International Construction Specialty Conference of the Canadian Society for Civil Engineering (ICSC) (5th : 2015)

A smart mobile app for site inspection and documentation Nguyen, Long D.; Koufakou, Anna; Mitchell, Colin Jun 30, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52660-Nguyen_L_et_al_ICSC15_119_Smart_App_Site_Inspection.pdf [ 3.01MB ]
52660-Nguyen_L_et_al_ICSC15_119_Smart_App_Site_Inspection_slides.pdf [ 10.8MB ]
Metadata
JSON: 52660-1.0076358.json
JSON-LD: 52660-1.0076358-ld.json
RDF/XML (Pretty): 52660-1.0076358-rdf.xml
RDF/JSON: 52660-1.0076358-rdf.json
Turtle: 52660-1.0076358-turtle.txt
N-Triples: 52660-1.0076358-rdf-ntriples.txt
Original Record: 52660-1.0076358-source.json
Full Text
52660-1.0076358-fulltext.txt
Citation
52660-1.0076358.ris

Full Text

5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   A SMART MOBILE APP FOR SITE INSPECTION AND DOCUMENTATION Long D. Nguyen1, Anna Koufakou2 and Colin Mitchell3 1 Department of Environmental and Civil Engineering, Florida Gulf Coast University, USA 2 Department of Software Engineering, Florida Gulf Coast University, USA 3 Department of Software Engineering, Florida Gulf Coast University, USA Abstract: Extensive time and effort are spent on inspecting and documenting construction defects for facilities under construction or in use. The goal of this study is to provide construction engineers and experts with a smart mobile application to efficiently record and document construction defects. This study first investigated typical inspection processes, data, and reports which were used in site inspections in practice. An Android-based mobile application called InSite Inspector (Intelligent Site Inspector) was then developed to facilitate site inspection and documentation. This smart app is able to: (i) take images and catalog details of construction defects such as defect types, construction trades, building components, and date and time; (ii) automatically locate defects using the global positioning system (GPS); and (iii) produce various types of reports for different inspection purposes such as punch lists and defect reports. The inspectors are allowed to customize the app features such as text entries, type of data to be recorded and/or reported to meet their specific inspection requirements. Additionally, the InSite Inspector allows engineers and experts to manage different construction sites, inspections, and to keep track of past and ongoing inspections. Finally, the app can be used to save, email, or upload the resulting reports to cloud-based repositories. An inspection case study was used to demonstrate the application and utilities of this development. The InSite Inspector is expected to significantly streamline site inspection and documentation processes. 1 INTRODUCTION In the construction industry, extensive time and effort are spent on inspecting and documenting construction defects for buildings under construction or in use. When a building project is almost completed, project parties conduct site inspections to prepare punch lists and repair defects before the building is handed over to its owner. For an occupied building, the owner, designer, contractor, and other parties may conduct site inspections when substantial defects are identified. Construction defects are therefore a major concern for all stakeholders, including owners, designers, builders, and insurance companies. Literature indicates the ratification cost of defects in between 2% and 12.4% of the construction cost (Lundkvist et al., 2014). With the construction value of $975 billion in 2014 (Census Bureau, 2015), the ratification of defects could cost the U.S. taxpayers in a range of $20 – $120 billion last year. In addition to the cost of repairs, defects are one of the major sources of costly and lengthy construction dispute and litigation in the U.S. In a construction defect dispute or litigation, both claimant and defendant inspect the property and report their findings and results of the inspection. Unfortunately, the current inspection, defect classification, and documentation process is time-consuming and expensive.  119-1 The Android-based platform and JAVA/XML technology were adopted for developing the first and current version of InSite Inspector. As there are various platform versions of Android available (Figure 3), the evaluation and selection of the right platform was conducted in this study because choosing the right version target is critical. Choosing too high versions would result in more phones being incompatible, isolate more potential users, and hence decrease the user base, while choosing too low versions would result in the app missing important built-in features of the newer versions. In other words, choosing the right platform version is the trade-off of increasing the user base and utilizing newer and better features.  Figure 3: Distribution of the Android platform versions in early January 2015 Figure 3 displays the percentage of devices running different versions of the Android platform in early January 2015 (Android, 2015). The shares of these versions were 0.4% (2.2, API level 8), 7.8% (2.3.x, API level 10), 6.7% (4.0.x, API level 15), 19.2% (4.1.x, API level 16), 20.3% (4.2.x, API level 17), 6.5% (4.3, API level 18), and 39.1% (4.4, API level 19) (Figure 3). The current InSite Inspector employs Android 4.1.x version with application programming interface (API) level 16 that was released in July 2012. It should be noted that Android versions are backwards compatible. That is, Android version 4.1.x with API level 16 is compatible with its own version and newer versions (4.2.x, API level 17; 4.3, API level 18; and 4.4, API level 19). The Android version 4.1.x with API level 16 was selected for InSite Inspector because it is relatively new while it supports 85.1% of current devices (Figure 3). Creating InSite Inspector involved designing and developing the user interface, with capability of taking and storing photos, as well as text entry in form-type fields. Multiple site inspection tests were conducted to evaluate and improve InSite Inspector. Finally, the current version was used and tested in inspecting the envelope of an academic building on the authors’ campus.     4 DEVELOPMENT OF THE INSITE INSPECTOR 4.1 Overview of the InSite Inspector A few site inspection applications are available in different platforms. Perhaps the most related app was Defects app. The Defects app required iOS 6.0 or later (Contractors Apps, 2015). While this app can generate reports in PDF format, it seemed not able to generate a more flexible and customizable format, such as csv (comma-separated values) files. The Defects app also seemed not to allow to associate the responsible parties with defects. InSite Inspector includes those features not available in the current apps 0.4% 7.8% 6.7% 19.2% 20.3% 6.5% 39.1% 85.1% Version 2.2, API Level 8 Version 2.3.x, API Level 10Version 4.0.x, API Level 15 Version 4.1.x, API Level 16Version 4.2.x, API Level 17 Version 4.3, API Level 18Version 4.4, API Level 19119-4 and other features that assist efficient site inspection. Figure 4 displays a simplified Unified Modeling Language (UML) class diagram of InSite Inspector.   Main Menu Defect ScreenInspection ScreenConcerned Party ScreenReport ScreenLocation ScreenDatabase Data File ManagementGPSImage HandlingAndroid Activities Figure 4: Simplified UML class diagram of InSite Inspector Within Android activities, there are six classes, including Main Menu, Inspection, Location, Concerned Party, Defect, and Report (Figure 4). These Android activities interact with the built-in global positioning system (GPS) package to locate site location and defects. They interact with the Image Handling package when a defect image is taken by the camera. The Image Handling is used for image subsampling. That is, InSite Inspector will automatically estimate the memory usage of loading a defect and resize the image file and/or change resolution if necessary. Finally, the Android Activities interact with Data, Database and File Management packages for restoring and reporting inspection data and information. A SQLite database was developed to store the inspection data. It stores these data as “.db” files hidden on the system. The File Management package is used to write the inspection data (e.g., defect images, locations) to or read them from internal/external device storage. The next two sections describe the two major modules, namely Site Inspection and Documentation and Reporting in details through a case study. The case study was an inspection of the envelope of an academic building on the authors’ campus in January, 2015. 4.2 Site Inspection Module The site inspection module includes processes used by the site engineer or manager in InSite Inspector in order to collect and store inspection data. Figure 5 demonstrates four screenshots of the major processes for site inspection with InSite Inspector. An inspector may create a new inspection, continue previous inspection, or generate reports from previous inspections. These are features in the Main Menu class (not shown in Figure 5). If an inspector starts a new inspection, he/she may first assign his/her inspection name, location, inspector profile.      (a)             (b)     (c)      (d) Figure 5: Site inspection module 119-5 The inspection can proceed by adding a new defect (Figure 5a). The defect recording process allows an inspector to take multiple images of a defect and to choose up to two images representing the defect (Figure 5b). An inspector may provide entries such as type and location of defect (Figure 5b) and additional defect information (not shown). These entries can be recorded either by text or voice to text with the Google Voice typing feature. InSite Inspector will automatically capture the inspection time and date of the defect and locate the defect location using GPS. It utilizes the built-in GPS for Android in two different ways for two scenarios. If the device and hence InSite Inspector search and find at least four GPS satellites of the available 24, a defect can be located by all three: latitude, longitude, and altitude. The “Window sill crack” defect in Figure 5b was located by these three coordinates. If not, InSite Inspector can determine a defect location through a network provider based on availability of cell towers or WiFi access points. In this case, only latitude and longitude of the defect will be determined. As the current defect management systems tended to error (Park et al., 2013, Lundkvist et al., 2014), the proposed automatic positioning of defects will eliminate this drawback.  InSite Inspector allows site engineers and managers to associate a defect with concerned and/or responsible parties. Defects occurred due to many reasons such as design, specification, materials, workmanship, and managerial errors and malpractice. The reasons are all related to human errors and more specially, parties involved throughout the project life cycle, including owners, designers, general contractors, subcontractors, suppliers, and users. During or after an inspection, site inspectors can add parties involved (Figure 5c) and relate them with any defect already investigated (Figure 5d). For example, D&W Sub C installed windows for the building investigated and thus might be associated with the “window sill crack” defect (Figure 5d). The association of defects and concerned parties facilitates engineers and managers in customizing inspection reports as discussed in the next section.   4.3 Documentation and Reporting Module Figure 6 presents the selected snapshots of the inspection documentation and reporting module. An inspector can customize inspection reports in different ways. All or a subset of defects can be easily chosen before generating reports (Figure 6a). For example, an inspector may only select defects related to stucco work (Figure 6a) for a report and send it to Stucco Sub B (Figure 5c). For a certain defect, an inspector may also choose information to include or not in an inspection report, such as defect type, location, associated parties (Figure 6b).        (a)             (b)     (c)      (d) Figure 6: Documentation and reporting module After customizing information for the report, a few export options are available for the report (Figures 6c and 6d). The output types include the portable document format (pdf) files or csv files where inspectors 119-6 can further edit and format the report (Figure 6c). Finally, inspectors have different options to export their report. They can save it in their current device, email to related parties, or upload to clouded-based repositories such as Dropbox (Figure 6d).  Figure 7 presents portions of two reports, one created with a specialized software on a PC by a professional inspector for an actual project in California (Figure 7a) and one created by InSite Inspector in the inspection case study in Florida (Figure 7b). A professional inspection report such as one shown in Figure 7a may take hours or days to be prepared. InSite Inspector efficiently creates professional-like reports (Figure 7b) with more details, especially those automatically captured on site in a few seconds or a few minutes depending on the magnitude of the inspection and the capacity of the mobile device. Substantial time savings and more accurate defect information can be achieved with InSite Inspector. (a) Professional inspection report       (b) InSite Inspector generated report  Figure 7: Real-world vs. InSite Inspector generated reports 5 LIMITATIONS AND FUTURE WORK There are a few limitations of the present version of InSite Inspector. First, it is currently available only on Android although the share of Android was 84% as of the third quarter of 2014 (Barrie, 2014). Second, to position accurately the location and including all latitude, longitude, and altitude of a defect, an inspection device has to detect at least four GPS satellites. From our case study, this could only work for locating defects outside of the building (e.g., exterior walls, roofs, windows). For defects inside the building, the approximation of the defect location was retrieved from the cellphone or WiFi network provider and only included latitude and longitude. Future work will include employing techniques to overcome the above limitations and, especially in the longer term, utilize image processing and machine learning techniques for automatic defect classification (Step 2’, Figure 2). The ultimate goal of this ongoing research is to develop InSite Inspector that allows engineers and managers to take pictures of defects and automatically classify the types of defects.  Image processing and machine learning algorithms and techniques (e.g., Fast Fourier Transforms (FFT), Butterworth filters, Artificial Neural Networks (ANN)) may make the automatic classification of construction defects possible. Future research will also utilize the defect classifications available for different types of construction such as industrial construction (Fayek et al., 2003), timber construction (Johnson and Meiling, 2009) and bridge construction (Cheng and Leu, 2011). 6 CONCLUSIONS An Android-based smart mobile app named InSite Inspector was developed for site inspection and documentation. It is degined to be effectively used for site inspection in both construction and operation phases. A case study involving the inspection of the envelope of a building was conducted with InSite Inspector. This inspection demonstrated that InSite Inspector facilitates engineers and managers to record defect information, parties involved, and automatically locates defects using GPS. Inspectors can 119-7 customize report information and formats and generate reports and documents for different purposes. The documentation and reporting process with InSite Inspector can significantly save site engineers’ and managers’ time and eliminate errors that potentially occur in current site inspection practices. Future work will include improving the accurate positioning of the defects and utilizing image processing and machine learning techniques to automatically classify defects captured on site. References  Android 2015. Dashboards. Retrieved from http://developer.android.com/about/dashboards/index.html (last accessed on 27 January 2015). Barrie, J. 2014. After Years of Losses Apple is Finally Clawing Back Some Market Share against Android. Business Insider. Retrieved from http://www.businessinsider.com/apple-v-android-market-share-data-2014-12 (last accessed on 29 January 2015). Census Bureau 2015. November 2014 Construction at $975.0 Billion Annual Rate. U.S. Department of Commerce, Washington, D.C. Cheng, Y.-M. and Leu, S.-S. 2011. Integrating Data Mining with KJ Method to Classify Bridge Construction Defects. Expert Systems with Applications, 38(6): 7143-7150.  Contractors Apps 2015. Defects. Retrieved from http://www.contractorsapps.com/our-apps/defects (last accessed on 27 January 2015). Cox, S., Perdomo, J., and Thabet, W. 2002. Construction Field Data Inspection Using Pocket PC Technology. In Proceedings of International Council for Research and Innovation in Building and Construction, CIB W78 Conference, A  Dong, A., Maher, M. L., Kim, M. J., Gu, N., and Wang, X. 2009. Construction Defect Management Using a Telematic Digital Workbench. Automation in Construction, 18(6): 814-824. Fayek, A.R., Dissanayake, M. and Campero, O. 2003. Measuring and Classifying Construction Field Rework: A Pilot Study. Research Report, University of Alberta, Edmonton, Canada. Gordon, C., Akinci, B., and Garrett, J. H. 2007. Formalism for Construction Inspection Planning: Requirements and Process Concept. Journal of Computing in Civil Engineering, 21(1): 29–38. Johnsson, H. and Meiling, J. H. 2009. Defects in Offsite Construction: Timber Module Prefabrication. Construction Management and Economics, 27(7): 667–81. Kwon, O. S., Park, C. S., and Lim, C. R. 2014. A Defect Management System for Reinforced Concrete Work Utilizing BIM, Image-Matching and Augmented Reality. Automation in Construction, 46: 74-81. Lundkvist, R., Meiling, J. H., and Sandberg, M. 2014. A Proactive Plan-Do-Check-Act Approach to Defect Management Based on a Swedish Construction Project. Construction Management and Economics, 32(11): 1051-1065. Park, C.-S., Lee, D.-Y., Kwon, O.-S., and Wang, X. 2013. A Framework for Proactive Construction Defect Management Using BIM, Augmented Reality and Ontology-based Data Collection Template. Automation in Construction, 33: 61-71. Wang, L.-C. 2008. Enhancing Construction Quality Inspection and Management using RFID Technology. Automation in Construction, 17: 467-479.  119-8  5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   A SMART MOBILE APP FOR SITE INSPECTION AND DOCUMENTATION Long D. Nguyen1, Anna Koufakou2 and Colin Mitchell3 1 Department of Environmental and Civil Engineering, Florida Gulf Coast University, USA 2 Department of Software Engineering, Florida Gulf Coast University, USA 3 Department of Software Engineering, Florida Gulf Coast University, USA Abstract: Extensive time and effort are spent on inspecting and documenting construction defects for facilities under construction or in use. The goal of this study is to provide construction engineers and experts with a smart mobile application to efficiently record and document construction defects. This study first investigated typical inspection processes, data, and reports which were used in site inspections in practice. An Android-based mobile application called InSite Inspector (Intelligent Site Inspector) was then developed to facilitate site inspection and documentation. This smart app is able to: (i) take images and catalog details of construction defects such as defect types, construction trades, building components, and date and time; (ii) automatically locate defects using the global positioning system (GPS); and (iii) produce various types of reports for different inspection purposes such as punch lists and defect reports. The inspectors are allowed to customize the app features such as text entries, type of data to be recorded and/or reported to meet their specific inspection requirements. Additionally, the InSite Inspector allows engineers and experts to manage different construction sites, inspections, and to keep track of past and ongoing inspections. Finally, the app can be used to save, email, or upload the resulting reports to cloud-based repositories. An inspection case study was used to demonstrate the application and utilities of this development. The InSite Inspector is expected to significantly streamline site inspection and documentation processes. 1 INTRODUCTION In the construction industry, extensive time and effort are spent on inspecting and documenting construction defects for buildings under construction or in use. When a building project is almost completed, project parties conduct site inspections to prepare punch lists and repair defects before the building is handed over to its owner. For an occupied building, the owner, designer, contractor, and other parties may conduct site inspections when substantial defects are identified. Construction defects are therefore a major concern for all stakeholders, including owners, designers, builders, and insurance companies. Literature indicates the ratification cost of defects in between 2% and 12.4% of the construction cost (Lundkvist et al., 2014). With the construction value of $975 billion in 2014 (Census Bureau, 2015), the ratification of defects could cost the U.S. taxpayers in a range of $20 – $120 billion last year. In addition to the cost of repairs, defects are one of the major sources of costly and lengthy construction dispute and litigation in the U.S. In a construction defect dispute or litigation, both claimant and defendant inspect the property and report their findings and results of the inspection. Unfortunately, the current inspection, defect classification, and documentation process is time-consuming and expensive.  119-1 The Android-based platform and JAVA/XML technology were adopted for developing the first and current version of InSite Inspector. As there are various platform versions of Android available (Figure 3), the evaluation and selection of the right platform was conducted in this study because choosing the right version target is critical. Choosing too high versions would result in more phones being incompatible, isolate more potential users, and hence decrease the user base, while choosing too low versions would result in the app missing important built-in features of the newer versions. In other words, choosing the right platform version is the trade-off of increasing the user base and utilizing newer and better features.  Figure 3: Distribution of the Android platform versions in early January 2015 Figure 3 displays the percentage of devices running different versions of the Android platform in early January 2015 (Android, 2015). The shares of these versions were 0.4% (2.2, API level 8), 7.8% (2.3.x, API level 10), 6.7% (4.0.x, API level 15), 19.2% (4.1.x, API level 16), 20.3% (4.2.x, API level 17), 6.5% (4.3, API level 18), and 39.1% (4.4, API level 19) (Figure 3). The current InSite Inspector employs Android 4.1.x version with application programming interface (API) level 16 that was released in July 2012. It should be noted that Android versions are backwards compatible. That is, Android version 4.1.x with API level 16 is compatible with its own version and newer versions (4.2.x, API level 17; 4.3, API level 18; and 4.4, API level 19). The Android version 4.1.x with API level 16 was selected for InSite Inspector because it is relatively new while it supports 85.1% of current devices (Figure 3). Creating InSite Inspector involved designing and developing the user interface, with capability of taking and storing photos, as well as text entry in form-type fields. Multiple site inspection tests were conducted to evaluate and improve InSite Inspector. Finally, the current version was used and tested in inspecting the envelope of an academic building on the authors’ campus.     4 DEVELOPMENT OF THE INSITE INSPECTOR 4.1 Overview of the InSite Inspector A few site inspection applications are available in different platforms. Perhaps the most related app was Defects app. The Defects app required iOS 6.0 or later (Contractors Apps, 2015). While this app can generate reports in PDF format, it seemed not able to generate a more flexible and customizable format, such as csv (comma-separated values) files. The Defects app also seemed not to allow to associate the responsible parties with defects. InSite Inspector includes those features not available in the current apps 0.4% 7.8% 6.7% 19.2% 20.3% 6.5% 39.1% 85.1% Version 2.2, API Level 8 Version 2.3.x, API Level 10Version 4.0.x, API Level 15 Version 4.1.x, API Level 16Version 4.2.x, API Level 17 Version 4.3, API Level 18Version 4.4, API Level 19119-4 and other features that assist efficient site inspection. Figure 4 displays a simplified Unified Modeling Language (UML) class diagram of InSite Inspector.   Main Menu Defect ScreenInspection ScreenConcerned Party ScreenReport ScreenLocation ScreenDatabase Data File ManagementGPSImage HandlingAndroid Activities Figure 4: Simplified UML class diagram of InSite Inspector Within Android activities, there are six classes, including Main Menu, Inspection, Location, Concerned Party, Defect, and Report (Figure 4). These Android activities interact with the built-in global positioning system (GPS) package to locate site location and defects. They interact with the Image Handling package when a defect image is taken by the camera. The Image Handling is used for image subsampling. That is, InSite Inspector will automatically estimate the memory usage of loading a defect and resize the image file and/or change resolution if necessary. Finally, the Android Activities interact with Data, Database and File Management packages for restoring and reporting inspection data and information. A SQLite database was developed to store the inspection data. It stores these data as “.db” files hidden on the system. The File Management package is used to write the inspection data (e.g., defect images, locations) to or read them from internal/external device storage. The next two sections describe the two major modules, namely Site Inspection and Documentation and Reporting in details through a case study. The case study was an inspection of the envelope of an academic building on the authors’ campus in January, 2015. 4.2 Site Inspection Module The site inspection module includes processes used by the site engineer or manager in InSite Inspector in order to collect and store inspection data. Figure 5 demonstrates four screenshots of the major processes for site inspection with InSite Inspector. An inspector may create a new inspection, continue previous inspection, or generate reports from previous inspections. These are features in the Main Menu class (not shown in Figure 5). If an inspector starts a new inspection, he/she may first assign his/her inspection name, location, inspector profile.      (a)             (b)     (c)      (d) Figure 5: Site inspection module 119-5 The inspection can proceed by adding a new defect (Figure 5a). The defect recording process allows an inspector to take multiple images of a defect and to choose up to two images representing the defect (Figure 5b). An inspector may provide entries such as type and location of defect (Figure 5b) and additional defect information (not shown). These entries can be recorded either by text or voice to text with the Google Voice typing feature. InSite Inspector will automatically capture the inspection time and date of the defect and locate the defect location using GPS. It utilizes the built-in GPS for Android in two different ways for two scenarios. If the device and hence InSite Inspector search and find at least four GPS satellites of the available 24, a defect can be located by all three: latitude, longitude, and altitude. The “Window sill crack” defect in Figure 5b was located by these three coordinates. If not, InSite Inspector can determine a defect location through a network provider based on availability of cell towers or WiFi access points. In this case, only latitude and longitude of the defect will be determined. As the current defect management systems tended to error (Park et al., 2013, Lundkvist et al., 2014), the proposed automatic positioning of defects will eliminate this drawback.  InSite Inspector allows site engineers and managers to associate a defect with concerned and/or responsible parties. Defects occurred due to many reasons such as design, specification, materials, workmanship, and managerial errors and malpractice. The reasons are all related to human errors and more specially, parties involved throughout the project life cycle, including owners, designers, general contractors, subcontractors, suppliers, and users. During or after an inspection, site inspectors can add parties involved (Figure 5c) and relate them with any defect already investigated (Figure 5d). For example, D&W Sub C installed windows for the building investigated and thus might be associated with the “window sill crack” defect (Figure 5d). The association of defects and concerned parties facilitates engineers and managers in customizing inspection reports as discussed in the next section.   4.3 Documentation and Reporting Module Figure 6 presents the selected snapshots of the inspection documentation and reporting module. An inspector can customize inspection reports in different ways. All or a subset of defects can be easily chosen before generating reports (Figure 6a). For example, an inspector may only select defects related to stucco work (Figure 6a) for a report and send it to Stucco Sub B (Figure 5c). For a certain defect, an inspector may also choose information to include or not in an inspection report, such as defect type, location, associated parties (Figure 6b).        (a)             (b)     (c)      (d) Figure 6: Documentation and reporting module After customizing information for the report, a few export options are available for the report (Figures 6c and 6d). The output types include the portable document format (pdf) files or csv files where inspectors 119-6 can further edit and format the report (Figure 6c). Finally, inspectors have different options to export their report. They can save it in their current device, email to related parties, or upload to clouded-based repositories such as Dropbox (Figure 6d).  Figure 7 presents portions of two reports, one created with a specialized software on a PC by a professional inspector for an actual project in California (Figure 7a) and one created by InSite Inspector in the inspection case study in Florida (Figure 7b). A professional inspection report such as one shown in Figure 7a may take hours or days to be prepared. InSite Inspector efficiently creates professional-like reports (Figure 7b) with more details, especially those automatically captured on site in a few seconds or a few minutes depending on the magnitude of the inspection and the capacity of the mobile device. Substantial time savings and more accurate defect information can be achieved with InSite Inspector. (a) Professional inspection report       (b) InSite Inspector generated report  Figure 7: Real-world vs. InSite Inspector generated reports 5 LIMITATIONS AND FUTURE WORK There are a few limitations of the present version of InSite Inspector. First, it is currently available only on Android although the share of Android was 84% as of the third quarter of 2014 (Barrie, 2014). Second, to position accurately the location and including all latitude, longitude, and altitude of a defect, an inspection device has to detect at least four GPS satellites. From our case study, this could only work for locating defects outside of the building (e.g., exterior walls, roofs, windows). For defects inside the building, the approximation of the defect location was retrieved from the cellphone or WiFi network provider and only included latitude and longitude. Future work will include employing techniques to overcome the above limitations and, especially in the longer term, utilize image processing and machine learning techniques for automatic defect classification (Step 2’, Figure 2). The ultimate goal of this ongoing research is to develop InSite Inspector that allows engineers and managers to take pictures of defects and automatically classify the types of defects.  Image processing and machine learning algorithms and techniques (e.g., Fast Fourier Transforms (FFT), Butterworth filters, Artificial Neural Networks (ANN)) may make the automatic classification of construction defects possible. Future research will also utilize the defect classifications available for different types of construction such as industrial construction (Fayek et al., 2003), timber construction (Johnson and Meiling, 2009) and bridge construction (Cheng and Leu, 2011). 6 CONCLUSIONS An Android-based smart mobile app named InSite Inspector was developed for site inspection and documentation. It is degined to be effectively used for site inspection in both construction and operation phases. A case study involving the inspection of the envelope of a building was conducted with InSite Inspector. This inspection demonstrated that InSite Inspector facilitates engineers and managers to record defect information, parties involved, and automatically locates defects using GPS. Inspectors can 119-7 customize report information and formats and generate reports and documents for different purposes. The documentation and reporting process with InSite Inspector can significantly save site engineers’ and managers’ time and eliminate errors that potentially occur in current site inspection practices. Future work will include improving the accurate positioning of the defects and utilizing image processing and machine learning techniques to automatically classify defects captured on site. References  Android 2015. Dashboards. Retrieved from http://developer.android.com/about/dashboards/index.html (last accessed on 27 January 2015). Barrie, J. 2014. After Years of Losses Apple is Finally Clawing Back Some Market Share against Android. Business Insider. Retrieved from http://www.businessinsider.com/apple-v-android-market-share-data-2014-12 (last accessed on 29 January 2015). Census Bureau 2015. November 2014 Construction at $975.0 Billion Annual Rate. U.S. Department of Commerce, Washington, D.C. Cheng, Y.-M. and Leu, S.-S. 2011. Integrating Data Mining with KJ Method to Classify Bridge Construction Defects. Expert Systems with Applications, 38(6): 7143-7150.  Contractors Apps 2015. Defects. Retrieved from http://www.contractorsapps.com/our-apps/defects (last accessed on 27 January 2015). Cox, S., Perdomo, J., and Thabet, W. 2002. Construction Field Data Inspection Using Pocket PC Technology. In Proceedings of International Council for Research and Innovation in Building and Construction, CIB W78 Conference, A  Dong, A., Maher, M. L., Kim, M. J., Gu, N., and Wang, X. 2009. Construction Defect Management Using a Telematic Digital Workbench. Automation in Construction, 18(6): 814-824. Fayek, A.R., Dissanayake, M. and Campero, O. 2003. Measuring and Classifying Construction Field Rework: A Pilot Study. Research Report, University of Alberta, Edmonton, Canada. Gordon, C., Akinci, B., and Garrett, J. H. 2007. Formalism for Construction Inspection Planning: Requirements and Process Concept. Journal of Computing in Civil Engineering, 21(1): 29–38. Johnsson, H. and Meiling, J. H. 2009. Defects in Offsite Construction: Timber Module Prefabrication. Construction Management and Economics, 27(7): 667–81. Kwon, O. S., Park, C. S., and Lim, C. R. 2014. A Defect Management System for Reinforced Concrete Work Utilizing BIM, Image-Matching and Augmented Reality. Automation in Construction, 46: 74-81. Lundkvist, R., Meiling, J. H., and Sandberg, M. 2014. A Proactive Plan-Do-Check-Act Approach to Defect Management Based on a Swedish Construction Project. Construction Management and Economics, 32(11): 1051-1065. Park, C.-S., Lee, D.-Y., Kwon, O.-S., and Wang, X. 2013. A Framework for Proactive Construction Defect Management Using BIM, Augmented Reality and Ontology-based Data Collection Template. Automation in Construction, 33: 61-71. Wang, L.-C. 2008. Enhancing Construction Quality Inspection and Management using RFID Technology. Automation in Construction, 17: 467-479.  119-8  Smart Mobile App for Site Inspection and Documentation Long D. Nguyen1, Anna Koufakou2 and Colin Mitchell2 1 Department of Environmental and Civil Engineering, Florida Gulf Coast University, USA 2 Department of Software Engineering, Florida Gulf Coast University, USA  ICSC15, Vancouver, Canada June 9, 2015 Background •  Extensive effort spent on inspecting and documenting construction defects. •  For a 20,000-m2 conference center, •  41 pre-inspections conducted in a year (Lundkvist et al., 2014).  2 Research Objectives •  Developing a smart application called InSite Inspector (Intelligent Site Inspector): i.  taking images and catalog construction defects;  ii.  automatically locating defects using GPS; and iii.  producing various types of reports for different inspection purposes.  3 Current Construction Defect Inspection and Documentation Site inspection (photos, field notes)Transferring photos to PCs Engineers reviewing photos and describing defectsRequest for inspecting defectsPrepare reports of site inspectionExpert testimony in arbitration and litigation1 2 34564 Proposed Construction Defect Inspection and Documentation Site inspection with defect classification and reporting mobile appRequest for inspecting defectsPrepare reports of site inspectionExpert testimony in arbitration and litigationDefect image databaseImage processingMachine learning1 2342'5 InSite Inspector Current version Methodology •  Review of the state of practice in site inspection •  Collection of different types of inspection documents •  The Android-based platform (4.1.x with API level 16) •  JAVA/XML technology 6 Distribution of the Android Platform Versions in early January 2015 0.4% 7.8% 6.7% 19.2% 20.3% 6.5% 39.1% 85.1% Version 2.2, API Level 8 Version 2.3.x, API Level 10 Version 4.0.x, API Level 15 Version 4.1.x, API Level 16 Version 4.2.x, API Level 17 Version 4.3, API Level 18 Version 4.4, API Level 19 7 InSite Inspector: Simplified UML Class Diagram Main Menu Defect ScreenInspection ScreenConcerned Party ScreenReport ScreenLocation ScreenDatabase Data File ManagementGPSImage HandlingAndroid Activities8 Site Inspection Module 9 Documentation and Reporting Module 10 Documentation and Reporting •  Real-world vs. InSite Inspector generated reports Professional inspection report       InSite Inspector generated report 11 Conclusions and Future Work •  InSite Inspector: •  facilitates inspectors to record defect information, •  automatically locates defects using GPS, •  allows to customize report information and formats, •  generates reports for different purposes. •  Future work: •  improving the accurate positioning of the defects, •  utilizing computer vision techniques to automatically classify defects. 12 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52660.1-0076358/manifest

Comment

Related Items