 4G/LTE networks provide enough band-width near densely populated areas, but in the sparsely populatedareas the needed high data rates may cause problems.

Table 1The most important requirements for the developed method.

Need Explanation

Semi-real-time Response time should be within tens ofseconds.

Easy system integration of differentactors needed without customprogramming

Common data models for both design andrun-time process data provider parties areneeded.

Automated decision making Data formats supporting machine basedinferring.

Non-system dependent Using information management systemshould be done using common, platformindependent technologies.

Possibility for decentralized system No need for common repository forprocess data.

Selected process parameter monitoring Necessary function of the concept,concerning proper site progress monitoring.

2.9. Research methods

The research method used in this research was constructive [49].The objective of this research was firstly to develop tools for advancedprocess control and afterwards subsequently to implement a technicalsolution for validation of the developed method. As inputs to construc-tive method were state-of-the-art study and technology reviews andthe contractor and their employee interviews, industry partner work-shops and worksite visits.

The need for the developed method and suitability of relatedtechnology types was researched in several different phases. In thebeginning of the research the state-of-the-art survey was done tostudy similar systems among the commercial products and academyresearches to verify the need of new method and to potentially findalready developed corresponding methods. In order to collect domainknowledge and information about current management practicessome field study observations were conducted. During observationsthe daily routine of building site management personnel was followedand the information collection and utilization in building site manage-mentwere studied. Altogether threemodern building siteswere visitedin Finland, as those used the most recent automation systems in build-ing process. Two workshops among the project partners to discuss thetopic were also held. After perceiving the overall picture of the buildingsitemanagement, potential technology enablerswere studied to ensurethe method real-life implementation feasibility.

In addition to the state-of-the-art review and field study observa-tions, personnel of the involved project partners were interviewed toget a better idea of the developedmethod requirements. The interviewsof the study covered personnel basically from each sector of typicalactors from larger scale infrastructure building project. Among theinterviewed key personnel there were representatives from a builder,a designer, and a main contractor as well as subcontractor system sup-plier companies. Subcontractor system supplier companies covered adesign tool provider, a mass transportation management system pro-vider, a machine control system provider and a road measurementsystem as well as end users of those systems. The amounts of theinterviewed personnel were only 16 area experts altogether includingdevelopment managers (3), development engineers (5), a worksite op-erator (1), worksite supervisors (5) andoperationalmanagers (2). Sincethe amount of the interviewed personswas quite few, the interview re-sults are not all covered, merely giving a basic idea of the system re-quirements and a direction to proceed. Sub-topics of the interviewscovered partner's recent official practices concerning project-wide andinter-partner information flow, information access and management,future plans and interviewee's own opinion of the way practices shouldbe realized. The interviews and the questions were tailored accordingthe interviewee type. The answers included direct suggestions on howto improve practices as well as indirect observations that were translat-ed to requirements by authors. E.g. one of the answers of the maincontractor explained what the most important observed process pa-rameters are andwhat the preferred update interval for them is. The in-terview results completed the information received from worksiteobservations, a partner workshop, the state-of-the-art review and theauthors' experience from the other related projects.

All mentioned steps lead to a gap analysis where the current prac-tices in the field and desired features of future system were identifiedand analyzed and the most important of them were selected to beused as method requirements.

3. Results

3.1. Method requirements

The requirements of the method and information delivery systemwere gathered mainly during interviews of the personnel of the in-volved project partners. The requirements of the system contain boththe exploitation possibilities of the information and the overall integra-tion system requirements. The most important requirements for theintended control system are listed in Table 1. The list reveals that an ef-ficient control system for the purpose includes at least some of the fol-lowing features: soft real-time, decentralization, system independency,pre-agreed subcomponent interfaces and decision-making capabilities.State-of-the-art survey revealed to us that there is general purpose in-formation integration method developed in academia suitable for theinfrastructure building processes [16]. However, there was no such acase found where the general information integration method wasused for construction time process control purposes. There are alsomany different examples on how different sub-processes or special fo-cused needs could be covered, but none reallymatching the needsmen-tioned above.

3.2. Method description

Themethod presented is based on the previously described require-ments. Themethod developed in this research is allocated especially forinfrastructure building, but the solutions made can be applied also toother domains, especially regarding the information integration. Afterstate-of-the-art study and interviews the most suitable technologieswere selected to be used for the base of themethod. As studied in earlierchapters, one of the most important matters in order to realize anadvanced process control system seems to be general data share and in-teroperability that can be realized e.g. with pre-agreed common dataformats that enable assembling of smaller portions of data to usablewholeness. Those data formats should also bemachine understandable.Also important properties of the control system are system platformindependency and possibility to realize system using decentralizedarchitecture. The main idea of the developed method is to developand use shared ontology model that contaINS sub-ontology foreach sub-process of the infrastructure building process. This divisionmainly follows the subcontractor types, but when it is possible, somesubcontractor information could be included in the same sub-ontology. For example there should be different sub-ontologies forthe design phase and work progress. In the case of mass transport or

Fig. 3. Initial prerequisites of the control method.

130 E. Viljamaa, I. Peltomaa / Automation in Construction 39 (2014) 126–133

excavating companies the combining of sub-ontologies is possible. Fig. 2illustrates a simple snippet of sub-ontology of transportation and workmachine where the information of different subcontractors has beencombined under Resource concept in ontology. Sub-ontologies containonly relevant data instances and their interrelations.

This research uses single ontology approach as ontology architecture;each source is preserving its own ontology and global, shared ontologyis constructed to act as common vocabulary. The shared ontology guar-antees the information interoperability and possibility to do inferringbased on the data relations. Using the developed method, all the ontol-ogies were described using OWL Lite as ontology description languageto enable easy ontology combination from the different information frag-ments. The shared ontology was created using Protégé framework [50].All the process data from the design to the real-time process data isstored to the triplet database to be easily deployed also using inferringfeatures when needed. Sesame [51] is used as a database and for query-ing and updating the information SPARQL 1.1 query language [52] isused.

3.3. Method utilization prerequisites

In order to realize an efficient method for process managementbased on system requirements described in the previous section,there are some initial prerequisites to be fulfilled. These prerequi-sites should be taken into consideration in the very early phase ofagreement concluding between the project participating companies.The most important prerequisites are explained in Fig. 3 and Table 1.These prerequisites are needed since at present there are no stan-dard ontology templates for different actor types.

3.4. System architecture

In the developedmethod, themain contractor is a party that is coor-dinating data collection and reuse. Both decentralized and centralizedmethods can be used. In decentralized method, all the other participat-ing companies just give the main contractor an access to their remoteinformation repository containing generated process data in some se-mantic e.g. OWL format and the main contractor fetches data fromthere and combines them when necessary. In our study however, thecentralized method is the one that is in focus. In the centralized method,all the parties send their design and process data in correct format to

Fig. 2. A sub-ontology snippet represe

the server provided by the main contractor in agreed intervals and themain contractor takes care of the data extraction, transformation andexploitation.

Fig. 4 shows a simplified architecture and the data transfer principlesof the centralized version of the developed information integrationmethod.

Designers deliver design related information, like constructionand schedule design, to main contractor's server as well as informa-tion about any changes made to the design information during theproject. Subcontractors transfer process related data to the maincontractor's server; this information may contain e.g. excavationstatus and mass transfer data with related machine location data.Subcontractors are able to receive realization information about pro-ject and for example work instructions through information system.Main contractor adds its own process related data, which is com-bined to subcontractors' data. The developed method offers integrat-ed information to main contractor through information integrationsystem's user interfaces. All the data transferred to main contractor'sserver is formatted according to agreed ontology model. The ontolo-gy in this study was realized using OWL.

nting ontology concept Resource.

Fig. 4. The control system architecture.

131E. Viljamaa, I. Peltomaa / Automation in Construction 39 (2014) 126–133

3.5. Illustrative use case of the method

An illustrative use case in Fig. 5 describes an explanatory use of thenew control method concept. The use case presents the use of the con-trol method in a road construction case. The use case is divided into

Fig. 5. An explanatory usage

three parts and it considers three actor types, designer, main contractorand subcontractors. First phase of the use case is initial phase and thefollowing phases are operational and post-operational of their nature.The presented use case is ideal and heavily simplified. The use casewill be implemented for real-life implementation in the future research.

3.5.1. Initial phaseIn the initial state all needed road designs,machine guidancemodels,

mass hauling plans and schedules are uploaded into the server of themain contractor following the researched prerequisites described inTable 2. Initial state models, plans and schedules have designed bygroup of designers and verified by construction site operating officer.Naturally all the initial data is following the shared ontology formatand is therefore compatible with each other.

3.5.2. Operational phase, task executionExecution phase is divided into two parts; earth removal and mass

hauling. Therefore execution phase considers only two actor types, ex-cavator and truck. An excavator of a subcontractor is equipped withstate-of-the-art machine guidance system with wireless data transfertechnology. Earth removal is done following the instructions fromama-chine guidance models downloaded from the main contractor server.An excavator can confirm sub-tasks completed manually or using auto-matedmachine control model andmeasurement result basedmethods.Subcontractor's mass hauling trucks record every load they transportusing e.g. a mobile application. Recorded mass hauling data is sent in

of the control method.

Table 2Explained method prerequisites.

Prerequisite Explanation

1 The builder should order a product design either directly in commonontology format or easily converted one supporting also machinecontrol system conversions for better design data compatibility,exploitability and better general reuse. Designs or more specificallymachine guidance models should be divided into parts suitable fordeveloped concept.

2 The builder should order themain contractor to log themost importantprocess parameters in common ontology format for more efficientquality control, traceable billing and maintenance purposes.

3 The main contractor should demand subcontractors to provide theproduced process parameters to be available in shared ontology forbetter information and projectmanagement, safety, cost-efficiency andeco-efficiency.

132 E. Viljamaa, I. Peltomaa / Automation in Construction 39 (2014) 126–133

real time to the main contractor server, where it can be compared toplan exploiting ontology model semantics if wanted.

In the execution, the main contractor can monitor and control thetask progress by comparing the schedules and realization modelssince they are part of the same common ontology model. Progressmay be presented graphically enabling easy project status observa-tion. If realization seems too fast or slow the supervisor can react ac-cordingly. The actual control of the process may be realized throughmanual actions by work supervisor or by pre-defined automatic ac-tions by software.

3.5.3. Operational phase, quality assurance and task approvingIn a quality assurance and task approving phases the executed work

is recorded by machine guidance and mass hauling systems and resultsare delivered to the server as an as-buildmodel. After comparing the as-designed and as-built models the main contractor can approve the taskand give the permission to start the next phase of the construction.

3.5.4. Post-operational phase, exploiting collected dataSince all the design and process data are stored in semantic format it

may be possible to easily utilize it for purposes other than building timeprocess control. Using intelligent queries and semantics of data, the col-lected data may be helping in further tasks like cost and other estima-tions for future projects.

4. Discussion

The driving force of our research was the lack of proper multi-actorgeneral tool for project management concerning the site officer, whois in charge of daily construction actions. The general process manage-mentmethodwas developed for that purpose. Utilization of themethodcould also affect inmanydifferent sub-topics in the infrastructure build-ing sector as well as to other domains with scattered contractor chainsand not formalized production methods. The gained benefit comesdue to formalized, easily combined data formats ensuring better inter-operability and enabled automatic inferring and decision-makingbased on the collected data. The design data could be used more effec-tively when it is used as an initial data to be compared to the realizationdata which improves process follow-up. Also the collected process datais easier to integrate in order to enhance the processmonitoring, qualitycontrol, faster reaction, safer and more efficient processes and after theproject for maintenance focusing and improvement. The collected datacould also be used after the project as a reference for a more efficientoffer requesting processes and descriptions. The semantic way to pres-ent project data with internal relations eases also the exploitation ofnovel PDM-methods.

The developed method is quite scalable and general purpose in itsnature. Each company taking part to a construction project could usetheir best known software technology and platform and take part inthe information chain. The method is also usable in other related tasks

e.g. invoicing process as for example the realization of a certain masshauling task may be allocated to a certain subcontractor. This reducesthe need of human work in invoicing as it reduces paper handling.This also gives better opportunities to main contractor to order extramachines for example for mass hauling.

Previous and currently running research projects have been studied.Methods, how to monitor and manage processes and how to gatherliable status information from the infrastructure building projects wereresearched. In many cases automated data gathering and GNSS-positioning systems are the main factors of the developed concepts,but theway onhow this gathered data is analyzed and used again variesbetween different projects. Nowadays the common way is to developan algorithm that monitors automatically where the assets are movingand realizes which part of process is in progress. The control methoddeveloped in this research exploits also GNSS-positioning and othernovel ICT technology, but the main focus is in the ontology based infor-mation formalization and its utilization for information integrationpurposes. Several lately conducted researches have the same kind of in-tegration approach where enterprise level systems and software are in-tegrated to concept database. Developed methods have also the samekey features like web-based platforms, semantic databases and mobileaccessorieswhich are noteworthy technologies because of the latest de-velopments in wireless network and other ICT-related areas. There arealso examples where ontologies are used for information integrationpurposes in the field of infrastructure building processes. These exam-ples however did not point out how to use them to process control pur-poses and neither was there an explanation as to what kind of extraprerequisites was needed to fulfill to really deploy them.

5. Conclusion

During the research process a couple of challenges were recognized.Since the infrastructure building process is quite scattered concerning alevel of the used technology among the subcontractors there will beproblems to add all the needed parties to the system in practice. Alsothe utilization of the semantic methods demands high technology ori-entation from the using partywith the skilled personnel before the rou-tines to use the method evolve. Due to quite heavy data formats andsystem structure the method does not allow hard real-time utilization.Of course the possible production system using the developed methodwould demand heavy standardization process for development of realshared ontology. One challenge is also the lack of clear earning princi-ples. Of course, if the benefits are high enough for the main parties,the demand for method from the main builder side would speed upthe deployment process. At themoment the current subcontractor soft-ware data formats do not support the ontologies described in this re-search, causing an issue demanding data format converters or specificinterfaces for every subcontractor counterpart.

As pointed out the need for additional research is clear. Since thereare already few main example ontologies developed, the proof-of-concept implementation system should be done including the integra-tion of the design data, scheduling data and some important processdata. A tool in improving the construction time process managementwith user interface should also be included. The implementation shouldbe tested concerning the performance and utilization experiencesfrom real users. Also the business model to motivate additionalfunding for research should be clearly pointed out. Only comprehen-sive testing will reveal real benefits and application possibilities ofthe developed method in the construction and other related processcontrol applications.

Acknowledgments

This work was done in “Control of dynamic infrastructure construc-tion processes using dynamic co-operation network” DIGIINFRA pro-ject funded by TEKES (Finnish Funding Agency for Technology and

133E. Viljamaa, I. Peltomaa / Automation in Construction 39 (2014) 126–133

Innovation). The contribution of the participating companies in the pro-ject is also highly appreciated.

References

[1] S. Boddy, Y. Rezgui, G. Cooper, M.Wetherill, Computer integrated construction: a re-view and proposals for future direction, Adv. Eng. Softw. 38 (2007) 677–687.

[2] W. Shen, Q. Hao, H. Mak, J. Neelamkavil, H. Xie, J. Dickinson, et al., Systems integra-tion and collaboration in architecture, engineering, construction, and facilitiesmanagement: a review, Adv. Eng. Inform. 24 (2010) 196–207.

[3] R. Navon, E. Goldschmidt, Y. Shpatnitsky, A concept proving prototype of automatedearth-moving control, Automat Constr. 13 (2) (2004) 1225–1229.

[4] R. Navon, Research in automated measurement of project performance indicators,Automat Constr. 16 (2) (2007) 176–188.

[5] S. Kirn, C. Frantzen, M. Kaak, A. Sturm, AutoBauLog— autonomic control in construc-tion logistics, Proc. of 2nd International Conference on Machine Control & Guidance,ETH Zürich, 2008.

[6] A. Borrman, Y. Ji, I.-C. Wu, M. Obergriesser, E. Rank, C. Klaubert, W. Günthner,ForBAU — the virtual construction site project, Proc. of the 24th CIB-W78 Confer-ence on Managing IT in Construction. Istanbul, Turkey, 2009.

[7] C. Klaubert, M. Schorr, W. Günthner, Real time construction progress control usingNFC, Workshop on Smart Objects: Systems, Technologies and Applications (RFIDSys Tech2010), Ciudad, Spain, June 2010.

[8] A. Borrmann, M. Schorr, M. Obergriesser, Y. Ji, I.-C. Wu, W. Günthner, et al., Usingproduct data management systems for civil engineering projects — potentials andobstacles, Proc. of the 2009 ASCE International Workshop on Computing in Civil En-gineering. Austin, TX, USA., 2009.

[9] K. Kirchbach, C. Runde, Optimized work flow through VR and AR technology on con-struction sites, Proc. of the 15th International Conference on Information Visualization,IEEE Computer Society, London, United Kingdom, ISBN: 978-0-7695-4476-2, July 2011.

[10] C.L. Perkinson, M.E. Bayraktar, I. Ahmad, The use of computing technology in highwayconstruction as a total jobsite management tool, Automat Constr. 19 (2010) 884–897.

[11] S. El-Omari, O. Moselhi, Integrated automated data acquisition technologies forprogress reporting of construction projects, Automat Constr. 20 (2011) 699–705.

[12] P. Rausch, F. Schreiber, M. Diegelmann, Closed loop controlling approaches for pro-ject in the earth moving and construction industry, Proc. of 2nd International Con-ference on Machine Control & Guidance, March 9–11 2010.

[13] B. Kosovac, A Framework for Managing Information from Heterogeneous, Distribut-ed, and Autonomous Sources in the Architecture, Engineering Construction, and Fa-cilities Management Domain, (PhD Thesis) Dept. of Civil Engineering, University ofBritish Columbia, Canada, 2007.

[14] T.A. El-Diraby, C. Lima, B. Feis, Domain taxonomy for construction concepts: towarda formal ontology for construction knowledge, J Comput Civil Eng. 19 (4) (2005)394–406.

[15] R.V. Vallejos, C.P. Lima, G. Varvakis, Towards the development of a framework to cre-ate a virtual organisation breeding environment in themould and die sector, J. Intell.Manuf. 18 (2007) 587–597.

[16] T.E. El-Diraby, K.F. Kashif, Distributed ontology architecture for knowledge manage-ment in highway construction, J Constr Eng M ASCE 131 (5) (2005) 591–603.

[17] T. Berners-Lee, J. Hendler, O. Lassila, The semantic web, Sci. Am. 284 (5) (2001) 34–43.[18] P. Hitzler, S. Bader, A. Garcez, Ontology learning as a use-case for neural-symbolic

integration, Proc. of IJCAI05 Workshop on Neural-symbolic Learning and Reasoning,NeSy'05, Edinburgh, UK, 2005.

[19] M. Hepp, F. Leymann, J. Domingue, A. Wahler, D. Fensel, Semantic business pro-cess management: a vision towards using semantic web services for businessprocess management, Proceedings of IEEE International Conference on e-BusinessEngineeringICEBE05, Beijing, China, October 12–18 2005, pp. 535–540.

[20] J.T. Pollock, R. Hodgson, Adaptive Information — Improving Business through Se-mantic Interoperability, Grid Computing, and Enterprise Integration, John Wiley &Sons, 2004.

[21] S. Ram, J. Park, Semantic Conflict Resolution Ontology (SCROL): an ontology fordetecting and resolving data and schema level semantic conflicts, IEEE Trans.Knowl. Data Eng. 16 (2) (2004) 189–202.

[22] J.T. Pollock, The big issue: interoperability vs. integration, eAI J. 3 (2001) 48–52.[23] T.R. Gruber, Towards principles for the design of ontology used for knowledge shar-

ing, Int. J. Human–Computer Studies 43 (5/6) (1995) 907–928.[24] M. Uschold, M. Gruninger, Ontologies: principles, methods, and applications, Knowl.

Eng. Rev. 11 (2) (1996) 93–155.

[25] N.F. Noy, D.L. McGuinness, Ontology Development 101: A Guide to CreatingYour First Ontology, Stanford Knowledge Systems Laboratory, Available athttp://protege.stanford.edu/publications/ontology_development/ontology101-noy-mcguinness.html2001.

[26] A. Gómez-Pérez, M. Fernández-López, O. Corcho, Ontological Engineering: With Ex-amples from the Areas of Knowledge Management, e-Commerce and the SemanticWeb, Springer, 2004.

[27] W. Shen, Q. Hao, Y. Xue, A loosely coupled system integration approach for decisionsupport in facility management andmaintenance, Automat Constr. 25 (2012) 41–48.

[28] O. Zimmermann, M. Tomlinson, S. Peuser, Perspectives on Web Services: ApplyingSOAP, WSDL and UDDI to Real-world Projects, Springer, 2005.

[29] M. Rebstock, J. Fengel, H. Paulheim, Ontologies-based Business Integration,Springer-Verlag, Berlin Heidelberg, 2008.

[30] G. Antoniou, F. van Harmelen, A Semantic Web Primer, second ed. The MIT Press,Cambridge, Massachusetts, 2008.

[31] H. Wache, T. Vögele, U. Visser, H. Stuckenschmidt, G. Schuster, H. Neumann, et al.,Ontology-based integration of information — a survey of existing approaches, in:H. Stuckenschmidt (Ed.), IJCAI01 Workshop Proc. Ontologies and Information Shar-ing, 2001, pp. 108–117.

[32] A. Pradhan, B. Akinci, C.T. Haas, Formalisms for query capture and data source iden-tification to support data fusion for construction productivity monitoring, AutomatConstr. 20 (2011) 389–398.

[33] V. Alexiev, M. Breu, J.D. Bruijin, D. Fensel, R. Lara, H. Lausen, Information Integrationwith Ontologies: Experiences from an Industrial Showcase, John Wiley & Sons,Chichester, UK, 2005.

[34] J.D. Bruijn, C. Feier, D4.6.1.1 Report on Ontology Mediation for Case Studies V1.SEKT-project Report 2005-08-05, Available at http://www.sekt-project.com/rd/deliverables/wp04/sket-d-4-6-1-1-Ontology%20Mediation%20for%20Case%20Studies%20V1.pdf2005.

[35] F. Manola, E. Miller, R.D.F. Primer, W3C Recommendation, Available at http://www.w3.org/TR/rdf-primer/February 10 2004.

[36] D.L. McGuinness, F. van Harmelen, OWL Web Ontology Language Overview, W3CRecommendation, Available at http://www.w3.org/TR/owl-features/February 102004.

[37] E. Prud'hommeaux, A. Seaborne, SPARQL Query Language for RDF, W3C Recommen-dation, Available at http://www.w3.org/TR/rdf-sparql-query/January 15 2008.

[38] InfraModel2 Specification Base on LandXML v. 1.0, Available at http://cic.vtt.fi/projects/inframodel2.

[39] Maurizio Lenzerini, Data integration: a theoretical perspective, Proceedings of theTwenty-first ACM SIGMOD-SIGACT-SIGART Symposium on Principles of Database Sys-tems (PODS '02)ACM, New York, NY, USA, 2002, pp. 233–246, http://dx.doi.org/10.1145/543613.543644, (http://doi.acm.org/10.1145/543613.543644).

[40] M.M. Kwakye, A Practical Approach toMergingMultidimensional DataModels, Doctor-al Thesis, retrieved August 2, 2013, from http://www.ruor.uottawa.ca/en/bitstream/handle/10393/20457/Mireku_Kwakye_Michael_2011_thesis.pdf?sequence=12011.

[41] K.W. Chau, Ying Cao, M. Anson, Jianping Zhang, Application of data warehouse andDecision Support System in construction management, Autom. Constr. (ISSN:0926-5805) 12 (2) (March 2003) 213–224.

[42] Weiming Shen, Qi Hao, Yunjiao Xue, A loosely coupled system integration approachfor decision support in facility management and maintenance, Autom. Constr.(ISSN: 0926-5805) 25 (August 2012) 41–48.

[43] Novatron Machine Control Systems, retrieved August 2, 2013, from http://www.novatron.eu/en/.

[44] Trimble Machine Control Systems, retrieved August 2, 2013, from http://www.trimble.com/construction/heavy-civil/machine-control/grade-control/.

[45] Topcon Machine Control, retrieved August 2, 2013, from http://www.topconpositioning.com/products/machine-control.

[46] Topcon Sitelink 3D, retrieved August 2, 2013, from http://www.topconpositioning.com/news-events/news/product-news/topcon%E2%80%99s-sitelink-3d-connects-range-grade-control-products.

[47] Trimble Connected Site, retrieved August 2, 2013, from http://connectedsite.com/#3.[48] Caterpillar Fleet Management Solutions, retrieved August 2, 2013, from http://

www.cat.com/technology/fleet-management-solutions.[49] Eero Kasanen, Lukka Kari, Arto Siitonen, The constructive approach in management

accounting research, J. Manag. Account. Res. 5 (1) (1993) 243–263.[50] Protégé, Available at: http://protege.stanford.edu/.[51] OpenRDF.org — Home of Sesame, Available at: http://www.openrdf.org/.[52] S. Harris, A. Seaborne, SPARQL 1.1 query language,W3C Proposed Recommendation,

November 8 2012. (Available at http://www.w3.org/TR/2012/PR-sparql11-query-20121108/).