P1: OSO
JWDD004-18 JWDD004-Halpin-v6 August 17, 2005 20:38
Chapter 18
Construction Operations
PROSIDYC: Simulation Program for Construction Operations
The Need
Currently 3D-modelling is the trend in the simulation area. However,
Construction of floating caissons utilized
PROSIDYC as a tool to increase
production.
developing 3D models of construction operations is very complex and time
consuming. In general, the study of construction operations requires a tool
that provides solutions without requiring the input of copious amounts of
data. In order for a construction company to use a simulation tool, the
methodology has to be presented in a very simple and graphical context.
Pictorial and schematic tools are easily accepted. In contrast, if the
methodology appears to be too theoretical or analytical it will be avoided
by construction practitioners.
The Technology
PROSIDYC is a system for simulating construction operations jointly developed by the
Planning and Methods Unit of Dragados y Construcciones, Madrid, Spain and the
Division of Construction Engineering & Management at Purdue University.
PROject SImulation Dragados Y Construcciones (PROSIDYC) is a computer based
system for analyzing construction job site production processes. It is used to improve
productivity in the field by studying resource utilization and cycle times and identifying
opportunities for production improvement. PROSIDYC uses the CYCLic Operations
NEtwork (CYCLONE) modeling format. A set of graphical modeling elements are
utilized to develop a network model of the process of interest. The model identifies
waiting or delay states as well as active productive states. The computer program allows
the modeler to identify resources which are underutilized and bottlenecks in the process.
The use of this approach has achieved 100% success in productivity improvement on
the processes studied. Improvements range from 30% to 200%. Data support the fact that
Harbor site layout.
PROSIDYC/CYCLONE
Flow diagram.
Caisson Construction Valencia, Spain.
18-1
18-2 Chapter 18 Construction Operations
for every hour of analyst time used, a saving of $2,000 is realized. Therefore, for 100
hours of engineering-time invested, a saving of $200,000 is achieved. PROSIDYC was
used to achieve major cost savings in the massive breakwater in Valencia shown here.
18.1 MODELING CONSTRUCTION OPERATIONS
In Chapter 1, the hierarchy of construction management was described as shown in Figure
18.1. Activities define the structure of projects. The basic building block required to understand
and analyze construction operations is the work task. A meaningful description
of a construction operation requires the definition of the basic work tasks and the manner
in which the available resources (e.g., cranes, crews, materials, etc.) perform or process
through the work tasks. In this sense individual resources can be said to traverse or flow
through work tasks. The sequential and logical relationships between the various work tasks
define the technology being used. The actual working of the operation can then be described
by locating and monitoring, from time to time, the various resource entities as they dynamically
traverse the static structure of the operation. A simple graphical modeling system
can be used to analyze the work flow and develop the productivity for a given construction
operation.
18.2 BASIC MODELING ELEMENTS
A modeling format for flow modeling construction operations can be developed using four
graphical symbols:
Figure 18.1 Hierarchical levels in construction management.
18.3 Building Process Models 18-3
Modeling Element Name of Element
NORMAL
COMBI
QUEUE
ARROW
Description of Modeling Element
The normal work task modeling element can
commence as soon as a unit (e.g. resource)
arrives from a preceding element; it is
unconstrained.
The constrained work task modeling element
requires multiple resources (e.g. cranes,
crews) before it can begin. A combination of
resources is required to start. Otherwise similar
to the normal work task modeling element.
The idle state of a resource entity symbolically
represents a waiting location (i.e. a queue)
where resources wait prior to being combined.
The directional flow modeling element shows
the logical flow of resources.
Figure 18.2 Basic modeling elements.
1. Active-state square node representing a work task
2. Idle-state circle representing a delay or waiting position for a resource entity
3. Directional flowarrowrepresenting the path of a resource entity as it moves between
idle and active states
The symbols used (see Fig. 18.2) for each modeling element are designed to be simple
and helpful in developing schematic representations of the construction operation being
modeled. Two basic shapes (squares and circles) are used to model active and waiting resource
states; together with directed arrows (arcs) for resource flow direction, they help to
provide a quick visual grasp of the structure of a construction operation. These symbols
are the basic modeling elements of the CYCLONE (CYCLic Operations Network) modeling
system. They are used to build networks of active and idle states to represent cyclic
construction processes.
It is convenient to distinguish between the unconstrained (i.e., normal) work task and
the constrained (i.e., requiring the initial satisfaction of conditions) work task. While all
work tasks are modeled schematically as square nodes, the constrainedwork task is modeled
as a square node with a corner slash. Thus a total of four symbols is required for the modeling
of the structure and resource entity flow of construction operations (see Fig. 18.2).
The active working-state models are the NORMAL and COMBI modeling elements.
Both have a square-node format and model work tasks. Since the work task is the basic
component of a construction operation, it should be chosen so that its name or description
is sufficient to convey to a crew member or supervisor the nature, technology, work content,
and resources needed to fulfill the work task.
Simple examples of work task activities are breaking open brick pallets, preparing
column formwork, and loading trucks with front-end loaders. The definition of a work task
thus requires a verbal description, an indication of the resource entities involved, and a
definition of the time required (duration) to complete the task.
18.3 BUILDING PROCESS MODELS
The relative sequence and logic of the work tasks and processes that make up a construction
operation constitute the technological structure of the operation. The modeling elements
can be used in a variety of patterns to model construction operations.
18-4 Chapter 18 Construction Operations
Figure 18.3 Schematic outline of earth-moving operation.
As an example, consider the development of a model for an earth-moving operation
that involves the loading of trucks with earth for transport to a dump area. A pictorial
representation of the operation is shown in Figure 18.3; it uses a front-end loader, some
trucks, and earth.
In order to develop the framework of the earth-moving operation, it is necessary to
identify the major resources involved (i.e., trucks, front-end loader, and soil) and establish
the various states (i.e., both the active working states and the passive waiting states) that
the resources traverse in their work assignment paths and cycles. Finally, the integration of
the resource paths and cycles establishes the basic structure of the operation.
Each truck, for example, is idle while it waits (i.e., queues) for loading; it enters active
working states when it is being loaded, dumping, traveling loaded to the work site, and
returning empty for another load. A simple model of this work cycle is shown in Figure
18.4a using a single COMBI ÒLoad truckÓ work task that requires earth and a front-end
loader for initiation; three NORMAL work task elements, ÒLoaded truck travel,Ó ÒTruck
dump activity,Ó and ÒEmpty truck returnÓ; a single QUEUE element, ÒJoin truck queueÓ;
and five arrows indicating the logical relationships between the various truck states.
18.4 STRUCTURE OF CONSTRUCTION OPERATIONS
The front-end loader can be initially modeled by a unit cycle involving the active-state
COMBI element ÒFEL (front-end loader) loading,Ó the idle QUEUE element ÒFEL idle,Ó
and two entity flow directional logic arrows (see Fig. 18.4b).
In Figure 18.4c, a soil path model is shown that uses a source QUEUE node Òsoil
stockpileÓ and a sink destination soil dump QUEUE node together with a COMBI work
task ÒLoaded into truckÓ and NORMAL work tasks ÒTransport by truckÓ and ÒDumpedÓ
to portray the soil involvement in active work states. Finally, four directional arrows are
required to develop the path structure.
The integrated model incorporating the truck and front-end loader cycles together
with the soil path from stockpile to dump is shown in Figure 18.4d. Model integration is
achieved by combining or ÒoverlayingÓ active states which are common to two or more
resource cycles. For instance, Òload truckÓ in the truck cycle is the same active state (work
task) as Òloaded into truckÓ in the soil cycle and ÒFEL LoadingÓ in the loader cycle. These
three states are combined in the integrated model to be one active state ÒLOAD.Ó
18.5 Modeling Procedure 18-5
Figure 18.4 Development of operational structure; (a) truck cycle, (b) loader cycle, and
(c) earth-moving operation.
This model can be used as the basis for further development involving dump area
spotters and queues, dozer stockpiling operations, and truck maintenance, as well as the
basis for further detail such as a more precise description of the front-end loader loading
cycle. An extension of the skeletal structure of the earth-moving operation to include dozer
stockpiling and spreading operations together with a dump spotter foreman is shown in
Figure 18.5. A counter element (represented by a flag) has been added to note the point in
the network at which production will be measured.
The foregoing presentation illustrates that the structure of construction operations can
be developed and illustrated through the proper use and labeling of the basic modeling
elements. The model structure can be used in explaining the construction technology and
construction method of the construction operation to field personnel and managers.
18.5 MODELING PROCEDURE
The procedure for modeling a given construction process involves four basic steps. The
steps, as shown in Figure 18.6, are as follows:
1. Flow Unit Identification. As a first step, the modeler must identify the system
resource flow units (e.g., resources such as earth, cranes, crews, etc.) that are
18-6 Chapter 18 Construction Operations
Figure 18.5 Model of earth-moving operation.
Figure 18.6 Steps in model formulation.
18.7 Concrete Pouring Using a Crane and Bucket 18-7
relevant to system performance and for which transit time information is available
or obtainable from the field. The selection of the flow entities is very important
since it dictates the degree of modeling detail incorporated into the operation model.
2. Development of Flow Unit Cycles. Following identification of the flow units that
appear relevant to the process being modeled, the next step in model formulation
is to identify the full range of possible states that can be associated with each flow
unit and to develop the cycle through which each flow unit passes.
3. Integration of Flow Unit Cycles. The flow unit cycles provide the elemental building
components of the model. The structure and scope of the model are obtained by
the integration of activities that are common to two or more flow unit cycles.
4. Flow Unit Initialization. In order to analyze the model and determine the response
of the system model, the various flow units involved must be initialized, both in
number and initial location. Flow units are initialized at idle or waiting states (i.e.,
Queue Nodes).
Models developed using these basic steps must also be modified to provide for monitoring
of system performance. This leads to a fifth stage of system design in which special
elements for determining system productivity, flow unit characteristics, and other pertinent
information are included in the model structure. These features as well as the use of the
Web-CYCLONE program are described on the web at www.wiley.com\college\halpin.
18.6 TYPICAL REPETITIVE OPERATIONS
The key to modeling operations to determine productivity and balance among resources is
to identify processes that are linear and repetitive. At the production level many processes
are cyclic in nature and can be readily modeled using the CYCLONE modeling format.
In general, processes that are linear or evidence linear characteristics are repetitive and
good candidates to be modeled using a cyclic modeling environment. Some examples of
repetitive or cyclic construction processes are:
1. Concrete pouring
2. Structural steel erection
3. Slurry wall construction
4. Pile Construction (Driven and Augered)
5. Caisson Construction
6. Pipe laying
7. Brick and masonry work
8. Reinforced earth construction
9. Exterior panel installation
10. Window or glass-curtain wall installation
11. Tunneling and tunnel excavation
12. Precast concrete member erection
18.7 CONCRETE POURING USING A CRANE AND BUCKET
Concrete is one of the most commonly used materials in construction. Its versatility and
ease of placement have enabled it to hold its place in the market despite the development
of more sophisticated materials. Concrete may be used in many different ways on a project
to include monolithic foundation pours and precast panels to form the building exterior.
18-8 Chapter 18 Construction Operations
Concrete can be either cast in place on-site or precast off-site and transported to the project
for installation. The methods of placement at the site vary and depend on a number of
considerations such as (1) the placement location, (2) the desired speed of placement, and
(3) the types of equipment available. In this section, a simple model for the placement of
cast in place concrete using a crane-bucket system will be discussed.
It is assumed that forms and steel reinforcement are in place and that the system is not
constrained by the batch plant (i.e., the quantity of concrete available is not a constraint).
Required resources for the model include concrete hauling trucks, a crane with bucket(s),
vibrating and finishing apparatus, and a crew of laborers at the placement site.
To provide a context for this model, assume that the concrete is to be placed on
a paving job and that the concrete is batched at a site 2 miles from the paving site.
Rather than arriving at the site as a wet mix, five-at-a-time dry batches are carried in
an open-bay truck (with appropriate compartments) to a mixer near the paving site. The
batches are then dumped individually and sequentially into the skip of the mixer and
mixed sequentially, starting with batch 1 and ending with batch 5. As each wet batch
exits the mixer, it is dumped into a concrete bucket and lifted by a crane to the placement
location where it is dumped, spread, vibrated, and finished by a concrete crew.
A schematic diagram of the process is shown in Figure 18.7a. Although dry-batching
operations of this type are not common, this situation provides a good opportunity to
utilize various modeling features. A crane and bucket placement operation is shown in
Figure 18.7b.
A CYCLONE model of the process is shown in Figure 18.8. The model consists of six
cycles representing the various flow units involved. The units and cycles of interest are:
1. Batch plant
2. Trucks
3. Mixer
4. Crane
5. Bucket
6. Laborer crew for spreading, vibrating, and finishing
The process begins with trucks being loaded at the batch tower (COMBI 2). The trucks
consist of five compartments, which are defined by baffles or dividers that are pinned in such
a way that they can be released individually (one at a time) when the truck bed is elevated.
Dry batches are loaded into each of the five compartments. This requires five individual
loads at COMBI 2. The demand for five loads is generated using the GENERATE function
at QUEUE node 9. The GENERATE function takes a single truck unit arriving at 9 and
splits it into 5 units to be processed (i.e., compartments to be loaded).
When five batches are loaded, the CONSOLIDATE at FUNCTION node 3 assembles
the five loads into a single truck for travel to the mixer. Upon arriving at the mixer, the truck
is again reconfigured to represent the five dry batches using the GENERATE function at
QUEUE node 5. Each of the five dry batches are dumped sequentially into the skip of the
mixer.
The mixer processes the batches in sequence and converts them into wet batches for
transport to the placement site. The space in the mixer is represented by a flow unit at
QUEUE node 13. If one unit is initialized at QUEUE node 13, this means that the mixer is
a single-drum unit and only one batch at a time can be processed. If two units are defined
at QUEUE node 13, then the mixer is a dual-drum unit and two batches can be processed
simultaneously (in tandem).
Once a batch is dumped into the skip of the mixer, it is moved to the drum, where water
is added and it is mixed (NORMAL 10). Following mixing, it occupies space in the drum
18.7 Concrete Pouring Using a Crane and Bucket 18-9
Figure 18.7 (a) Dry-batch delivery and placement and (b) crane-bucket concrete placement.
18-10 Chapter 18 Construction Operations
Figure 18.8 Dry-batch and delivery placement model.
(QUEUE node 11) until it can be dumped to the concrete bucket. Therefore, the space in
the drum is not free until the bucket is filled at COMBI 12, and this is represented by the
feedback loop to QUEUE node 13 (the space QUEUE). That is, space becomes available
after the bucket is filled. This availability of space is required before the next dry batch can
be loaded into the mixer at COMBI 6.
Once all five batches on the truck have been loaded to the mixer at COMBI 6 is free to
return to the batch plant. The fact that the truck is empty (five batches loaded into the mixer)
is established by the CONSOLIDATE at FUNCTION node 7. The empty compartments are
reassembled into a single empty truck that returns to the batching power.
When the bucket is filled, it is available to be lifted by the crane to the placement
location in the pavement. If only one bucket is used, the crane and bucket can be considered
a single unit, and the separate QUEUE node 17 for ÒCrane readyÓ is redundant.
However, if two buckets are used, the crane is a separate unit. After swinging back, it
drops one bucket and picks up the other. This is more efficient since the bucket at the
mixer can be filled while the crane is swinging and placing the other. The second bucket
provides a storage point and keeps the crane active, allowing it to pick up the next loaded
bucket without having to wait during the ÒFill bucketÓ activity at 12. For this reason,
...
Januszek66