Improving On-Site Construction Productivity Using the WRITE System Seonghoon Kim Department of Construction Management and Civil Engineering Technology Georgia Southern University P.O. Box 8047, 1006C Carruth Bldg. Statesboro, GA 30458
[email protected] Yong Bai Department of Civil, Environmental, and Architectural Engineering University of Kansas 1530 W. 15th Street, 2150 Learned Hall Lawrence, KS 66045
[email protected]
ABSTRACT Existing construction productivity measurement techniques are not capable of providing the real-time productivity data to project managers and engineers for analyses and sharing the data among participants involved in construction operations. As a result, actions to address the on-site productivity problems cannot be taken just in time. To address these shortfalls, the Wireless Real-time Productivity Measurement (WRITE) System was developed to measure the on-site construction productivity in real time. In addition, an on-site construction productivity improvement model using the WRITE System and the benchmark data was developed. Field experiments were conducted on a bridge construction project to determine the accuracy of the developed model. During the experiments, the real-time productivity data measured by the WRITE System was compared to the benchmark productivity data. The results of the comparison provided the necessary information for project managers and engineers to determine if immediate actions should be taken to improve the on-site construction productivity. The success of this research project made several major contributions to the advancement of the construction industry. First, it advanced the application of wireless technology in construction operations. Second, it provided an advanced technology for engineers and project managers to determine on-site construction productivity in real time. Thus, actions on improving on-site construction productivity could be taken just in time if needed. These advancements enhance the contractors’ capability of managing construction projects. Key words: benchmark—construction—measurement—productivity—wireless
Proceedings of the 2009 Mid-Continent Transportation Research Symposium, Ames, Iowa, August 2009. © 2009 by Iowa State University. The contents of this paper reflect the views of the author(s), who are responsible for the facts and accuracy of the information presented herein.
INTRODUCTION Productivity data have been widely used as performance indicators to evaluate construction operations throughout the entire phase of construction. Construction companies must continuously track productivity in order to estimate their performance to maintain profitability and to prepare future biddings (Ghanem and Abdelrazig 2006). Hence, measuring productivity at a project site has been an important task in the construction industry (Chang 1991). Over the years, on-site productivity measurement techniques have been developed, including questionnaires, stopwatch studies, photography, time-lapse videos, and videotaping (Oglesby, Parker, and Howell 1989). In recent years, real-time monitoring systems have become key methods to reduce the gap between actual and planned production rates in a timely manner (Navon and Shpatnitsky 2005; Sacks et al. 2005; Peddi et al. 2009). A real-time video system was developed by Everett and Slocum to monitor lifting activities of crane in attempts to improve both productivity and safety of crane operations (Everett and Slocum 1993). Since 2000, wireless technologies, such as global positioning system (GPS) and radio frequency identification (RFID) system, were utilized to track the current status of the resources and activities. A GPS technology was used to automatically measure earthmoving performance by identifying the locations of equipment at regular time intervals and converting the information into project productivity (Navon and Shpatnitsky 2005). A web-based camera was used to monitor interior construction operations. This web-based network technology produced an opportunity to avoid using a wired network connection in such congested construction jobsite (Kang and Choi 2005). Existing on-site construction productivity measurement methods have some common limitations in providing the real-time productivity data to project managers and engineers for analyses and sharing the data among participants involved in construction operations. As a result, opportunities to improve construction operations are lost. To address these shortfalls, there is an urgent need to develop new technologies that can be used to collect and analyze the on-site construction productivity data in real time. OBJECTIVES The first objective of this research project was to develop the Wireless Real-time Productivity Measurement (WRITE) System that could be used to collect and analyze the on-site construction productivity data in real time. Using the real-time productivity data, engineers and project managers may be able to accurately determine the bridge replacement progress and easily share the information with all parties involved in the bridge replacement project. Thus, the wireless real-time productivity measurement technology has a great promise to improve construction schedule forecasts and to increase emergency response capability after extreme events. The second objective was to develop a model for improving onsite construction productivity in real time utilizing the data collected by the developed WRITE System and the benchmarking data gathered from experts in the construction industry. RESEARCH METHODOLOGY The first phase of this research project was to conduct the literature review followed by the development of the WRITE System. During the development phase, the researchers identified the necessary hardware and software for the system and outlined a framework to show the connection of major hardware components. The third phase was to develop the model for construction productivity improvement using the WRITE System and the productivity benchmark data. A survey methodology was used to obtain the productivity benchmark data from bridge construction experts. The fourth phase was to conduct the field experiments to test the developed model. During the field experiment, productivity data collected using
Kim, Bai
2
the WRITE System at the construction site was compared to the benchmark data to form the basis for the project managers to make productivity improvement decision in real time. Finally, research findings and recommendations for future research were outlined. Field experiments were conducted in a bridge reconstruction project to demonstrate how this procedure was used by the construction project managers to identify on-site productivity problems and to take immediate action to address these problems. DEVELOPMENT OF WRITE SYTEM Overview of the WRITE System The developed WRITE system includes a video camera, a digital camera, a data processor, an AC transformer, two antennas, and a laptop computer as shown in Figure 1. The preliminary test results indicated that the developed system can measure the on-site construction productivity accurately (Kim 2008).
Figure 1. Major components of the WRITE system WRITE System Framework Figure 2 presents the framework of the WRITE System that was developed during the process of this research project. Once the video camera takes pictures from the construction site, the data processor immediately saves the pictorial data into files. Then, these files are transmitted in real time via wireless modems. An engineer or a project manager with the IP address at another location can access the data files via a wireless modem or a local area network (LAN) to conduct productivity analyses using a computer with VM View software. After finishing the data analysis, productivity data and live pictures can be presented in a website so that other users, such as the owner, engineers, contractors, and material suppliers, can share the information.
Kim, Bai
3
Figure 2. Framework of the WRITE System Development of the Productivity Improvement Model After building the WRITE System, a model for the on-site construction productivity improvement was developed as shown in Figure 3. In this model, the first task is to collect pictorial data in the construction site using the WRITE System. The second task is to determine the real-time productivity data which is the ratio of working and nonworking time. The third task is to compare the real-time productivity data with the productivity benchmark data. During the comparison, a project manager or an engineer will answer two questions, and then make productivity improvement decisions accordingly. The first question is whether the real-time productivity data is higher than the benchmark data at which action should be taken. If the answer for this question is no, management needs to take action immediately to improve the on-site productivity. If the answer for the first question is yes, management goes to the next stage to compare the real-time data with the acceptable benchmark data. If the real-time data is higher than the acceptable benchmark data, no action is needed. Otherwise, management needs to be aware that action may be needed in the near future and close monitoring is necessary at the construction site. The developed model can be utilized for the entire period of construction or for the segments of construction.
Kim, Bai
4
Figure 3. Productivity improvement model using the WRITE System and benchmarking data
Kim, Bai
5
FIELD EXPERIMENTS The researchers conducted field experiments to demonstrate how the developed model can be utilized to improve on-site construction productivity. The field experiment was performed in a steel girder bridge reconstruction project over Interstate 70 in Lawrence, Kansas. On-site productivity data were collected for two months from March 24 to May 23 in 2008. Prior to the field experiments, the researchers reviewed plans, specifications, the project cost and schedule, and other publications to obtain information on project uniqueness, crew size, and historical daily productions. In addition, the researchers visited the jobsites to gather the geographical information in order to develop the field experimental plan. Work Breakdown Structure (WBS) The work breakdown structure (WBS) has been widely used to manage the project. WBS is defined as “a deliverable-oriented grouping of project elements,” which organizes and defines the hierarchical structure of the entire project (Jung and Woo 2004). It is often used in the complex construction projects to identify project information and improve the efficiency of control processes (Chua and Godinot 2006). A WBS shows the relationship of all project elements at different levels and makes them more manageable and measurable. The number of levels depends on the size and complexity of the projects (U.S. Department of Energy 1997). The bridge reconstruction project used in the field experiments was broken down into four levels, including Level 1 (project), Level 2 (work zone), Level 3 (activity), and Level 4 (operation). Examples of the levels of steel girder bridge WBSs are shown in Table 1.
Kim, Bai
6
Table 1. WBS for steel girder bridge reconstruction Level 1 (Project)
Level 2 (Work Zone)
Steel Girder Bridge
General Abutment Pier 1 Pier 2 Pier 3 North side South side Span 1 Span 2 Span 3 Span 4
Level 3 (Activity) Mobilization Traffic Control Demolition Excavation Abutment 1 Abutment 2 Pier Drill Shafts Pier Columns Pier Cap Slope protection Beam Setting Deck Forming Reinforcing Deck Bridge Barrier Rail Concrete Barrier Backfill Abutments Approach road
Level 4 (Operation) Set up Crane Moving concrete safety barrier Driving pile Forming Structural excavation Slope protection (filter fabric and rock) Set bearing devices Unload beams Set beams Install diaphragms Bolting and tightening splice Ground splice Prepare deck material Prepare deck forming Overhangs Strip Place backwall (strip drain & backfill) Tying rebar Pouring and curing Strip and check elevation Others
Determining Productivity Benchmark Data Benchmarking has been used as a tool to improve productivity since the early 1980s (Jackson et al. 1994). The Construction Industry Institute (CII) has established construction productivity metrics and a reporting format for construction productivity benchmarking and improvement (Han et al. 2005). Actual working time of construction workers is at 56% in nuclear plant construction projects (Hewage and Ruwanpura 2006). Christian and Hachey (1995) studied concrete-placement operations, and the finding was 61% working time and 39% nonworking time. According to the previous research projects, the ratio of working time and nonworking time ranges approximately from 50:50 to 60:40. However, there is no consensus on the acceptance ratio of working time verse nonworking time in the construction industry because construction projects have different natures such as different types of projects, activities, and operations. In this research project, working and nonworking time for five bridge operations were identified, including deck forming, tying rebar, installing finisher, backfilling, and placing approach road footing. A total of 66 hours of video tapes were recorded using the WRITE System to determine the productivity rates for the five bridge operations. These videos were all taken zoomed-in so that the researchers could clearly identify the working and nonworking time for each operation. The ratio of working and nonworking time was 86% and 14% on average as shown in Table 2.
Kim, Bai
7
Table 2. Ratio of working and nonworking time determined by the WRITE System Time (Second) Operation
Working Time
Percentage (%)
Nonworking Working Nonworking Time Time Time
Deck forming
24,720
2,160
92
8
Tying rebar
40,320
5,880
87
13
Installing finisher
71,230
21,100
77
23
Placing backwall, strip drain, and backfill
44,850
1,950
96
4
Grade and tie approach road footing
21,275
2,725
89
11
Total
202,395
33,815
86
14
To identify the benchmarking data, an e-mail survey form shown in Figure 4 was developed and distributed to four construction professionals in the bridge construction. Table 3 shows names of construction professionals and their company names, specialties, and positions. The benchmark data were based on professional intuitions about rates of working time and nonworking time for each of the five bridge construction operations.
Figure 4. Survey form for collecting benchmark data
Kim, Bai
8
Table 3. List of survey construction professionals Name
Company
Construction Specialty Position
Ken Johnson
BRB contractors, Inc.
Bridge
Project Manager
Mike Laird
BRB contractors, Inc.
Bridge and Plant
Project Manager
Ray Rinne
A.M. Cohron & Son, Inc.
Bridge
Superintendent
Christopher J. Rech
A.M. Cohron & Son, Inc.
Bridge
Project Manager
Table 4 shows acceptable ratios provided by four survey participants. The overall average ratio for working time (WT) was 81% and overall average ratio for nonworking time (NWT) was 19%. Tying rebar had the highest nonworking ratio of 21%, while deck forming had the lowest rate of 16%. According to the survey participants, they can accept the working time ratio of at least 79% for these bridge operations. Table 5 presents ratios at which action should be taken by project managers to improve on-site construction productivity. The overall average ratio for WT was 75% and overall average for NWT was 25%. Tying rebar had the highest nonworking time rate of 28%, while deck forming had the least nonworking time rate of 23%. Table 4. Acceptable ratio BRB 1 Operation
BRB 2
A.M. Cohron 1
A.M. Cohron 2
Average
WT NWT WT NWT WT NWT WT NWT WT NWT (%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
Deck forming
85
15
80
20
85
15
85
15
84
16
Tying rebar
80
20
75
25
80
20
80
20
79
21
Installing finisher
85
15
75
25
85
15
85
15
82
18
Placing backwall, strip drain, and backfill
70
30
80
20
85
15
85
15
80
20
Grade and tie approach road footing
80
20
80
20
85
15
85
15
82
18
Average
80
20
78
22
84
16
84
16
81
19
Note: WT–Working Time; NWT–Nonworking Time
Kim, Bai
9
Table 5. Ratio at which action should be taken BRB 1 Operation
BRB 2
A.M. Cohron 1
A.M. Cohron 2
WT NWT WT NWT WT NWT WT NWT (%) (%) (%) (%) (%) (%) (%) (%)
Average WT (%)
NWT (%)
Deck Forming
75
25
75
25
80
20
80
20
77
23
Tying rebar
70
30
70
30
75
25
75
25
72
28
Installing finisher
75
25
70
30
80
20
80
20
76
24
Placing backwall, strip drain, and backfill
60
40
75
25
80
20
80
20
74
26
Grade and tie approach road footing
70
30
75
25
80
20
80
20
76
24
Average
70
30
73
27
79
21
79
21
75
25
Note: WT–Working Time; NWT–Nonworking Time
Table 6 presents the results of the comparison between the benchmarking data from the survey and the real-time productivity data determined by the WRITE System. For the operation of installing finisher, the nonworking ratio of 24% was equal to the ratio at which action should be initiated by the construction manager. The rest of operations had larger working time ratios than the minimum required working ratios. Table 6. Data comparison between the WRITE System and the benchmarks Acceptable Ratio Operation
Ratio at which action should be taken
WRITE System
WT (%)
NWT (%)
WT (%)
NWT (%)
WT (%)
NWT (%)
Deck Forming
84
16
77
23
92
8
Tying rebar
79
21
72
28
87
13
Installing finisher
82
18
76
24
76
24
80
20
74
26
96
4
82
18
76
24
89
11
81
19
75
25
88
12
Placing backwall, strip drain, and backfill Grade and tie approach road footing Average
Note: WT – Working Time; NWT – Nonworking Time
By comparing the rates from the WRITE System to the benchmark data, project managers can take actions for improving on-site construction productivity in real time. As shown in Table 7, there are three cases that project managers can make decision using the developed model. First, if the productivity ratio measured by the WRITE System is higher than the acceptable ratio, then, no action is required. Second, if the ratio is between acceptable ratios and ratios at which action should be initiated, then, management needs to be aware that an action may be needed in the near future. Finally, if the ratios are lower than the minimum required rate, then, the project manager needs to take actions immediately.
Kim, Bai
10
Table 7. Making management decisions using the WRITE System No. 1 2 3
Ratios from the WRITE System
Action
Higher than acceptable ratios Between acceptable ratios and ratios at which action should be taken Lower than ratios at which action should be taken
No action needed Aware that action may be needed Action is required
CONCLUSIONS AND RECOMMENDATIONS This research project made several major contributions to the advancement of knowledge in the construction industry. First, it advances the applications of wireless technologies in construction operations. As a result, all participants involved in construction projects can monitor construction projects at any location and any time. In addition, productivity data gathered by the WRITE System can be sent to a website so that owners, engineers, contractors, and material suppliers can share data and make necessary actions in remote locations. Second, due to the fact that the developed WRITE System is capable of continuously collecting and sending the on-site construction productivity data, construction managers and engineers now have a new technology to determine the on-site construction productivity in real time. Third, integrating the benchmarking data and the WRITE System data makes it possible for the construction managers and engineers to continuously improve the on-site construction productivity in real time. With advancements made by utilizing the WRITE System, communication and coordination will be improved at the construction project sites, which enhance the contractors’ capability of managing construction projects. This research project can be extended in several ways. First, because the process of determining working status using the live images obtained from the WRITE System is time-consuming and subject to human error and bias, there is a need to develop an algorithm to automatically classify working or nonworking status. Second, currently no database has been established for real-time productivity improvement using the WRITE System and benchmark data. There is no standard process for management to increase or decrease the crew size during the construction. Third, validation for the developed procedure needs to be performed on other construction operations in the future research projects.
Kim, Bai
11
ACKNOWLEDGEMENTS This research project was funded by the Kansas University Transportation Research Institute, the Kansas Department of Transportation, the Federal Highway Administration, and the National Science Foundation under Grant No. 0741374. The financial support from these agencies is greatly appreciated REFERENCES Chang, L.-M. 1991. Measuring construction productivity, Cost Engineering 33(10):19–25. (periodical) Chua, D.K.H. and M. Godinot, 2006. Use of a WBS Matrix to Improve Interface Management in Projects, Journal of Construction Engineering and Management 132 (1):67-79. (periodical) Everett, J.G., and A.H. Slocum, 1993. CRANIUM: Device for Improving Crane Productivity and Safety, Journal of Construction Engineering and Management 119 (1):23–39. (periodical) Ghanem,A.G., and Y.A. Abdelrazig, 2006. A framework for real-time construction project progress tracking, Proceedings of the 10th Biennial International Conference on Engineering, Construction, and Operations in Challenging Environments, Vol. 2006, American Society of Civil Engineers, League City/Houston, TX. (paper presented at a meeting) Han, S., M. Park, and F. Pena-Mora, 2005. Comparative Study of Discrete-Event Simulation and System Dynamics for Construction Process Planning, San Diego, USA: Construction Research Congress. (paper presented at a meeting) Hewage, K.N. and J.Y. Ruwanpura, 2006. Carpentry workers issues and efficiencies related to construction productivity in commercial construction projects in Alberta, Canadian Journal of Civil Engineering 33:1075–89. (periodical) Jackson A.E., and R.R. Safford, and W.W. Swart, 1994. Roadmap to Current Benchmarking Literature, Journal of Management in Engineering 10 (6):60–67. (periodical) Jung, Y. and S. Woo, 2004. Flexible Work Breakdown Structure for Integrated Cost and Schedule Control, Journal of Construction Engineering and Management 130 (5):616–25. (periodical) Kang, J., and J.-W. Choi, 2005. Observation Error of Time-Lapsed Photos in Construction Operation Monitoring, Computing in Civil Engineering 179:100. (periodical) Kim, S. 2008. Development of a Wireless Real-Time Productivity Measurement System for Rapid Construction Operations, PhD dissertation, University of Kansas. (thesis or dissertation) Navon, R., and Y. Shpatnitsky, 2005. Field Experiments in Automated Monitoring of Road Construction, Journal of Construction Engineering and Management 131 (4):487–93. (periodical) Oglesby, C.H., H.W. Parker, and G.A. Howell, 1989. Productivity Improvement in Construction, New York: McGraw-Hill, Inc. (book) Peddi, A., L. Huan,, Y. Bai, and S. Kim, 2009. Development of Human Pose Analyzing Algorithms for the Determination of Construction Productivity in Real-time. 2009 Construction Research Congress, Seattle, Washington, Vol 1, pp 11-20. (paper presented at a meeting)
Kim, Bai
12
Sacks,R., R. Navon, I. Brodetskaia, and A. Shapira, 2005. Feasibility of Automated Monitoring of Lifting Equipment in Support of Project Control, Journal of Construction Engineering and Management 131(5):604–614. (periodical) U.S. Department of Energy, 1997. Cost Codes and the Work Breakdown Structure, 1997. (book)
Kim, Bai
13
