ArchiInspection Project: Integrated Non Destructive Testing, A Building Information Model Approach

ArchiInspection ProjectIntegrated Non Destructive Testing, A Building Information ModelApproach Farid Mokhtar Noriega1, Victor Garcia Barba2, Jose Antonio Merino3, Jose Julio Zancajo4, Teresa Mostaza Pérez5 1,2,3 Escuela Superior de Arquitectura y Tecnología, Camilo Jose Cela University, Spain, 4,5 Escuela Politécnica Superior de Avila, Salamanca University, Spain 1,2,3 http://www.ucjc.edu, 4,5http://www.usal.es 1 fmokhtar@ucjc.edu, 2vgarcia@ucjc.edu, 3Jamerino@ucjc.edu, 4jzancajo@usal.es, 5 teresamp@usal.es Abstract. A non destructive testing process is becoming a technical need, thousands of buildings and huge urban areas will have to be adapted to restrictive energy-saving standards and sustainability criteria. Analysis and diagnostics are required on a massive scale. Building Information Modeling seems to be the adequate environment to assemble huge amounts of data. At this moment both hardware and software technologies are performing moderately well separately. The challenge is to connect them and in the long run automate data collection and conversion to a unified model that could be maintained during the programmed building life cycle. The aim of this research is to discuss the challenge of NDT hardware and BIM software systems integration and define the basic steps for the best practices to undertake it in a fast and accurate manner as well as to define the present and future connections to be developed. A 3 phase joint research project is proposed here and basic needs are analysed. Many lessons have been learned from field work, data translation and data incompatibilities with many shortcomings being detected. Keywords. BIM; Architectural Non Destructive Testing; Architectural conservation databases; information interoperabilityINTRODUCTIONArchitecture is becoming a strategic asset for com- required extensive rehabilitation or restoration. Thismunities and countries with a rich heritage. Mil- short-term thinking is now obsolete, sustainabilitylions of houses and buildings were erected during is becoming a constant concern and energy-savingthe 20th century when, until recently, energy sav- regulations are imposing strict standards for build-ing and sustainability were not a global concern ing efficiency.(ECOFYS, 2005). Traditionally, buildings were not Architects and building engineers require inputmaintained periodically by their users or owners data to produce accurate diagnostics and to planbut left abandoned till they reached a state that efficiency and conservation protocols (Koo and Van Information/Knowledge Architecture & BIM - eCAADe 29 383

Hoy, 2000). Monitoring and observing building ef- the gathered data and turn it into a feasible analysis ficiency requires scientific methodology and peri- model. Actually, this task requires highly skilled tech- odic updating for rapid intervention to take place. nicians to perform it. Although the final result is a rudi- Till now building pathologies have been inspected mentary BIM representation of a building, skilled ND visually and maintenance and repair proposals for- technicians and proficient BIM users are required. De- mulated on an expertise and personal proficiency fining NDT activities, working protocols and standard- basis rather than on scientific and objective diagno- ization of data organization are required for system- sis. NDT technology is reaching an advanced level of atic execution of NDT tasks by any type of users. NDT accuracy in this field (Maierhofer, 2009). is a collaborative activity that should and will always BIM technology is not only a new formula for rep- be performed by a team of proficient technicians; resenting new building proposals, used as a building connecting the results and defining the big picture is information database, it could also help architects and necessary for building long term corrective measures. builders to represent built buildings and act as a main- Hardware and software should be adapted to the NDT tenance and life cycle monitor to control and measure requirements: ease of use, transportability, accuracy, performance with the support of periodic inspections interoperability, modularity, connectivity and mobil- and integrated monitoring devices. Creating as built ity. Shared and analyzed real world case studies are representation databases might seem easier combin- necessary and a very close collaboration between ing NDT technologies (such as 3D scanners, ultrasonic software, hardware and users is also necessary. testing, Ground Penetrating Radar, thermography and photomodelling) with BIM technology. The latter THE AIM OF THE ARCHIINSPECTION seems to be the adequate platform for creating build- PROJECT ing conservation databases based on the information The aim of the ArchiInspection project is to test the gathered by NDT systems. It is assumed that these interoperability of BIM and NDT technology. The high tech hardware and software solutions could planned activities are: the analysis of some non de- help research teams to accelerate their data collec- structive hardware and software solutions, establish- tion process and simplify the analysis, diagnostics and ing data gathering protocols with available onsite decision making for accurate repair and maintenance technologies, translation and validation, interpreta- activities in cities and urban areas. It is also supposed tion, integration of some data sets and geometric that the major benefit will be in the following stages modelling in BIM software. when periodic inspection is performed and hidden The main questions that drive this research are: anomalies are easily detected; the availability of stan- • Are these technologies capable of reproducing dardized data will help in performing long term plan- exact replicas of existing buildings in commer- ning for new and less aged buildings. Applications cial BIM environments? and long term benefits are huge. • What are the possible weaknesses of these tech- Till now, however, the building data represen- nologies and what are the strategies, processes tation and analysis gathered by NDT has been per- and solutions to be developed in the future? formed by skilled technicians each working indepen- dently, using different types of diagnostic tools, pro- THE COLLABORATIVE PROPOSAL ducing incompatible data sets. Valuable information The University of Salamanca (USAL) and Camilo José about the state of buildings is stored in isolated files Cela University (UCJC) were aware that they lack ex- or unconnected databases. NDT building represen- perience in integrated NDT and BIM interoperability. tation in BIM software is still a promise not a reality, Till now both teams had performed these activities huge post processing work is required to interpret separately and not as an integrated process.384 eCAADe 29 - Information/Knowledge Architecture & BIM

Salamanca University has a specialized team of advanced and non destructive conventional in NDT equipped with data collection technolo- methods. Using a combination of BIM and NDT gies such as 3D-scanning, drone-driven aerial technologies for existing building representa- photography for photogrammetric modelling, tion was a new step forward and a challenge thermal detection and ground detection radar to meet. An analysis framework and work pro- exploration. The USAL team has performed sev- tocols were proposed. Their aim was to develop eral non destructive tests on historic buildings in systematic data collection, information organi- Avila, Spain. zation and data analysis for accurate problem The Camilo Jose Cela research team (arquiT- definition and solution testing (Garcia, et al, ICs) has a long experience in three dimensional 2008). The 210 King Street project [2] was one data representation and use BIM in Building of our references for building our framework of modelling, they developed an as built represen- representing existing buildings with BIM based tation project of a community centre using a mix NDT technologies [FIGURE 1].Figure 1Proposed BIM-NDT interop-erability framework. Information/Knowledge Architecture & BIM - eCAADe 29 385

A two stage plan was designed: 7. Data analysis and solution seeking workshop to • The Stage One aim was to generate a full data be organized set of an existing building and perform an in- 8. An experts meeting and a second test for proto- tegration trial. col testing to be planned • The Stage Two aim was to create a collaborative community based on the original source data On project start, both research teams were aware and research results as an open source informa- that a successful digital NDT project depends on four tion repository. major factors: hardware, software, information and users. Considering the NDT hardware component, Stage One: the equipment within the researchers’ reach was: 3D- In Stage One both teams established a joint collab- scanning, drone-driven aerial photography, photo- orative activity to perform a real NDT case study. A grammetric modelling, thermo graphic detection and one-storey 40 year-old building on the University B ground detection radar. Other existing but not avail- campus outside Madrid’s urban area was selected able NDT technologies can be added in the future. As [FIGURE 2]. The size and situation allowed activity to for the software, the applications used were donated be less restricted (many authorizations and permis- by manufacturers (Quantacad) or were beta solu- sions are required) for this early phase. Initial testing tions available on the web (Photofly) and others were protocols, working conditions, problems, weakness- in-house. The third component was the NDT data. es and needs were defined. Two major raw data groups (vector and raster) were obtained from the hardware-software surveying op- Work was divided into the following steps: erations. Several import-export strategies were tested 1. Roles and schedules to be defined to analyse the filtering, organizing and processing 2. Non Destructive Hardware tweaking activities involved. Most of the information acquired 3. Software evaluation matrix to be built and solu- couldn’t be directly linked to the BIM model. The user, tion selection started the fourth component is also critical in the NDT pro- 4. Data extraction protocols to be established cess. Skilled technicians and proficient BIM users are 5. Data extraction with NDT to be planned in sev- needed for data collection, organization, processing eral stages and running scientific analysis applications. User abil- 6. First attempts for data integration to be per- ity for activity planning, data organization and accu- formed rate data processing is essential. Figure 2 The one story building, sub- ject of the NDT study, real (left) and scanned (right)386 eCAADe 29 - Information/Knowledge Architecture & BIM

The selected NDT [1] technologies were: Other NDT technologies such as thermography1. 3D scanning with a Faro Scanner capable of and photomodeling were also tested. Raster based producing a point cloud data set of information information obtained with the thermography imag- of the existing building ery offered an interesting vision of some of the hid-2. 3D photogrammetric photo modelling based den aspects of the building structure. Although use- on ground and aerial (microdrone) photo- ful as an independent NDT surveying technology, graphs this information was limited. Specialized software3. Thermography for testing building efficiency produced accurate information for the main pur-4. Ground Penetrating Radar analysis for under- pose of the technology: detecting hidden defects ground data extraction. through infrared imagery. Transferring this data to BIM programs isn’t possible; there is no mechanism Results obtained from the first pilot tests were we know of that could map this information overnot valid for BIM modelling; many problems were the walls and installations of the building model.encountered especially during geometry transla- Photomodelling didn’t fulfil the promised resultstion to commercial BIM software. The BIM platforms either. Tested solutions required highly accuraterequired auxiliary plugins to import point cloud photographs, which required special optics andfiles. These conversion techniques were not able to definition. Technical photographic techniques wereturn the information into parametric geometry and required. Models generated with image recognitionbuilding types. BIM operators are required to model techniques are of limited use and are difficult to ex-the raw data to create an idealized building repre- port to BIM programs. The lack of direct conversionsentation. There is no standardized data model to to universal BIM formats such as IFC standards andsupport existing buildings. the limited capacity of BIM software to import non REVIT was the selected BIM platform for the parametric data turned out to be a disadvantage.first attempts of geometric modelling; the 210 King Photomodelling could, however, help in determin-Street project was the reference example that sup- ing visible anomalies and building pathologies forported our research. Some BIM auxiliary applications monitoring and documenting; mapping visual in-and third party solutions are being analyzed for fur- formation could help in conserving the visible com-ther work (IMAGINIT) [3]. During the research an al- ponents of the building and photogrammetry couldternative to point clouds was introduced in the anal- add interesting data to building information.ysis: the Quantapoint laser scanning based on a Faro Connecting raster data sets with other analysisscanner used a different approach [4]. The Quanta- applications and mapping them in BIM models is stillpoint suite of building surveying tools turns REVIT hard and imprecise. Building a unified model is theinto a modelling BIM environment for scanned data. only way to build an integrated view of the complexLaser scanning technology offers accurate, light, information hidden in an existing building. Unfortu-visually clear and relatively small files that don’t re- nately BIM data models have not yet been adaptedquire point servers. Using this technology supposed for interoperability of NDT data and building de-a paradigm shift, and some additional training was scription databases. Different types of data aboutnecessary to use it. building behaviour will remain in parallel data bases Other BIM platforms were not rejected as point until designers solve their association problem withclouds could also be introduced in them. A plan was BIM models.therefore drawn up for testing other BIM platforms A hidden disruptive factor is the site, the build-to analyze their interoperability with NDT technolo- ing surroundings, obstacles and additional ele-gies and data. ments, which in some cases add value to the NDT Information/Knowledge Architecture & BIM - eCAADe 29 387

Table 1 Factors BIM oriented NDT Technologies, ArchiInspection Research Comparative study of NDT Technology 3D Scanning Photomodelling GEO Radar Thermography ArchiInspection NDT technologies’ BIM level of Professional Level Highly skilled Highly skilled Highly skilled Highly skilled integration Training Advanced Intermediate Advanced Advanced Training Needs Obligatory Obligatory Obligatory Obligatory Task Complexity Complex Intermediate Complex Complex Duration Of Work Depends on Depends on Depends on Depends on On Site Building size & com- Building size & com- Site size& complexity Building size & complexity plexity plexity In Office Building size Building size Building size Building size Investment Required Hardware High High High High Software High High High High Personnel Training Intermediate Intermediate Intermediate Intermediate Data Intermediate Intermediate Intermediate Intermediate Hardware-Bim Interop- Low Very Low Very Low Very Low erability Tested Apps QuantaCAD Photomodeler SPIVIEW Build-IR Realworks, ScantoBIM Photoscene editor Problems Encountered New point cloud stan- Obstacles in front of Ground obstacles and Obstacles on site re- dards, Translation building reduce accu- terrain levels reduce duce accuracy Obstacles on site racy or abort process accuracy Bim Interoperability Intermediate Low None Low Evolution Each generation is Through generic Through generic better models models Data Use Modeling Geometric Geometric Modeling Reference Medium Res Medium res template Low res template Decal on site model Decal on building Intgration Level In Bim Low Very low None None Database Captured Data Vector Raster Sound waves Infra red Data Representation Point Cloud Pixels High Res. Images Pixels Low Res. Images Pixels Low Res. Images Pixels Mid Res. Images Most Common Data Ascii, ptc jpg Jpg Jpg Accuracy Intermediate/high Low Low Intermediate/Low Ndt Management High, Areas, references, High, Building & ac- High, Site & activities High, Building & ac- Requirements activities tivities tivities Bim Connection Intermediate, Evolving Low, Undetermined Not connected Not connected Data Life Span Short Short Intermediate Short Change Monitoring Progressive changes Progressive changes Variable Constant change Data Maintenance Each inspection Each inspection Each inspection Each inspection Frequency388 eCAADe 29 - Information/Knowledge Architecture & BIM

survey, and in others add complexity due to unnec- The research team obtained from the workshop essary data. Some NDT activities such as photomod- some important information relating to organiza- elling lose their accuracy and even fail if the site is tion, equipment, software, information and person- full of obstacles (vegetation or fixed elements of any nel [TABLE 2]: type). Photo retouching doesn’t solve the issue: the 1. BIM integration in monument conservation is still software produces incorrect results. Raster based hard and inaccurate, heavily ornamented build- technologies require clean sites and interiors, an ide- ings are difficult to reproduce, sculpted details are alized situation unavailable in the real world. not possible to introduce in parametric models Both teams looked for possible unconventional based on idealized building elements and con- solutions for data integration for long term analysis struction techniques. Buildings with reduced ar- and monitoring of the building life cycle. The ONUMA tistic content are easier to represent and monitor. [5] system, a building life cycle cloud computing in- 2. Systematic observation and analysis is needed, tegration platform, showed great potential. Its open buildings have hidden anomalies and weakness- architecture paradigm could help the data integration es. Many building pathologies are not detected challenge in the future. The vRIS project is a reference with established digital NDT techniques and for real time cloud-based BIM and near-real time infor- proficient technicians are capable of observing mation about the building environment and energy failures hard to detect with scanning or imaging. performance [FIGURE 3]. These kinds of projects will 3. NDT training is hard; inspectors should acquire require the intense support of NDT surveying to de- advanced knowledge in construction tech- tect on site the building anomalies that produce per- niques, material science, photogrammetry, formance anomalies (Lavell, 2010). image processing, geometric modelling, data management, BIM use and programming. PANEL OF EXPERTS: 4. Valuable information is retained in NDT data Due to the multidisciplinary nature of the files and inspection reports. There is no frame- NDT activity and the high level of proficiency that work to assemble it and use it as a reference. this demands, the two research teams interviewed 5. Digital equipment has a limited quality; it does many professionals. A collaborative workshop on not reach the level of precision of analogue NDT data interoperability and integration was held. equipment which is still useful and capable of Some NDT experts attended the event. Discussions capturing relevant information. were held on surveying protocols, equipment use, 6. Conserving digital equipment is a hard task; it software application in documenting, analysing requires special care during handling as well as and diagnostics formulation. constant tweaking and calibration at the startFigure 3vRIS near-real time infor-mation about the buildingenvironment and energy per-formance system , courtesy ofOnuma Inc. Information/Knowledge Architecture & BIM - eCAADe 29 389

of each inspection and surveying task. and manufacturers. Intensive collaboration is 7. Some of the NDT hardware defects relate to its necessary to enhance these emerging applica- closed architecture; adding accessories and mod- tions and technologies. ular components is limited and many cases use- less. Upgrading is a missing property in such ex- Stage Two: pensive equipment. Hardware should be able to The USAL and UCJC teams agreed to create a wiki integrate different surveying technologies, which repository to share expert opinions, raw NDT data, would allow users to perform standardized tests. and test results for public research under the cre- 8. NDT equipment produces heterogeneous infor- ative commons license. This contribution could help mation classified into three major data sets: vec- researchers, hardware manufacturers and software tor, raster and numeric. Connecting all of them developers who are interested in BIM and NDT data in a unified BIM model is required. For the time interoperability and integration and help develop fur- being integration is relatively easy in pairs, the ther research. The repository could be used as a sand- challenge is to connect them simultaneously. box to test solutions with the help of the community. 9. Digital photography based raster data collection Complex and complete examples could be is still in its early stages with regard to the infor- added in the future, these could help in different mation it offers compared to traditional tech- ways: understanding the whole non destructive data niques. Its major weaknesses are Optics (lens ac- acquisition process, simplifying data extraction accuracy) and data capturing (CCD precision). tivities for professionals, easy interpretation of the 10. NDT protocols are needed; monitoring activi- output data, accurate representation of building ties, data collection, database creation, analysis information and analysis and performance compari- methodologies and systematic documentation son of building life cycles. of actions are essential. Sharing this informa- ArchiInspection collaborative networking will tion regarding problems and their solutions benefit researchers who are looking for complete could provide valuable input for researchers data sets to develop their research work. Full sets Table 2 Component Proposals Selected Recommendations Hardware Modular components of equipment proposed by the experts team Easy upgrade for longer hardware life cycle NDT equipment integration (two or three devices working together) Data collection accuracy Software Capable of directly transferring and converting data to BIM standards Bidirectional BIM Data Base connection Lifecycle DB Standard data models BIM Data standardized transfer Open platform design for easy integration of future generations of NDT technologies Ability to store relevant physical information and represent it graphically and numerically Ability to map data over simplified building geometry Represent and store as built information for analysis Personnel Standardized NDT protocols BIM Data transfer Accredited Training Data Standardization of NDT data for long term data conservation and analysis390 eCAADe 29 - Information/Knowledge Architecture & BIM

of NDT samples are not available due to the copy- interoperability of all the systems required to devel- right conditions imposed by the clients who finance op NDT surveys [TABLE 3]. them. Sharing data is crucial, learning how to cap- ture, clean and organize the information is critical. MAJOR RESEARCH FINDINGS Some manufacturers are trying to introduce their 1. Extensive work should be performed to en- own standards and translating information from uni- able NDT technologies to become an accurate versal formats to their proprietary file system. Non input device for BIM software. A modular and standardized data formats could harm the informa- flexible design paradigm is required to keep it tion owners, taking into consideration the informa- functional and technically accurate over a long tion life span. Many workarounds are required to time frame. change from one standard to another and probably 2. Commercial BIM needs to standardize its data data will be lost. models and adopt sustainable long-term open A wish list was created to help the all the architecture that can be maintained during the agents involved in the process especially hardware building life cycle by consecutive management and software developers in their quest for a better teams and different levels of users.Table 3 NDT STATE Actual (Present) Expected (Future)NDT present situation anddesired future Hardware Highly specialized, independent Specialized, multi-functional Software Highly specialized, independent and uncon- Specialized, integrated and connected nected Architecture Closed, platform dependent Open, platform independent Location of NDT information Desktop Mixed, desktop and cloud and tools Interoperability of Low level of integration High level of integration hardware, software and data Life cycle Orientation Short term, Long term Time span Valid for a limited time, Valid for long time Await for next inspection 7/24 monitored Information use Non accumulative Accumulative Data structures Fragmented, Integrated, stored in independent databases organized in connected databases Data type Independent non standardized standardized Data structure Fragile Robust Information sharing User owned, closed, unshared Team Owned, open, shared Durability Short durability Long durability Practice Periodic Continuous User knowledge impact User dependent User independent Level of user knowledge Expert Proficient user User participation Physical, in site Virtual, online monitoring Information/Knowledge Architecture & BIM - eCAADe 29 391

3. Some emerging surveying techniques like laser García Barba, Víctor and Mokhtar Noriega, Farid scanning or cloud-based BIM could dramati- 2008, Análisis y diagnóstico de la eficiencia ener- cally enhance the NDT surveying process. gética de edificios existentes mediante sistemas 4. Hardware and software are designed as inde- no destructivos, CONAMA 9, Madrid. pendent systems; connecting and performing Koo, WL and Van Hoy, T 2000, Determining the Eco- accurate building life cycle monitoring is a big nomic Value of Preventive Maintenance, Jones challenge. Lang LaSalle, Chicago. 5. BIM and NDT manufacturers should join ef- Lavelle, Michael 2010, Virtual Building Energy Man- forts and produce standard data for represent- agement, Moving to Cloud-based Building Energy ing and connecting surveying and monitoring Management, AutomatedBuildings.com, http:// work. This would ease the analysis and compari- www.automatedbuildings.com/news/may10/ son process for long-term sustainable design. articles/lavelleenergy/100427104606lavelle. 6. Building conservation and maintenance could htm. be an extended professional service and a good Maierhofer, C 2009, Non destructive testing, Heverlee, job opportunity for many EU countries. On the http://precomos.org/index.php/library/library_ other hand training programs should be put extended/non-destructive_testing_ndt/. into practice due to the complexity of the activ- [1] Photomodeler Scanner http://www.photomod- ity and the level of proficiency required. Users eler.com/; Flir thermographic camera http:// should be proficient technicians and advanced www.flir.com/; Faro Laser Scanner http://www. BIM users to be able to transfer the captured faro.com/; structure scan http://www.geophysi- data to the basic model. All the activities are cal.com/whatsnewatgssi.htm. critical; errors could result in misleading judg- [2] http://www.digital210king.org/index.php ments. NDT is not a one-man task; it is a team- [3] http://s3.amazonaws.com/ScanToBIM/ScanTo- based activity, with each member needing to BIM2012.zip be proficient in his field. The human factor is http://www.youtube.com/watch?v=bp4J- still critical and required for controlling the in- a_3188&feature=related put and output quality at all stages. [4] Quantapoint http://www.quantapoint.com/re- 7. BIM and NDT integration and interoperabil- sources/whitepapers/ ity require normalized protocols, which could [5] Onuma cloud computing Integrated BIM http:// guarantee the effectiveness of the inspection www.onuma.com/. and surveying processes and accurate analysis of the building in each revision period. The lack of adequate feedback reduces the value of the tests performed. 8. There is widely available expertise to draw on in some European countries to contribute to our proposed ArchiInspection repository but there remain many problems to identify and to solve in the future. REFERENCES ECOFYS, 2005, Cost effective retrofit in buildings, ECO- FYS, Cologne.392 eCAADe 29 - Information/Knowledge Architecture & BIM

ArchiInspection Project: Integrated Non Destructive Testing, A Building Information Model Approach

by Farid Mokhtar Noriega on Dec 26, 2011

Accessibility

Categories

Tags

Upload Details

Usage Rights

1 Embed 297

Statistics

ArchiInspection Project: Integrated Non Destructive Testing, A Building Information Model Approach — Document Transcript