Remote Monitoring of Dynamic Construction Processes Using Automated Equipment Tracking

1360 1 Remote Monitoring of Dynamic Construction Processes Using Automated Equipment Tracking Reza AKHAVIAN 1 and Amir H. BEHZADAN 2 M.S.C.E. Student, Department of Civil, Environmental, and Construction Engineering, University of Central Florida, Orlando, FL 32816-2450, USA; PH (407) 823-2480; FAX (407) 823-3315; email: reza@knights.ucf.edu 2 Wharton Smith Faculty Fellow and Assistant Professor, Department of Civil, Environmental, and Construction Engineering, University of Central Florida, Orlando, FL 32816-2450, USA; PH (407) 823-2480; FAX (407) 823-3315; email: amir.behzadan@ucf.edu ABSTRACT Construction resource planning and control is traditionally done using static data and information available from similar projects. However, the uniqueness of and uncertainties involved in each construction project may require that field data from equipment is dynamically collected, analyzed, and integrated into the decisionmaking process in order to achieve the best possible operational plan. The collected data can be used to predict the performance of a construction system based on the latest status of the project, as well as to monitor if all pieces of equipment are operating according to the plan and if any corrective action is needed. This paper presents the results of a remote tracking technique developed to capture field data from construction equipment in real time for short term monitoring and control of construction operations. The developed technique uses a.net environment thus providing a convenient means for data collection, sorting, filtering, and interpretation. The collected data is time-stamped and thus, can be used to create a real time 3D animation stream of the ongoing operation. This facilitates the communication of project details and can be ultimately used as a verification and validation tool for the underlying simulation model. INTRODUCTION Construction resource planning and control at the operations level are key factors in managing the performance of ongoing activities in a project (Halpin and Riggs 1992). Studies conducted on a wide range of construction projects in the United States shows that a comprehensive operations level plan can help project decisionmakers foresee potential problems such as spatial conflicts and resource utilization before the actual operation takes place (Ibbs 1997, Cox et al. 2003, NRC 2007). This can dramatically influence the time and money that would have been otherwise spent on reworks, resolving conflicts, and performing change orders. Cox et al. (2003) suggested that rework is typically responsible for 6-12% of the overall expenditure for a construction project. Construction Industry Dispute Avoidance Task Force (DART) reported that annually more than $60 billion was spent on change orders in US (Ibbs 1997). Also, according to the Federal Facilities Council (FFC), in 10-30%

1361 of all construction projects serious disputes are estimated to arise with a total cost of resolution between $4-12 billion each year (NRC 2007). In a typical construction project, effective planning is primarily dependent on proper management of a large volume of information that can include inputs from alternative project designs, material properties, labor productivity, equipment specifications, and project schedules. Furthermore, dynamic nature of construction projects introduces several layers of uncertainty that can range from internal factors (e.g. project time and cost variations, equipment breakdowns, contractor claims) to external events (e.g. weather conditions, financial market stability). Computer applications have thus evolved during the past several years to facilitate the process of project planning by providing a convenient and reliable means for modeling, simulating, and visualizing project activities (Paulson et al. 1987, Oloufa 1993, Shi 1999, Behzadan and Kamat 2009). In particular, advances in automation and information technology resulted in new approaches for collecting and managing construction data. Automated tracking systems have been employed to determine the location and status of resources (e.g. construction equipment and personnel). RESEARCH MOTIVATION Locating resources and knowing the context surrounding the operations by using the field data in a timely manner are valuable for monitoring the workflow of activities. Having the filed data during operations can help decision-makers predict the performance of a construction system based on the latest status of the project. Also, field data can be collected in order to create visual presentations of various activities taking place in a construction site. 3D visualization can serve as a convenient tool for decision-makers to get a real insight of what is exactly happening on a jobsite (particularly for operations that are hard to quantify or represent in a parametric model). It is also a substantial help for verification and validation of the underlying simulation model. This is especially important because decision-makers often do not have the time and knowledge to confirm the accuracy and validity of simulation models and thus do not usually rely on the results (Shi 1999). However, in the absence of a methodology that enables planners to collect real time data from the field, most project decision-makers rely on readily available project information and their own subjective judgment in evaluating uncertainties and forecasting future project performance (Song 2008, Akhavian and Behzadan 2011). This of course cannot assist in short-term planning due to the dynamics of construction projects. Hence, the presented research is mainly motivated by this need and is aimed to investigate the requirements and applicability of an integrated framework that uses the paradigm of Dynamic Data Driven Application Simulation (DDDAS) to address the problem of short-term operational level planning and control. The focus of this paper is on presenting the results of a remote tracking technique developed to capture field data from construction equipment in real time and create live 3D animated scenes of the operations. The developed technique uses a.net environment thus providing a convenient means for data manipulation. The presented study has been investigated and validated in an indoor laboratory scale setting. However, the developed methodology is generic in nature and can be adopted for use in outdoor situations with minimum modifications.

1362 IDENTIFIED GAP IN THE STATE OF KNOWLEDGE Despite the advancements of construction measurements and sensing technologies in recent decades, having accurate and latest information about the status of construction operations remains an issue in construction industry (Saidi et al. 2003). Therefore, automated resource (personnel, equipment, and material) tracking has been the subject of many studies in construction research and facility management. Different technologies in indoor and outdoor environments have been developed for resource tracking. A variety of outdoor and indoor location tracking and remote sensing technologies exist with significantly different characteristics, infrastructure, device requirements, and data protocols. In an early attempt, Radio Frequency Identification (RFID) technology was used for tracking high-valued materials on construction jobsites (Jaselskis et al. 1995). Global Positioning System (GPS) was another popular technology in a number of previous studies mainly due to its capability in accurate tracking of construction labor and equipment in outdoor environments and construction sites (Oloufa et al. 2003, Navon and Shpatnitsky 2005, Caldas et al. 2006). Even, there have been some attempts in combining RFID with GPS technology for tracking components, data transfer between maintenance workers, and tracking material on jobsites (Song et al. 2006, Ergen et al. 2007). Another technology called ZigBee, overcame the drawbacks of GPS and RFID systems in terms of accuracy and cost and has been used in an Automated Material Tracking (AMTRACK) system (Jang and Skibniewski 2007). In the context of active work zone, safety, and location tracking Ultra Wide Band (UWB) has been deployed (Teizer et al. 2007). Automated tracking had some applications in indoor environments as well. In an indoor environment, where Global Navigation Satellite System (GNSS) data is not available, indoor positioning technologies are used. Indoor GPS, Wireless Local Area Network (WLAN), Inertial Navigation Systems (INS), Bluetooth, Infrared, and Ultrasonic are some of the available technologies for indoor tracking purposes and a number of applications have been developed based on these technologies (Khoury and Kamat 2009, Razavi and Moselhi 2011). Table 1 summarizes the results of an extensive literature review conducted by the authors on tracking and localization technologies and their application in construction research. As Table 1 suggests most of the current applications of remote data collection systems are related to material tracking in order to enhance the project productivity and also safety and monitoring purposes. To the authors best knowledge, the application of real time automated equipment tracking for the purpose of short-term planning and control has not yet been investigated. The presented research aims to fill this gap by investigating the extent to which remote data collection and resource tracking can be used to assist in short-term planning and control and decision-making.

1363 Table 1. Previous Research on Tracking Applications in Construction Jobsites Study Application Technology Jaselskis et al. 1995 Material Tracking RFID Song et al. 2006a Material Tracking RFID Navon and Shpatnisky 2005 Monitoring of Road Construction GPS Oloufa et al. 2003 Equipment Tracking/safety GPS Caldas et al. 2006 Material Tracking GPS Song et al. 2006b Material Tracking RFID + GPS Ergen et al. 2007 Material Tracking RFID + GPS Jang et al. 2007 Material Tracking ZigBee Teizer et al. 2007 Safety UWB Khoury and Kamat 2009 Tracking mobile users UWB/Indoor GPS/WLAN/ METHODOLOGY Dynamic Data-Driven Application Simulation (DDDAS) The fundamental concept that this research has been built upon is DDDAS. The paradigm of DDDAS refers to the dynamic incorporation of additional data into a running application, and the ability of an application to dynamically steer the measurement process (Darema 2005). In construction operations, a DDDAS model dynamically measures site data and integrates this collected data with the corresponding simulation model in form of a new information layer to constantly adapt the model to the dynamics of the construction system and update it based on the latest collected operational data. DDDAS enables a more accurate prediction of how a dynamic construction system behaves based on the current status of its constituents (i.e. resources). The system architecture of the developed framework built around the concept of DDDAS is shown in Figure 1. Figure 1. System architecture of the developed DDDAS framework.

1364 Data Collection and Visualization Procedure The laboratory scale validation experiments of the data collection and equipment tracking component of the presented framework have been successfully conducted to collect equipment motion data in real time in order to create precise 3D animated scenes of ongoing activities. Since these experiments were conducted in an indoor environment, GPS positional data were not integrated with the animation and it was assumed that equipment positions were known. However, as stated earlier, the presented methodology is generic and can be adopted for use in outdoor situations with minimum modifications. Data Collection Device: 3D Orientation Tracker Figure 2 illustrates a 3D orientation tracker mounted on a model excavator. The 3D orientation tracker used in these validation experiments was a TCM Prime 3- axis electronic compassing module. 3-aixs magnetic field sensing and 3-axis tilt sensing are incorporated together in this module to provide accurate and precise measurements. As shown in Figure 2, the module sends values for rotational angles of an object in the form of a vector of (yaw, pitch, and roll). Figure 2. A Prime 3D orientation tracker mounted on a model excavator Developed Algorithm for Data Collection Figure 3 illustrates the overall layout of the validation experiments. As shown in the schema of Figure 3, orientation trackers collect and transmit data from construction equipment models to a centralized computing unit. The transmitted data is then processed and used to create real time 3D visualization of the operations. In order to communicate with the data collection devices, a serial port communication algorithm was developed. Since the collected tracker data is in a binary format, the developed serial port communication algorithm contains methods to decode the transmitted data and convert them to a computer interpretable format.

1365 Figure 3. Data flow from model equipment to remote data processing unit The developed algorithm takes advantage of object-oriented programing in a C++.NET environment. Using serial communication dynamically linked libraries (DLLs), an initial communication with the port is established, the port is opened, data (i.e. three orientation angles) is received through the port, and the port is closed at the end of the experiment. The orientation data coming through different brands of orientation trackers may follow different transmission protocols. For example, the Prime module utilizes a binary data protocol to obtain, extract, and transmit the tracker data over an RS-232 interface. Using the binary data provides the system with a fast data transmission rate. However, this will in turn makes the communication very sensitive to data corruption (Proakis and Salehi 2008). As a result, a mathematical transformation method called the Cyclic Redundancy Check (CRC) is used to distinguish between useful and corrupted binary data packets. CRC is applied to a series of bytes and produces an integer result that can be used for error detection (PNI 2011). After data is received from the orientation tracker, the tracking application computes the CRC value of the packet with its original content and then matches it with the received CRC value. If the two values are not the same, the packet is considered as corrupted and will be disregarded and the application waits for the next data packet. If the two values are equal, the data is safe to be used and will be extracted into its components. Using a set of binary data manipulation statements provided in the orientation tracker s Application Programming Interface (API), the numerical values for each of the orientation angles (yaw, pitch, and roll) are obtained. The main functionalities of the C++ managed class developed for acquiring orientation tracker data through a serial port is shown the flowchart of Figure 4.

1366 Figure 4. Flowchart of 3D orientation tracker serial communication process Visualization A scene graph is a data structure aiming at organizing the logical and often spatial representation of a graphical scene. Scene graph application programming interface (API) provides a means for constructing scenes that follow a hierarchical data structure of objects (called nodes) (Kamat and Martinez 2002). Any transformation (i.e. translation, rotation, scaling) applied to a parent node will affect all its child nodes. Considering a scene consisting of a loader and a truck, Figure 5 shows the hierarchical scene graph and relationships between different nodes. In this research, OpenSceneGraph (OSG) is used inside the.net environment to create dynamic animations of ongoing equipment activities and to link each and every object motion inside the animation to the collected field data that represent the actual motions of that object. The scene graph developed in this research creates a real time animation based on the latest values of the transformation matrices of the scene objects. Figure 6 shows snapshots from a 3D animation created in real time based on the data collected from a model truck and a model loader. As shown in this Figure, the scene is continuously updated as a result of model construction equipment moving.

1367 Figure 5. Hierarchical scene graph and relationships between different nodes Figure 6. Real time animation of model equipment operations CONCLUSIONS AND FUTURE WORK Construction simulation systems traditionally used static or historical data for the purpose of early project planning and long-term scheduling. However, the dynamic environment of most construction projects requires that not only the main entities and logical relationships of the real system be modeled, but also the dynamics on the ground to be incorporated in real time into simulation models. As a result, collecting and manipulating field data was one of the major challenges of this research. The authors successfully developed an algorithm inside the.net environment that enabled data collection and transmission from data acquisition devices mounted on model construction equipment to a centralized computing unit where data was sorted, analyzed, and used to create real time 3D animations of concurrent construction processes. The presented data collection algorithm read and saved data in a sequential format which could possibly lead to synchronization and

1368 multithreading issues especially when multiple data acquisition devices were simultaneously used. The authors are currently working on a more efficient algorithm to resolve this potential problem. This paper mainly presented results of the real time data collection and visualization components of the developed data-driven simulation system. The authors are currently working on creating the requirements for a self-adaptive simulation modeling methodology that can be effectively used for short-term planning and control. The real time data will be used to alter the parameters of an existing simulation model based on the latest changes of the real operations. In this manner, it is possible to compare the predicted results such as production rate and cycle times obtained from simulation models with the actual conditions and to try to minimize this deviation. Using the most updated simulation model, what-if analysis will be conducted and the best plan of action will be sent to the field personnel. Field personnel can then evaluate the results, override or modify parts of the plan, if necessary, and send it to equipment operators. REFERENCES Akhavian, R., and Behzadan, A. H. (2011). "Dynamic Simulation of Construction Activities Using Real Time Field Data Collection." Proc., 18th Workshop of Intelligent Computing in Engineering and Architecture (EG-ICE), Enschede, Netherlands, 1-10. Behzadan, A. H., and Kamat, V. R. (2009). "Automated generation of operations level construction animations in outdoor augmented reality." J. Comput. Civil Eng., 23(6), 405-417. Caldas, C. H., Grau, D. T., and Haas, C. T. (2006). "Using global positioning system to improve materials-locating processes on industrial projects." J. Constr. Eng. Manage., 132(7), 741-749. Cox, R. F., Issa, R. R. A., and Aherns, D. (2003). "Management's perception of key performance indicators for construction." J. Constr. Eng. Manage., 129(2), 142-151. Darema, F. (2005). "Dynamic data driven applications systems: new capabilities for application simulations and measurements." Proc., Computational Science -- ICCS 2005, Atlanta, USA, 610-615. Ergen, E., Akinci, B., East, B., and Kirby, J. (2007). "Tracking Components and Maintenance History within a Facility Utilizing Radio Frequency Identification Technology." J. Comput. Civil Eng., 21(1), 11 20. Halpin, D. W., and Riggs, L. S. (1992). Planning and analysis of construction operations, Wiley, New York. Ibbs, W. (1997). "Quantitative impacts of project change: size issues." J. Constr. Eng. Manage., 123(3), 08-11. Jang, W. S., and Skibniewski, M. J. (2007). "Wireless sensor technologies for automated tracking and monitoring of construction materials utilizing Zigbee networks." Proc., ASCE Construction Research Congress, Grand Bahamas Island.

1369 Jaselskis, E. J., Anderson, M. R., Jahren, C. T., Rodrigues, Y., and Njos, S. (1995). "Radio Frequency Identification Applications in Construction Industry." J. Constr. Eng. Manage., 121(2), 189-196. Kamat, V. R., and Martinez, J. C. (2002). "Scene graph and frame update algorithms for smooth and scalable 3D visualization of simulated construction operations." Journal of Computer-aided civil and infrastructure engineering, 17(4), 228-245. Khoury, H. M., and Kamat, V. R. (2009). "Indoor User Localization for Context- Aware Information Retrieval in Construction Projects." Autom. Constr., 18(4), 444-457. Navon, R., and Shpatnitsky, Y. (2005). "A model for automated monitoring of road construction." Constr. Mange. Econ., 23(9), 941-951. NRC (2007). "Reducing Construction Costs: Uses of Best Dispute Resolution Practices by Project Owners." National Research Council (NRC), Federal Facilities Council Technical Report No. 149 Washington, DC. Oloufa, A. A. (1993). "Modeling operational activities in object-oriented simulation." J. Comput. Civil Eng., 7(1), 94-106. Oloufa, A. A., Ikeda, M., and Oda, H. (2003). "Situational awareness of construction equipment using GPS, wireless and web technologies." Autom. Constr., 12(6), 737-748. Paulson, B. C., Chan, W. T., and Koo, C. C. (1987). "Construction operation simulation by microcomputer." J. Constr. Eng. Manage., 113(302), 302-314. User Manual (2010). CompassPoint Prime, 3-Axis Electronic Compass Module. Proakis, J. G., and Salehi, M. (2008). Digital communications, McGraw-Hill, Boston. Razavi, S., and Moselhi, O. (2011). "Indoor Construction Location Sensing using Low Cost Passive RFID Tags." Proc., The ASCE Construction Research Congress, Ottawa, Ontario, CN1-10. Saidi, K. S., Lytle, A. M., and Stone, W. C. (2003). "Report of the NIST Workshop on Data Exchange Standards at the Construction Job Site." Proc., 20th International Symposium on Automation and Robotics in Construction (ISARC), Eindhoven, The Netherlands, 617-622. Shi, J. (1999). "Activity-Based construction ABC modeling and simulation method." J. Constr. Eng. Manage., 12 (354), 354-360. Song, J., Haas, C. T., and Caldas, C. (2006). "Tracking the location of materials on construction job sites." J. Constr. Eng. Manage., 132(9), 911-918. Song, L. (2008). "A framework for real-time simulation of heavy construction operations." Proc., 2008 Winter Simulation Conference (WSC), Miami, FL, 2387-2395. Teizer, J., Lao, D., and Sofer, M. (2007). "Rapid Automated Monitoring of Construction Site Activities using Ultra-Wideband." Proc., 24th International Symposium on Automation and Robotics in Construction, Cochin, Kerala, India, 23-28.