Smart office; a data-driven management tool for mechanized tunnel construction
Like other industries, construction firms today are capturing more data than ever before through information-sensing devices (such as job site sensors, smartphones and heavyequipment tracking devices). Though there is an enormous potential of leveraging the power of captured data to increased productivity, much of the data is siloed without being utilized. In mechanized tunneling construction with tunnel-boring machines (TBMs), utilization of data to improve the construction process is limited to a small portion consumed by the machine operator. Indeed, much of that data is stored and filed away once a project is completed. With today’s ease of data acquisition, in addition to TBM data, tunneling construction firms can capture more data than ever before through other information-sensing devices on the job site. Transforming these heterogeneous, seemingly unrelated data into coherent, visually immersive and interactive insights will enhance every aspect of the project execution.
Data-driven modeling systems (such as BIM, CIM) have become more popular in tunnel engineering construction. The Crossrail Tunnel in London (Heikkila and Makkonen, 2014), State Route 99 Tunnel in Seattle, WA (Lensing, 2016; Trimble, 2011), Hallandsas Tunnel in Sweden (Smith, 2014), Hangzhou Zizhi tunnel in China (Wang et al., 2015) and Mikusa Tunnel in Japan (Sugiura, S., 2015) are all examples of projects that have been executed by utilizing data-driven methods (Fig. 1). Despite the several advantages of BIM models in the design field, due to the disruptive nature of this system, there are several challenges in implementation for the construction firm (Davidson, 2009). For instance, BIM doesn’t help to facilitate communication in the construction field, and it should be implemented with its full capacity to be effective.
Sporadic attempts have been made to utilize the power of data on job sites, and a few companies (such as Babenderede Engineering and Tunnelware) have developed data management and visualization software compatible with the tunneling construction market. This commercial software can be used as a tool to make better decisions, increase productivity, control the quality of materials and improve safety at job sites. For instance, modular software was developed by Babenderede Engineering to support the tunneling construction process. Babenderede Engineering simply equips the contractor’s TBM with an external data acquisition system to pull relevant data from the machine PLC, and then, by processing raw data, helpful information is generated. The company has also developed software exclusively for tracking segmental lining through the lifecycle of the project (Cicinelli et al., 2017).
Tunnelware that is still in the development stage tries to consolidate data from several sources (such as TBM, site personnel reports, and job site sensors). According to the Tunnelware experts, the software will provide a robust tool for the constructors to visualize and analyze the tunnel excavation processes in a 5D format. Tunnelware is also adding other features such as cutter-tool management and virtual meeting rooms to their software package.
Although available commercial services have several benefits, there are drawbacks that make contractors hesitant to utilize these services in their projects. For example, the software visualization is old-fashioned and dissociated. Indeed, it is crucial to prepare an easy-touse interactive visualization platform that can facilitate construction processes for project personnel; otherwise, it will be abandoned throughout the construction period. In addition, because every tunnel project is unique, existing software on the market is not flexible enough to meet the uniqueness of the project.
In this article, the concept of a unified analytics center compatible for tunnel construction is presented. First, the Dugway Storage Tunnel project that is considered as a case study for testing our model is introduced. Later, the structure of the conceptual model, data collecting and required hardware and software are explained. Afterward, application of the proposed model throughout the construction period of the tunnel is partially examined. The advantages of implementing such a data-management system as a tool for project managers, engineers, safety superintendents and others are also addressed.
Dugway Storage Tunnel
Project description. The Dugway Storage Tunnel (DST) is the second of seven tunnel projects that will reduce 15.1 GL (4 billion gal) of pollution discharged into Lake Erie due to the seasonal overflows. The tunnel alignment is approximately 4.5 km (2.8 mile) in length with seven curves of variable radius, excavated using a single-shield, hard-rock TBM with 8 m (27 ft) excavation diameter. The finished internal tunnel lining diameter is 7.3 m (24 ft) using concrete segments of 0.3 m (1 ft) thickness. Depths of the tunnel invert ranges from 55 to 70 m (180 to 230 ft) below ground surface. The project includes a total of six deep shafts along the path of the tunnel with an internal lined diameter between 5 and 50 m (16 to 50 ft), and four adit connections between these shafts and tunnel of variable lengths between 15 and 304 m (50 and 1,000 ft) (Fig. 2). The 14 m (46 ft)-diameter shaft, known as DST-1, is the TBM launch shaft in which all the main conveyor systems will be installed. Part of the shafts were constructed through soft ground and part encountered Chagrin Shale bedrock. The project includes the construction of additional structures including diversion structures, gate structures, control vaults, ventilation vaults, drop manholes and modifications to existing regulatory structures. The project area is mainly older residential (pre-1950s interspersed with commercial properties and urban parks).
The TBM used to excavate the 8.2 m (27 ft)-diameter tunnel through the Chagrin Shale was a hard-rock, singleshield Herrenknecht machine type S-684 (Fig. 3). The machine was reconditioned on site by the contractor after excavating the first phase of the tunnel (Euclid Creek Tunnel). The TBM was partially assembled for the launch with only three of the six gantries in the starter tunnel that was previously excavated by employing drill-and-blast method. At this stage, materials were hauled out using locomotives and muck boxes. After the first 91 m (300 ft) of tunnel had been excavated, the TBM was assembled in its final setup with six gantries. The tunnel conveyor system was comprised of five sections transporting materials from the tunnel to the vertical conveyor to overland belt and finally to the stacker. There was a total of 2,911 rings installed at a total length of 4,523 m (14,840 ft). The production average was 17 rings per day. The average of the excavation parameters were as follows:
- Penetration 10.3 mm/rotation.
- Cutter-head rotation 6 rpm.
- Advance of the TBM 66.75 mm/min.
- Thrust force = 10,000 KN.
- Torque = 2,500 KNm.
There was no presence of water in the material, but the quantity of methane trapped between the layers was sometimes relevant.
Unified analytics center
Conceptual model. Managing this data and using it as a tool to make better decisions during the construction process is of utmost importance. The purpose of the conceptual model of the unified analytics center is to provide a robust tool to enhance decision-making for all levels of on-site and office personnel. The conceptual model comprises:
- A physical smart office equipped with digital display screens.
- Sensors, equipment and digital tools to gather data from the project site.
- A platform to integrate, analyze and contextualize this information into a visual, meaningful representation.
Figure 4 shows the data flow model (DFM) of the proposed conceptual model. Data captured from the TBM, the conveyor system, equipment operating, inventory, material used, site personnel reports, labor hours, project documents and online weather reports are structured and pushed into a cloud-based collaboration and sharing system. Microsoft Power BI, which is a suite of analytics tools, is utilized to create interactive visualization and dashboards to share insight with individuals across the project and company. The model can provide customized dashboards suited to the end user that can be updated in real time, near real time or over a longer period. For instance, real-time monitoring of critical parameters of the TBM (such as thrust force, torque) can be visualized in the form of interactive dashboards to assist with optimizing the excavation process. As another example, historical data that have been collected over the years on the project can be presented to the headquarters to give overall insight into project progress and assist decision-makers to gain a competitive advantage when estimating and bidding on a new project with similar characteristics.
Application of the model in construction of the Dugway Storage Tunnel. Throughout the excavation phase of the tunnel, data from TBM PLC was pulled out and pushed into the Power BI platform. This system provided a means to connect the project personnel (project manager, construction manager, project engineer, tunnel shift engineers and TBM operator) to a broad range of data via easy-to-use dashboards, interactive reports and meaningful interactive visualizations. Prepared customized dashboards helped to ease the decision-making process for those directly and indirectly involved in the project. Figures 5 to 7 show examples of crafted dashboards. For instance, as shown in Fig. 5, by integrating a real-time interactive map of the TBM location and data recorded from geotechnical instrumentations, a better understanding of the behavior of the ground subjected to excavation can be seen. This interactive dashboard can assist the TBM operator in adjusting steering parameters. Figure 6 outlines the volume of the injected accelerator admixture from each port for each ring separately. By analyzing this data an engineer or TBM operator can solve a problem related to accelerator injection and not just rely on the last ring data. Collected data can be formatted in a customized and easy-to-use dashboard to monitor production per shift (Fig. 7). Having access to historical data of shift production can help increase efficiency by switching between working shift patterns (rotating three eight-hour shift schedules and two 10-hour shift schedules).
Figures 5 to 7 are a few of many helpful dashboards that can be generated to get the right information to the right person at the right time. In order to have a better picture of the advantages of employing such a data-driven system, it is useful to address the model’s application as follows.
As a tool for engineers. With a Wi-Fi system in the tunnel, job site engineers have permanent access to the real-time dashboard of all critical parameters of excavation, grout utilization, and navigation. As shown in Fig. 8, grout utilization parameters (such as grout and accelerator admixture pressure and volume) are integrated into a customized easy-to-use dashboard that can be used as a robust tool by engineers to monitor grout utilization per installed ring. As another example, we integrated daily production data with push and ring build time to get a better real-time picture of the TBM utilization throughout the excavation phase (Fig. 9). We also formulated several other dashboards to facilitate monitoring TBM cylinder pressure and extensions, gas infiltration location and values, ring installation effect on navigation, and the variability of tendencies during the advance and many other parameters and behaviors.
As a tool for project managers. The physical smart office is where people can meet to discuss the ongoing tunnel construction process. All critical data are available in the shape of interactive dashboards that provide a unique tool for project managers to hover over all aspects of the project from different viewpoints. For example, a dashboard can be generated to illustrate the production progress on shift, daily, weekly and monthly bases. Monitoring the project progress through this dashboard is far more convenient than exploring archived hard copies of a personnel report (Fig. 10).
As a tool for reporting. Digital custom reports for the construction process based on different time scales (shift, daily, weekly, monthly, full project) can be generated in the unified analytics center automatically. This platform also allows for the sharing of automatic reports on predefined timeframes with coworkers and any other recipients outside the company by including the list of email addresses in the system. For instance, a dashboard comprising production progress, grout utilization, TBM utilization, and the average of TBM excavation parameters can be an informative means for program managers in a headquarter office as well as estimators when estimating and bidding on a new project with similar characteristics (Fig. 11).
Conclusions
This article presented the fundamental principles of a unified analytics center as a robust tool for the tunneling industry. The application of a data modeling system in the construction of the Dugway Storage Tunnel was examined. Although data management in mechanized tunneling construction is not a new concept, there are several features that make this proposed model different. Interpreting data into meaningful easyto- use, real-time visual dashboards that give all individuals insight into the project is the foremost advantage of the model. In addition, Microsoft Power BI as the backbone of the model is an analytics service entrenched in the Microsoft stack. Indeed, unlike other inflexible predefined data management and visualization platforms, Power BI compatibility increases adoption of data modeling for contractors. Although the system was implemented just by using TBM data, due to the flexible characteristic of the model, we could take advantage of the system in our decision-making process. The proposed system also allows for automatic generation of consistent reports on shift, daily, weekly, monthly scales. The system will be examined as a whole by inducing data from other sources (such as job site sensors, smartphones, and heavyequipment tracking devices). The all-inclusive system will allow for improved decision-making and accordingly increase efficiency and reduce construction costs.
References
Cicinelli, V., Stahl, F., & Gronbach, T. 2017. Copenhagen Cityringen Project: Big Data to Manage Quality Control in Megaprojects. RETC San Diego, June 4-7, 2017 page 190-202.
Davidson, A. 2009. A Study of the Deployment and Impact of Building Information Modelling Software in the Construction Industry, England University of Leeds, 2009. Available: http://www.engineering.leeds.ac.uk/eengineering/ documents/AndrewDavidson.pdf
Heikkilä, R., Kaaranka, A., & Makkonen, T. 2014. Information Modelling based Tunnel Design and Construction Process. The 31st International Symposium on Automation and Robotics in Construction and Mining (ISARC Proceedings2014). Sydney, Australia, July 2014.
Lensing, R. 2016. BIM and construction process data in mechanized tunnel construction; Milestone control for tunnel construction sites using automatically created process data in comparison with 4D BIM. Master of Science thesis (Geographical Information Science & Systems). MSc(GIS)” Heidelberg, 20.12.2016
Smith, C. 2014. BIM at Sweden’s Hallandsås Tunnel: Planning pioneer [Online]. New Civil Engineer. Available: http://www.newcivilengineer.com/features/geotechnical/ bim-at swedenshallandss-tunnel-planning-pioneer/8663146.article Sugiura, S. 2015. First Application of CIM to Tunnel Construction in Japan. Proceedings of International Conference on Civil and Building Engineering Informatics (ICCBEI 2015), 82. Tokyo, Japan, 2015.
TRIMBLE. 2011. Seattle’s Massive Tunnel Makes travel safer – with Tekla BIMsight [Online]. Available: https://www.teklabimsight.com/references/ seattles-massive-tunnel-makes-travel-safer [Accessed 24.07.2016 2016].
Wang, J., Hao, X., & Gao, X. 2015. The Application of BIM Technology in the Construction of Hangzhou Zizhi Tunnel. 3rd International Conference on Mechatronics, Robotics and Automation (ICMRA 2015). ISBN 978-94- 62520-76-9.
