Automatic data processing system for integrated cost and schedule control of excavation works in NATM tunnels.
Cho, Daegu ; Cho, Hunhee ; Kim, Daewon 等
Introduction
In the age of limitless competition, the construction industry has
been increasingly challenged to improve work performance. Traditional
methods such as organization or process changes seem to have reached
their limits and their leverage seems to be insignificant. With the
incredible rate of advancement in information technology, IT convergence
for the construction industry has been in the spotlight as a new
approach to lead the industry from its traditional labour-intensive
status to become technology or knowledge intensive. While the IT
convergence has been attractive to the industry, its influence has been
imperceptible compared with other industries. IT convergence must not
simply utilize some technology in projects, but must optimize core
technologies in the construction environment considering work processes,
field characteristics, business strategies, and so on.
Timely feedback on a project's status enables project managers
to identify problems early and make appropriate adjustments that can
keep a project on time and within budget. Cost and schedule control have
been regarded as two major management factors for successful performance
of construction projects. Because cost and schedule control are closely
interrelated in terms of sharing data and management process (Jung, Woo
2004), an integrated approach also promises a myriad of benefits for
information systems and management. While extensive research has
attempted to provide a systematic method to integrate cost and schedule
data, the integrated approach still seems to be a long way off in the
construction industry. Different levels of detail between cost and
schedule data structures generate many integrated Control Accounts
(CAs), which require considerable overhead effort from field managers to
acquire, track, and analyze data throughout the integrated data
processing. The additional overhead has generated much resistance to
adopting the integrated approach in the industry. As a result, despite
much effort, integrated cost and schedule control has been considered as
irritating work in construction fields. While previous effort to
integrate cost and schedule data have mostly focused on building
projects that consist of many elements, operations, and crews, it is not
widely applied to a large construction project such as a tunnel, road,
or bridge characterized by a relatively small number of information
units, horizontally repetitive operations, and construction in open
space.
Motivated by these challenges, this study hypothesizes that an
automatic data processing system (ADPS) applied to a large construction
project should help facilitate the integrated approach by reducing the
number of CAs and overhead effort in the data processing. The primary
objective of this study is to propose a new method to automate the
required processes for implementing cost and schedule integration. NATM
tunnel projects focusing on excavation works is considered to verify the
proposed method.
1. Cost and schedule integration
As construction projects become larger and more complex, a great
amount of data is generated. Project participants must rapidly
understand and leverage these data for efficient project management.
Because cost and schedule control are the two major functions
determining a project's success or failure, various document forms
and software applications are used to acquire, track, and analyze cost
and schedule data. These independent tools might be well customized to
represent original data, but there are many redundancies in using
different forms, structures, and perspectives (Rasdorf, Abudayyeh 1991).
They require considerable overhead effort to acquire, analyze, and
manipulate fundamentally the same data over and over again in the data
processing.
Because cost and schedule information are closely interrelated, it
would be ideal for both cost and schedule data to be represented by a
single hierarchy and controlled by a single parameter. Unfortunately,
the low-level items in a traditional bill of quantity (BOQ) representing
cost data and the low-level items in a scheduling diagram are in two
different levels of an information hierarchy and there is considerable
mismatch (Perera, Imriyas 2004). The different levels of detail have
been regarded as the main difficulty of cost and schedule integration
(Hendrickson, Au 1989; Rasdorf, Abudayyeh 1991; Jung, Woo 2004; Fleming,
Koppelman 2005). Many researchers have attempted to derive an
appropriate integration model for the industry. Teicholz's CBS to
WBS Mapping Model (Teicholz 1987), Hendrickson and Au's Work
Element Model (Hendrickson, Au 1989), Kim's Design Object Model
(Kim 1989), and Rasdorf and Abudayyeh's Work-packaging Model
(Rasdorf, Abudayyeh 1991) are well known. These models have tried to
find an integrated control point through distributing cost data into
schedule data or vice versa (Cho 2009) to remove mismatches between cost
and schedule data.
An integrated Control Account (CA) is derived from three
information structures, Work Breakdown Structure (WBS), Cost Breakdown
Structure (CBS), and Organization Breakdown Structure (OBS). Thus, the
integrated CA is referred to a combination of zone and element from the
WBS, corresponding operations from the CBS, and organizations from the
OBS responsible for the operations. In other words, an integrated CA
must encompass all relevant information units such as element,
operation, location, and organization to resolve the mismatch. As a
result, an integration model requires a large number of CAs, which form
a complex data structure. Additional effort to generate CAs (Deng, Hung
1998) and the overhead costs to control field data (Rasdorf, Abudayyeh
1991; Deng, Hung 1998; Jung, Woo 2004) created resistance to wide
adoption of the integrated approach. It is paradoxical that project
managers resist the integration because of the additional work required,
while the integration system was designed to reduce the managers'
workload.
Rasdorf and Abudayyeh (1991) proposed an automatic data acquisition
method using bar codes and a relational database management system to
reduce the overhead effort of data acquisition and processing. Jung and
Woo (2004) proposed a flexible WBS method to minimize the number of CAs.
On one hand, these approaches improved the practicability of an
integration system, but on the other hand, they were still limited to
apply to all construction projects.
Representations of cost and schedule control items and their levels
of detail vary according to both project characteristics including
project size, type, cost, duration, technical complexity, management
level, delivery system, contract type, and management policies (Jung,
Woo 2004), and situational characteristics including development stage,
delivery method, available resources and constraints, contractual
relationship, construction method, developer's experience,
perspective and knowledge, etc. Considering these characteristics, a
single representative integration method might not be suitable for all
construction projects.
While previous efforts towards integration have usually focused on
building projects, they are rarely applied to large construction
projects such as tunnels, roads, railroads, or bridges. These are
characterized by relatively small numbers of elements and organizations,
horizontally repetitive operations, and construction in open spaces. The
different characteristics of these large construction projects might
positively impact on the integration method. In the following sections,
we propose a method to reduce the required number of integrated CAs and
the overhead effort in data processing, focusing on excavation works in
NATM tunnel projects.
2. Cost and schedule control in an NATM tunnel
2.1. Characteristics of NATM tunnels
The New Austrian Tunnelling Method (NATM), also commonly referred
to as the Sequential Excavation Method (SEM) uses the inherent strength
in the rock mass to support the roof during excavation. Because this
self-supporting capability achieves economy, flexibility in uncovered
ground conditions, and dynamic design variability, NATM is widely
applied for underground structures. NATM tunnels are largely dependent
upon round length, types of support, and ground conditions such as shear
strength, deformation, and groundwater level. Excavation works
accounting for more than fifty percent of the project budget and
consuming roughly sixty percent of the total project duration are
critical management points determining a project's success.
Excavation methods and supporting patterns are predefined using
"types" in the planning phase. As ground conditions
deteriorate, the type number increases, the number of supporting
processes (such as rock bolts, steel plates, and forepoling) increases,
and the length of excavation per cycle decreases from 3.5 metres to 1.0
metres. For example, Type-I is the lowest support type; it is applied in
favourable ground conditions, allows full-face excavation, and usually
advances by 3.5 se or more per cycle. On the other hand, Type-VI allows
only partial excavation and advances by 1.0 metres or less. Type-VI
tunnels are excavated in two stages independently. First, the top half
of the tunnel (or heading) is excavated, and then the bottom half (or
bench) is excavated, because prompt reinforcements are required to
protect against unfavourable ground conditions. The standard types
provide a common platform for project participants to communicate in the
field. A tunnel project may have several drift patterns (e.g. Type-I,
Type-II), and more than six types have been recommended to cope more
flexibly with various ground conditions.
A standard type of excavation is implemented by cyclic operations
such as boring, charging, blasting, mucking, reinforcing, and
shotcreting operations. Each operation in the repetitive series must
usually be completed before the subsequent operation can start
(Department of Transportation 2009). Twelve-hour shifts and two cycles
of excavation per day are typical in a portal. Thus, four shifts of
excavation are performed in a two-portal tunnel. Because one set of
equipment and crews is usually responsible for the four cycles, smooth
operational transitions from one portal to the other might be quite
important for efficient cost and schedule control.
[FIGURE 1 OMITTED]
2.2. Type-based cost and schedule control
Because of the limitations in investigating precise ground
conditions, excavation plans in the design phase can change frequently
during the project execution. Project management schedules are used
widely to assess the change and to compare the current status with the
plan. Figure 1 shows a project management schedule used for underground
construction. The big table is divided into a plan table and an actual
table. The plan table includes: (1) a station representing the
excavation point; (2) a planning type denoting an excavation and support
method; and (3) the total excavation length for the planning type. The
actual table includes: (1) an actual start date for the corresponding
station; (2) the implemented type; and (3) the total excavation length
of the actual type.
Table 1 shows a type-based unit costing system that is used in a
real tunnel project. These type-based unit costs are derived from the
company's own historical database. In the type-based system, cost
and schedule control is implemented by comparison between planned and
actual type for a given construction period. For example, Type-VI and
Type-V were designated for the first 28 metres of excavation in the plan
in Figure 1. However, only Type-VI was actually performed because of
unfavourable ground conditions. On April 4, a project manager calculates
the change's negative impact: the tunnel is 3.2 metres behind
schedule and KRW 47 203 200 over budget. Because both cost and schedule
control are implemented based on types, it can be regarded as a
type-based cost and schedule control system; in short, a type system in
the paper.
2.3. Limitations of the type-based control system
The type system consists of several cost items listed in BOQ. A
cost item can be divided into several operations according to equipment
requirements. Table 2 describes the cost items used in the BOQ and their
required operations for the Type-IV top cycle. The type system can be
delineated into cost item levels composed of a bundle of operations.
Three limitations of the type-based control system were identified
in our study. First, the type system does not allow project managers to
track individual performances in particular periods of time. Cost
control relies on the level of detail used in the BOQ and schedule
control relies on the productivity rate of an operation performed by a
crew. Meanwhile, the type system's data are at a too broad level
because all operational performances are lumped together. This broad
level is rarely appropriate for bills of quantity and detailed schedule
control. Accordingly, instead of ad hoc management from the type system,
a more detailed level of control system is required for a more
systematic project control.
Second, the type system does not support smooth transitions between
operations. As previously mentioned, excavation works are implemented by
a series of independent operations, each of which must be completed
before the next can start (Department of Transportation 2009).
Usually, one set of equipment and crew implements a two-portal
tunnel. At the operational level, a charging car is used several times
at both tunnel faces to survey and mark, charge, scale, install steel
ribs, and install rockbolts. To complete two cycles within 12 hours, the
one set of equipment and crew must frequently move from one portal to
the other considering the operational sequence, resources availability,
and work space. Because the following factors influence the operational
planning, smooth transitions are generally quite complex:
1) Two portals: e.g. Southbound and Northbound Tunnel;
2) Eleven types: Type-I, II, III, IV-Top, IV-Bottom, V-Top,
V-Bottom, VI-Top, VI-Bottom, VI-1-Top, and VI-1-Bottom;
3) Thirteen operations: surveying and marking, explosive drilling,
charging and blasting, ventilating, scaling, loading and hauling, muck
cleaning, steel rib installing, cast lining, floor cleaning, rockbolt
drilling, rockbolt installing, and forepoling;
4) Different quantities according to the applicable type;
5) Inconsistent and ever-changing work conditions: the distance
from a tunnel face to a muck pile or a batch plant, access road
conditions, equipment efficiency and productivity rate according to
excavation progress, equipment failure, relationships with
subcontractors, complaints from neighbouring residents, etc.
In the type system, the assembly of operations is unable to
represent operational levels of detail, so ad hoc decision-making
relying on a project manager's experience, knowledge, and intuition
prevails in practice. Thus, a more detailed control system enabling
project managers to track current work status should be designed to
support the smooth transitions.
Third, the type system seems to be convenient to control a project,
but it requires considerable effort to derive meaningful data, e.g.
feedback on a subcontractor's performance, productivity rate,
equipment efficiency, and effective cycle times of crews. An efficient
control system enables project managers to understand who is going to do
what, and how, where, why, and when they are going to do it.
Unfortunately, supporting data in the type system are distributed in
various documents and software applications with different forms,
structures, and points of view. Although the type system might be quite
adequate to support a broad level of control, additional effort is
required to manipulate data for deriving meaningful feedback on project
performance. Eventually, the type system triggers inaccuracy,
inefficiency, and inconsistency in information flow and interrupts
systematic project management.
So far, we have reviewed several limitations of the current
approach to cost and schedule control. Excavation works are described by
several predefined supporting types and employ sequential operations at
multiple places at the same time. While various field characteristics
should be understood by planners and reflected by constructors in the
project execution phase, the type system does not allow project managers
to handle the characteristics. These limitations fundamentally originate
from the too broad level of cost and schedule control. Thus, an
operational-level cost and schedule control system should be considered
as a solution to overcome the limitations.
3. Information technologies required for integrated cost and
schedule control
An NATM tunnel is excavated by repetitive cyclic operations, as
described in Table 2. Every operation is performed at the tunnel face
and by a set of equipment. Thus, we hypothesize that the operations and
their related data can be identified by detecting which equipment is at
the tunnel face. For example, a project manager can figure out that a
shotcreting operation is in progress if an information system shows that
a shotcrete machine and a ready-mix concrete truck are located at the
tunnel face. The running time of a piece of equipment can provide the
start and end times of the operation: shotcreting.
Based on an analysis of the characteristics of excavation works, we
propose an automatic data processing system (ADPS) that identifies the
locations of equipment in tunnels. To accomplish this, the following
systems are required: 1) a real-time location system to detect an
equipment location; 2) a network system to deliver the location signals
from the tunnel face to a field office; and 3) an information
transformation system to convert the location signals to cost and
schedule data. Based on extensive reviews of currently available
technologies, this study adopts both a real-time location system (RTLS)
using a time of arrival (TOA) method to identify the location of
equipment in tunnels and a Wi-Fi wireless mesh network to deliver
signals to a field office.
3.1. Real-time location identification system
Location identification technologies include the global positioning
system (GPS) using satellites, location-based systems (LBSs) using a
mobile base station, and RTLSs using an indoor network (Kim 2011).
Because GPS and LBS are not available inside tunnels, an RTLS was
adopted in our proposed system for identifying equipment locations
within the tunnel, while a GPS receiver was used for locations outside
the tunnel. RTLS technology is used to monitor the locations of assets,
materials, and people in real time using location tags and readers
(Sadeghpour 2006). The RTLS transmits real-time location data using
readers from tags such as passive, semi-passive, or active
radio-frequency identification (RFID) tags. The real-time location data
showing the routes of tag movement are transmitted to a network system.
An active tag system was adopted for tracking locations of
equipment in a variety of situations, particularly those appropriate to
a tunnel's long, linear structure. According to measurement
methods, location identification technologies can be divided by angle of
arrival (AOA), time of arrival (TOA), time difference of arrival (TDOA),
and received signal strength indication (RSSI). Among these, TOA was
adopted for the proposed system because it allows relatively precise
measurement values in a wide range of areas. Figure 2 illustrates a
cell-based topology to utilize the TOA algorithm inside a tunnel. The x-
and j-axis values representing the equipment location are calculated by
the times of arrival of radio-frequency (RF) signals in a cell composed
of four anchor nodes and one mobile node.
[FIGURE 2 OMITTED]
3.2. Wireless mesh network system
A network system is necessary to transmit the location signals from
the RTLS to a field office. Tunnels are usually remote from towns and
pass under mountains, where installation of all means of communication,
including telephone and the Internet, is quite difficult. As a result,
communication problems are typical in tunnel projects and communication
is intentionally minimized. However, systematic project control relies
greatly on efficient communication among project participants. A
wireless network system was adopted to transmit location signals,
considering that tunnel projects include large, extensive construction
fields and an unfavourable work environment involving blasting,
rockslides, and underground water.
Examples of network technologies include wireless local area
networks, Bluetooth, ZigBee, and Ultra Wide-Band. Specifications
appropriate to tunnel projects include: a topology appropriate for a
long, narrow, linear structure; durability against blasting operations;
transmission distances as long as possible; low installation cost; and
ability to transmit signals in real time. A wireless mesh network (WMN)
was adopted as the best solution to match the specifications. A WMN is a
network system made up of routers, nodes, and gateways organized in a
mesh topology connecting all the terminals and relaying signals to
users. Advantages of WMNs include an ability to switch routes, allowing
failures in nodes, relatively low cost, and easy installation. Figure 3
illustrates the overall system architecture of a wireless mesh network
installed in a tunnel project.
[FIGURE 3 OMITTED]
4. Methods to control cost and schedule data from real-time
location of equipment
4.1. Operation-based information unit
To overcome limitations of the type system, an operation-based
information system was designed to measure cost and schedule data.
Figure 4 illustrates an example of an operational-level schedule,
presenting one cycle of excavation works for a two-portal tunnel. The
schedule shows that Type-I for the southbound tunnel and Type-IV-Top for
the northbound tunnel were implemented as of 30 April 2010.
[FIGURE 4 OMITTED]
The dotted line in Figure 4 indicates that the load and haul
operation in the southbound tunnel and the survey and mark operation in
the northbound tunnel were in progress at noon. Based on the southbound
tunnel schedule, project managers coordinate the northbound tunnel
schedule to eliminate a bottleneck in operations. Thus, the
operation-based schedule not only allows project managers to avoid
operational conflicts, but also shows the actual feasibility of daily
scheduling. A smoother transition might be possible depending on how the
project manager plans the operational sequences and utilizes float time
(for example, the prepare operation in Fig. 4). In addition, the
operation-based control system shows the productivity of an operation
and the efficiency of equipment usage, the fundamental data of
short-term project planning.
Table 3 describes the unit costs of one cycle of Type-IV-Top. This
unit cost is based on the cyclic quantities of Type-IV-Top and it can be
utilized to account for cost data in the project budget. For example,
when the "shotcrete clean" operation is completed (Table 2),
the one-cycle unit cost (NE1102; KRW 1 492.963) of the shotcrete
operation is added to the project budget. The one-cycle unit cost makes
cost control possible at the operational level, because the respective
types are performed by identical cyclic operations.
4.2. Conditions for cost and schedule integration
Integrated cost and schedule control can be defined as an effort to
combine both kinds of information in a single controllable account. As
previously mentioned, an integrated cost and schedule control system
requires a great number of integrated CAs, including spatial, elemental,
operational, and organizational information units. A single integrated
item for a building project frequently includes physical and functional
properties and sometime requires structural, architectural, and
electrical properties with various levels of detail. A building project
typically generates more than several thousand integrated CAs. Deng and
Hung (1998) mentioned that a large number of CAs requires an increase in
the labour force and heavy overhead costs. Consequently, the large
number of control items and a complex data structure are the main
obstacles to wide adoption of the integrated approach. Although the
integrated approach provides easy access to cost and schedule control
for project managers, considerable effort may still be required to
acquire, track, and analyze operational data
On the other hand, an integrated CA for excavation works requires a
single linear space represented as a station, no specific element, and
several cost items consisting of a bundle of operations. A small number
of CAs an integrated approach might be easily accomplished. Existence of
eleven types and on average twelve operations means that the integrated
approach to excavation works would require 132 CAs, which must be a
relatively small number compared with a building project.
In addition, the integration supports smooth transitions and
meaningful analysis during project execution because an integrated CA is
based on the operation level. In summary, the characteristics of tunnel
excavations make operational-level cost and schedule integration to be
possible with a relatively small number of integrated CAs.
However, despite a small number of CAs, data acquisition,
transaction, and analysis in the fields might be far from realistic,
when considering the large construction field, unfavourable access, long
distances from a field office to jobsites, and limited administrative
staff. In this respect, we propose an Automatic Data Processing System
(ADPS).
4.3. Algorithms to convert location signals to operational-level
cost and schedule data
In Section 3, we described a real-time location identification
system to track equipment near a tunnel face and a wireless mesh network
to transmit location signals from the tunnel to a field office. The
location signals of equipment are converted into operational-level cost
and schedule data in the proposed system. The following algorithms are
part of the proposed model:
1) A series of repetitive operations, each of which must be
completed before the subsequent operation can start;
2) Respective operations are implemented by a specific set of
equipment at the tunnel face;
3) The running time of an item of equipment can provide the start
and end time of an operation;
4) A cost item used in BOQ can be represented by a bundle of
operations;
5) A type cost can be represented by the sum of one cycle of unit
costs.
Figure 5 illustrates a prototype model for the ADPS that we are
proposing. The RTLS detects the locations of a shotcrete machine and
ready-mix concrete truck at the southbound tunnel face (Station 818 m)
and the WMN transmits the signals from the RTLS to the field office. The
prototype model automatically realizes that the cast-lining operation is
now in progress and the installing steel rib operation has just been
completed from the signals that show a charging car is moving back (Fig.
5-A). Then, the one-cycle unit cost of the installing steel rib
operation (KRW 1 825 567) described in Table 3 is included in the BOQ in
the prototype model (Fig. 5-B). Next, the accumulated unit costs
comprise the Type-IV-Top cycle cost (Fig. 5-C) and finally, the
type-based cost is included in the project budget (Fig. 5-D). The
proposed system thus enables project managers to access four levels of
detail including the operation, bills of quantity, type, and project
budget level; to accomplish integrated cost and schedule control; and to
automate all required processes including data acquisition, tracking,
and analysis.
5. Pilot project
The practicability of the proposed ADPS for integrated cost and
schedule control was tested using a pilot project. The Samtan 1 tunnel
is the longest tunnel (2 645 metres in the Samchuk direction and 2 619
metres in the Jeachen direction) in the Samchuk--Jeachen road
construction in South Korea. The NATM tunnelling method and seven
standard types were employed in the project. Total duration was 48
months for the two-portal tunnels with twolane roads. Figure 6
illustrates the prototypical longitudinal profile of the Samchuk-Jeachen
road construction and shows overall project deliverables of the pilot
project.
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
5.1. System application to tunnel fields
Figure 7 illustrates the overall system application in the pilot
project. The RTLS consisted of location anchors (Fig. 7-B) installed
inside the tunnel and a location tag installed in a sport utility
vehicle (Fig. 7-C). Six anchors were installed at 9 metre width and 30
metre, 50 metre, and 70 metre lengths respectively, as shown in Figure
7, to minimize location errors according to longitudinal changes. The
vehicle substituting for construction equipment moved within the
rectangular fixed areas (section) and its movement was measured by a
total station to check for errors in the RTLS. Anchor nodes were
designed to transmit real-time location data 30 times per minute.
Figures 7-E and 7-F illustrate the WMN installed inside and outside the
tunnel at 100 metre intervals. Table 4 describes system specifications
applied in the pilot project.
[FIGURE 7 OMITTED]
5.2. Test results
Several rounds of testing are summarized in the following material.
First, signals from the anchors in every section were detected to
identify equipment locations precisely, and their errors were minimum 45
centimetres, maximum 315 centimetres, and average 150 centimetres when
compared with measurements by a total station. The range of errors must
be low enough to identify equipment locations in the tunnel and to
regard a set of equipment at the tunnel face as an operation. Second,
the range of errors is slightly impacted by longitudinal changes. A
later test in a section of more than 100 metres and a test using an
extended Kalman filter to reduce errors were considered to improve
technical and economic feasibility.
[FIGURE 8 OMITTED]
Figure 8 shows the pilot test results in the 70 metre section. The
dotted line indicates the planned movements and the staggered line shows
the actual movements of the vehicle location tag in the narrow-width
tunnel and at the unfavourable ground conditions.
Conclusions
The purpose of cost and schedule control can be defined as the
optimization of the resources required by tunnel projects: large amounts
of material, labour, and equipment for long periods. Thus, cost and
schedule control must be realistic and must reflect all the restrictions
that are imposed on the project (Department of Transportation 2009).
Based on the analysis of characteristics of NATM tunnel excavation
works, this study raises several limitations of the type system: the
level of control is too broad; cost and schedule data are too
distributed; management is ad hoc; and it is difficult to derive a
meaningful database. The type system relies on a project manager's
experience, knowledge, and intuition, rather than quantitative analysis
of field data. This study emphasizes the necessity of operational-level
cost and schedule integration and control. While an integrated approach
might contribute greatly to the overall enhancement of project
management (Jung, Gibson 1999), it requires considerable overhead effort
to acquire, track, and analyze data in the project execution phase.
Based on the motivation that the overhead effort has been the main cause
of resistance to adopting the integrated approach, we have proposed an
ADPS for cost and schedule integration.
The proposed system is composed of an RTLS, a WMN, and a prototype
model for integrated cost and schedule control. The RTLS has location
tags installed in construction equipment and location anchors
calculating the relative position of a tag within a section. From the
signals transmitted by the WMN, the prototype model identifies equipment
locations in tunnels and converts the location signals into integrated
cost and schedule data through the inherent algorithms. Based on the
operational level of an integrated Control Account (CA) having cost and
schedule values, the prototype model provides various levels of
integrated control, i.e. operation, BOQ, type, and project budget level.
Effect of IT convergence using sensor technologies has been limited
in the construction industry. This study introduces novel uses of sensor
technologies to facilitate cost and schedule control. The proposed
system can reduce the considerable overhead effort of implementing cost
and schedule integration through automating data processes and reducing
the required number of integrated CAs. The integrated approach using
sensor technologies thus offers project managers easy access to
operational-level cost and schedule control, smooth transitions,
productivity analysis, and improved equipment efficiency. In addition,
the operational-level database allows efficient construction planning
and management considering field characteristics. Technical and economic
feasibility studies to install RTLSs and WMNs in construction sites,
extension of the application range to whole tunnel projects, and
quantitative measurements of advantages of the proposed system remain
future study areas. We expect that the algorithms proposed by the study
will become a capstone to an approach to more systematic construction
management.
doi:10.3846/13923730.2013.801907
Acknowledgement
This study is part of a research project sponsored by the Korea
Institute of Construction and Transportation Technology Evaluation and
Planning (KICTEP) under the Grant No. 09CCTI-B052843-01. The support is
gratefully acknowledged.
References
Cho, D. G. 2009. Construction Information Database Framework (CIDF)
for integrated cost, schedule, and performance control: PhD Thesis.
University of Wisconsin-Madison, United States.
Deng, M. Z. M.; Hung, Y. E. 1998. Integrated cost and schedule
control: Hong Kong perspective, Project Management Journal 29(4): 43-19.
Department of Transportation (DOT) 2009. Technical manual for
design and construction of road tunnels--civil elements. Report No.
FHWA-NHI-10-034. Washington, DC: National Highway Institute, U.S.
Fleming, Q. W.; Koppelman, J. M. 2005. Earned value project
management. 3rd ed. Project Management Institute, P.A.
Hendrickson, C.; Au, T. 1989. Project management for construction:
fundamental concepts for owners, engineers, architects, and builders.
Englewood Cliffs, NJ: Prentice Hall.
Jung, Y.; Gibson, G. E. J. 1999. Planning for computer integrated
construction, Journal of Computing in Civil Engineering 13(4): 217-225.
http://dx.doi.org/10.1061/(ASCE)08873801(1999)13:4(217)
Jung, Y.; Woo, S. 2004. Flexible work breakdown structure for
integrated cost and schedule control, Journal of Construction
Engineering and Management 130(5): 616-625.
http://dx.doi.org/10.1061/(ASCE)07339364(2004)130:5(616)
Kim, D. W. 2011. Safety management system for a large construction
project using real-time location system. PhD Thesis. Korea University,
South Korea.
Kim, J. J. 1989. An object-oriented database management system
approach to improve construction project planning and control. PhD
Thesis. University of Illinois, United States.
Perera, A.; Imriyas, K. 2004. An integrated construction project
cost information system using MS[TM] Access and MSTM Project, Journal of
Construction Management and Economics 22(2): 203-211.
http://dx.doi.org/10.1080/0144619042000201402
Rasdorf, W. J.; Abudayyeh, O. Y. 1991. Cost- and schedule-control
integration: issues and needs, Journal of Construction Engineering and
Management 117(3): 486-502.
http://dx.doi.org/10.1061/(ASCE)07339364(1991)117:3(486)
Sadeghpour, F. 2006. Real time location system for construction
site management, in Proc. of Joint International Conference on Computing
and Decision Making in Civil and Building Engineering, Montreal, 14-16
June, 2006, 37363741.
Teicholz, P. M. 1987. Current needs for cost control systems, in
Ibbs, C. W.; Ashley, D. B. (Eds.). Project controls: needs and
solutions, Proc. Specialty Conference, ASCE, 47-57.
Daegu CHO (a), Hunhee CHO (a), Daewon KIM (b)
(a) School of Civil, Environmental and Architectural Engineering,
Korea University, Anam-Dong, Sungbuk-Gu, Seoul, S. Korea
(b) Department of PLMInnovation Group, Samsung C&T Corporation
Yoksam 2-Dong, Gangnam-Gu, Seoul, S. Korea
Received 26 Jan 2012; accepted 07 May 2012
Corresponding author: Hunhee Cho
E-mail: Hunhee CHO hhcho@korea.ac.kr
Daegu CHO. Research Professor at School of Civil, Environmental,
and Architectural Engineering, Korea University, South Korea. He holds a
PhD from the University of Wisconsin-Madison, USA. His research has
centered around four areas: construction management, information
technology, database management, and Building Information Modelling
(BIM). With respect to BIM, he is developing an integrated cost and
schedule system appropriate to the intelligent objectoriented modelling.
Hunhee CHO. Associate Professor at School of Civil, Environmental,
and Architectural Engineering, Korea University, South Korea. Dr. Cho is
working on multiple research projects in South Korea: a conceptual cost
estimating system for buildings, an optimal temporary hoist planning
system in high-rise building construction, a synchronous hydraulic
jack-up system, and more. Fields of research interests include
construction automation and robotics, high-rise building construction,
and cost control system.
Daewon KIM. Senior engineer at Samsung Construction and Technology,
South Korea. Dr. Kim has extensive research experience, specifically in
a location identification system such as GPS, LBS, and RTLS. His main
research interests include a safety management system using information
and cutting-edge sensor technology.
Table 1. Type-based unit costs used in Samtan 1 Tunnel
Standard Excavation length Unit cost per
type per cycle (m) cycle (wons)
TYPE-I 3.5 9,607,500
TYPE-II 3 8,547,000
TYPE-III 2 7,342,000
TYPE-IV 1.5 7,726,500
TYPE-V 1.2 8,734,800
TYPE-VI 1 11,685,000
TYPE-VI-1 1 11,677,000
Table 2. Cost items and operations of Type-IV top cycle
Cost Item Operation Equipment Time
in BOQ (ea.) (min)
Survey & Charging car (1) 20
mark
NA1613 Drill Drilling jumbo (1) 80
Excavation Charge & Charging car (1) 60
blast
Ventilate 30
Scale Backhoe (1) 20
NA1920 Charging car (1)
Muck Load & haul Backhoe (1) 90
disposal Dump truck (5)
Clean Backhoe (1) 70
Loader (1)
NG2111 Install Charging car (1) 50
Steel ribs steel ribs
Cast Shotcrete machine (1) 80
NE1102 lining Ready-mix truck (1)
Shotcrete Clean Loader (1) 30
Dump truck (1)
NG1154 Drill Drilling jumbo (1) 20
Rock bolt Install Charging car (1) 60
rock bolts
Total 610
Table 3. One-cycle unit cost of Type-IV-Top
Code Operatio Unit Cycle
quantity
NA1613 Excavation [m.sup.3] 99.47
NA1920 Muck disposal [m.sup.3] 105.94
NX1304 Labour Times 1.00
NG1154 Rock bolt (L = 4.0 m) ea. 9.50
NG2111 Steel rib (50*20*30) NR 2.33
NE 1102 Shotcrete [m.sup.3] 7.87
Unit cost (Won) Total cost
Code
Materials Labour Equip. (wons)
NA1613 5,747 878 6,239 1,279,582
NA1920 1,328 649 1,051 320,786
NX1304 1,027,557 1,027,557
NG1154 21,396 129 29,700 486,638
NG2111 393,580 369,186 20,739 1,825,567
NE 1102 133,703 17,404 38,596 1,492,963
Table 4. System specifications applied in the pilot project
Applied System RTLS Location Anchor
Model GNC-TAG-V1 GNC-An-V1
Major 2.45 GHz RF 2.45 GHz RF (ISM
Specifications Wireless LAN Band)
Embedded Linux Atmega 128L
Cortex A8 (1 GHz) Main RS-232C JTAG
CPU
Applied System RTLS GPS
Model GNC-TAG-V1 UIGGUB02-ROO1
Major 2.45 GHz RF -160 dBM Sensitivity
Specifications Wireless LAN -160 dBM Tracking
Embedded Linux System
Cortex A8 (1 GHz) Main NMEA Protocol
CPU
Applied System RTLS WMN
Model GNC-TAG-V1 RouterStation Pro
Major 2.45 GHz RF SuperRange 2 Network Inter-
Specifications Wireless LAN face Card
Embedded Linux AR7100 Network Processing
Cortex A8 (1 GHz) Main Unit
CPU 100-500 m Coverage
