American Journal of Industrial and Business Management, 2012, 2, 116-127
http://dx.doi.org/10.4236/ajibm.2012.24016 Published Online October 2012 (http://www.SciRP.org/journal/ajibm)
Configuration Model for Automating Work System Design
Muhamad Arfauz A. Rahman1, John P. T. Mo2
1Faculty of Manufacturing Engineering, Universiti Teknikal Malaysia Melaka, Hang Tuah Jaya, Malaysia; 2School of Aerospace
Mechanical and Manufacturing Engineering, Bundoora, Australia.
Received May 5th, 2012; revised June 4th, 2012; accepted July 2nd, 2012
ABSTRACT
The design of an automation work system involves important choices concerning the type of system process as well as
the condition of the process. These are based on the requirements from the user. This work provides the development of
configuration model for automating the design of work system. The model consists of the extraction and execution
process of user requirements. It begins with the identification of the requirement statement. Once identified, the state-
ment is extracted and sorted accordingly into process type, number of item count and condition of the process. Cur-
rently, the model considers simple sorting and basic assembly process. The model continues to the selection of system
action and system component. At the end of the work, the user requirements are transformed into a set of system model
and eventually provide the desired system specification. At this stage, the model is represented in a symbolic flow
process.
Keywords: Automation; Configuration; System Actions; System Components; User Requirements; Work System
1. Introduction
In configuring the automation work system, Mo et al. [1]
believe that the goal is not about adapting one system to
another, but rather to develop an automation work system
from a task description independent of the other system,
and subsequently assign components to achieve the
specified function. Configuration of manufacturing sys-
tems is a strategic decision. However, many companies
have lost their production capabilities every time they
acquire a new system or modifying existing system. Re-
gardless of the cause, companies have a challenging
problem arise due to selecting new resources that fit their
future needs better [2,3]. This phase is a very critical
since each decision taken will directly affect the per-
formance of the new system at this level and therefore its
profitability in the future. Often, the information avail-
able at the early configuration stage is not detailed and is
sometimes uncertain. According to Matta et al. [4] this is
true especially when uncertainty of the market demand
need to be considered during the configuration of the
system. This is important since unexpected variations of
the volumes required by the market, or the introduction
of new products, can make the solution unsuitable to
fulfill the market requests. At the same time, the decision
must consider many system variables such as the process
type, number of iteration and the process condition. To
solve the problem, a simplified methodology that can
deal with all the aspects described above is necessary.
2. Configuration Phase
To understand the configuration process, phases involve
in the designing of work system must be understood and
followed. The following Figure 1 shows the configura-
tion and reconfiguration phases of work system.
The success of any automation work system configu-
ration and reconfiguration will depend upon how well the
system is executed at the beginning of the phase. The
figure shows the generic configuration process that was
concluded based on various methods on configuration
and reconfiguration process include Monfared and Wes-
ton [5], Stoin and Frumusanu [6], Wiendahl et al. [7],
Travaini et al. [8], and ElMaraghy et al. [9]. Variation in
customer demands indicates changes in the market. Spe-
cific method on the acquisition of the user requirements
and later transforming the information into system speci-
fications has not been done. In line with the changes,
Covanich and McFarlane [10] through their case study
believe that an easy and simple engine is required to ma-
nipulate the requirement from the customers. Changes in
market indicate changes in user requirements. Ferscha et
al. [11] have agreed that a low flexibility of the system is
amongst the challenges that limit the system’s capability
to adapt with new requirement. Therefore, new systems
need to be set up. However setting up a new system is
very costly. Often reconfigurable manufacturing system
(RMS) the current system is manipulated to suit with the
new requirement.

Configuration Model for Automating Work System Design 117
Figure 1. Work system configuration phases.
An appropriate system specification is an essential step
leading to the successful implementation of automation
work system configuration. Toni and Tonchia [12] had
earlier discovered that the system needs to be configured
and reconfigured accordingly from time to time in order
to adapt with the new user requirements.
In specifying system specifications, few components
need to be visualized. These include action of each proc-
ess known as system actions as well as the corresponding
components known as system component provided by
Rahman and Mo [13]. Once the specification has been
produced, the next process involves implementing the
system in term of hardware. The implementation stage
often requires manual step. At this stage, the research
requires diverse concept not limited to initial configura-
tion of a new automation work system but also to recon-
figure the existing system. According to Mo et al. [1], in
building an automation system, components required
may be associated through physical or non physical
specifications at different types. Due to various changing
needs, addition or removal of physical components in the
system will be severely affected, thus affecting the finan-
cial component as well. The first step to configure the
system is to design the system accordingly.
The next process involves the integration of the hard-
ware system with any software component. Various
method for integrating the component have been intro-
duced including through combined ontological represen-
tation of the low-level functionality at the high-level
control layer by Lepuschitz et al. [14] and through Re-
configurable Manufacturing Execution Systems Archi-
tecture (RMESA) by Huang et al. [15]. This includes
programmable logic controller or micro controller. The
final stage of the phase will be the execution part.
Currently, the acquisition process shown in Figure 1 is
conducted manually in which a group of system design
engineer organized many discussions and meetings to
understand the user requirements and specifications.
However there is no effort to tackle the capturing of the
requirements and transforming it into the system specifi-
cations using the mention method. This is essential be-
cause the system specification is scenario dependent in
which the requirements will provide more precise infor-
mation towards reconfiguration. Hence, without properly
capture the requirements; the system may have not been
designed correctly. It is obvious from the reviews that
various methods on configuration and reconfiguration
were created but there were no specific research con-
ducted on capturing the requirements. Capturing the re-
quirements or user requirements is essential steps to sim-
plify the configuration and reconfiguration works as
shown in Figure 1.
To act as fast as possible, a specific automated method
to capture and manipulate the user requirements and later
provide an optimum solution for the design of flexible
and reconfigurable manufacturing automation system is
essential to complement with the current effort. It is
noted that the outcome of this work will undoubtedly
provide highly flexible and easily platform to adapt with
various manufacturing conditions with also less human
involvement. This platform will not only cater for initial
system design and development but also for system re-
configuration as well.

Configuration Model for Automating Work System Design 118
3. Focus Development
In this work, the focus is on developing item in phase 1
and 2 of Figure 1. In order to come up with a suitable
method, a specific model for extraction and configuration
needs to be introduced. The actual theory of the model is
briefly explained in this section. The initial work need to
be initiated. The idea is to introduce the extraction proc-
ess. The detail activity for this work is shown in the fol-
lowing Figure 2.
In Figure 2, the configuration process starts after the
requirements are received from the customer, i.e. from
the market. The process then continues analyzing the
requirements and extracts the key elements from the in-
put information. The outcomes of the requirements are
passed to the configuration module which produces the
proposed configuration for implementation.
3.1. Requirements
The critical task in the automated configuration system is
recognition of user requirements. Most of the require-
ment information comes directly from the customer but
there are other channels such as media and market intel-
ligence sources that can be consolidated into some de-
scription for the system designer. In reality, these re-
quirements are expressed in documents which are vague
and often misleading. Therefore, many engineering de-
sign teams use the concept of quality function deploy-
ment (QFD) [16]. Jiang et al. [17] presented an overall
review of QFD in the past 30 years. They adopted the
“action research” approach interacting with product de-
velopment groups and organisation, and proposed a
model with 17 subsystems that linked system and prod-
uct design to quality. Ocak [18] studied 2 competing
companies using QFD method to provide a comprehen-
sive, systematic approach to ensure customer require-
ments and expectations are met via applying improve-
ments to design, production and management phases in
manufacturing system design. The results of the QFD
study could assist the companies to focus on specific
issues.
Traditionally, QFD is used to capture the voice of the
customer (i.e. requirements) and translates it into techni-
cal design requirements [19]. According to Mehrjerdi
[20], the source of information for determining require-
ments come marketing surveys and case studies. More
importantly, the requirements of the so-called unspoken
customers could be captured by the “house of quality”
(HoQ) [21]. The concept of HoQ originated from Toyota
Motor Corporation [22]. The tool is composed of a set of
matrices that represents the relationships between cus-
tomer requirements (CRs) and technical characteristics
(TCs). Once these relationships are quantified, a combi-
nation of different analysis and decision methods can be
used to determine the outcome.
However, the QFD concept and HoQ methodologies
have some drawbacks [23]. An inconsistent HoQ chart is
one in which the information from the roof matrix is in-
consistent with that from the relationship matrix. It is
necessary to establish processes through which the con-
sistency of information collected in HoQ and QFD is
checked [24]. The system reconfiguration projects that
this paper investigates certainly need information from the
user to determine the best option for the manufacturing
tasks. Capturing user requirements in the least restriction
manner is essential to facilitate accuracy through the de-
scriptions [25]. Hence, this research adopts a linquistic
approach to capture user requirements from sentences.
Examples of user requirements are illustrated in the
following sentences:
a) Sort 2 materials by weight;
b) Assemble 2 parts by inserting B on top of A;
c) Paint the 2 surface with 2 different colors;
d) Classify object according to 4 different colors;
e) 4 items need to be categorized according to height.
These requirements contain useful information for
Figure 2. Extraction and configuration activity.

Configuration Model for Automating Work System Design 119
system designer to configure a new system. According to
Rahman and Mo [13], a basic user requirement can be
divided into three main elements. There are type of
process (P), condition (C) and number of iteration (I).
According to Ratchev et al. [26], these elements will give
basic information required to configure a system and are
required to identify the general idea of the system to be
designed. A simple method to differentiate between all
the elements is described in the next section.
3.2. Extraction Module
To configure a system, user requirements need to be
clearly identified and simplified. At the beginning of the
process, an understandable set of user requirements is
required, either from verbal description or by some
documentation or statement, to formulate a conceptual
model of what the system is supposed to do. The key in
this process is the identification of user requirements in
sentence and keywords, which have unique meanings.
Later, the user requirement will be transformed into sys-
tem specifications. These forms are well presented by
Manesh [27].
From the examples in previous subsection, it is logical
to divide the user requirements into the three elements P,
I and C. The following describe examples of characteris-
tics of the elements.
a) Possible Process Type (P):
Sort—1, 2, 3, ···, n product
Assemble—1, 2, 3, ···, n part
Hence, the main characteristic of P is based on key-
words which are contained as verbs.
b) Possible Number of Item Count (I):
“n” number of product
“n” number of part
Hence, the main characteristics of I is integer numbers,
starting from 1, 2, 3, ···, n.
c) Possible Condition (C):
By weight, material, height
From side, top, bottom
Hence, the main characteristic of C is based on key-
words, but they are adjective.
Now, the extraction process can be defined with the
following rules:
For P, extract by a database of key verbs.
For I, look for numbers.
For C, look for adjective etc. in the database.
In order for the configuration module to derive the re-
sult, the elements P, I and C are required to be extracted
from the user requirements. For example, using user re-
quirement (a), the extraction logic can be shown in Fig-
ure 3.
Generalising, we can map any user requirement by the
following relation:




R P,I,C (1)
In our case,
R = Sort 2 materials by weight
P = Sort
I = 2
C = by weight
Similarly, analysing user requirement (b) gives the
following relation:
Sort 2 materials by weight
P IC
Figure 3. Extraction of user requirements.
Figure 4. Generalized transformation system model.

Configuration Model for Automating Work System Design 120
Figure 5. Sorting actions for two components.



Assemble2partsbyinsertingB ontopofA
Assemble, 2, by insertingB ontop ofA. (2)
3.3. Configuration Module
The generalized transformation of system model of this
research part is shown in the following Figure 4.
In the figure, the beginning part shows the user re-
quirements which can be categorized into three main
parts. The figure clearly shows the following user re-
quirements, P, I and C.
3.3.1. System Action
The first step to introduce a configuration model rose
after a series or a combination of system action (SA) is
created. In this case, the model can be initially shown in
the following sequence:
1234
SA SASASASAn 
The system action may consist of the following action
SET, RESET, RECOGNIZE, MOVE, EJECT, HOLD.
Selection and/or combination of the system action is
based on the process type acquired from the user re-
quirements commencing from 1 until nth number. This nth
number indicates the total number of system action re-
quired to complete the process. It will depend on the
number of item count (I) from the user requirement. The
following Figure 5 shows a combination of system ac-
tion for sorting process from a study conducted by Rah-
man and Mo [13]:
The combination shows sorting of 2 component of
different weight. In this case n = 2 and the condition, c is
weight. Further to the increment of the number of com-
ponent to be sorted for n = 3 or more, another similar
combination of system action was added as illustrated in
Figure 6 for nth.
From the combination, it shows a unique pattern that
can be used to generalize the system action. Therefore,
for “n” number of component to be sorted,


n
SA
N
5n14 
For sorting case the value of n ≥ 2, otherwise the
process will not doing any sorting. In term of the se-
quence of the system action, the sorting process chooses
alternative combination. The process will choose the
system action combination accordingly upon receiving
the information at the earlier recognition process at SA2.
The process will be decided to proceed with SA4 or SA6
immediately after SA3.
Figure 6. Sorting actions for five components.

Configuration Model for Automating Work System Design 121
Figure 7. Assembly actions for tw o subasse mblie s.
Figure 8. Assembly actions for five subasse mblie s.
Sep
SortBase2
SA SortBase1
Alternative
2SA
Sort 3SAor
4SA






In order to study the other type of processes, the fol-
lowing combination of system action for assembly can
also be shown in Figure 7 as follows:
The combination shows assembly of two subassem-
blies together for n = 2 case. For assembly of three sub-
assemblies of n = 3 or more, combination of four system
action, will occurs. Finally, Figure 8 shows combination
for nth.
Again, the combination shows a unique pattern than
can be generalized for indicating the total number of
system action required for this assembly process:


n
SA
N
5n14 
The value of n for assembly process is ≥2, otherwise
the process will not do assembling task. In term of the
sequence of the system action, the assembly process
chooses direct combination. The process will add the
system action combination accordingly upon receiving
the information at the earlier recognition process at SA2.
The process will add numbers of subassemblies accord-
ingly.


Seq AssyBase1AssyBase2
AssySA= 3SAn14SA2SA 
Each system action will indicate a need for certain
type of component. In the next section system component
(SC) is introduced to the current combination which will
further derive the model.
3.3.2. System Com p onent
Given a single combination of system action, each sys-
tem action will react and correspond to a specific system
component from the system component repository. In
this case, the total number of system component is simi-
lar to the total number of system action.
123 4
SA SASASAS
A

1234
n
SC SCSC SCSC
At this stage, list of component are required in order to
suit with the desired system component as well as corre-
sponding system action. Since the expected outcome of
the manufacturing system may differ from one to another,
extensive lists are required. A set of components will be
created which will store the database and will be known
as system component repository. This repository contains
numbers of components needed for setting up various
types of system. The repository will provide heaps of
data regarding various components required in the proc-
essing level of the proposed configuration work. The
following Table 1 shows an example of the developed

Configuration Model for Automating Work System Design 122
Table 1. Repository system for system component with corresponding system action.
SA Level SC Level (Generic) SC Level (Specific)
EJECT CYLINDER CL1, CL2, CL3,···, CLn
Others
HOLD CYLINDER CL1, CL2, CL3,···, CLn
Others
RECOGNIZE SENSOR SN1, SN2, SN3,···, SNn
Others
MOVE CONVEYOR CV1, CV2, CV3,···, CVn
TURN TABLE TT1, TT2, TT3,···, TTn
LIFTING MECHANISM
LM1, LM2, LM3,···, LMn
Others
SLIDE SLIDER SL1, SL2, SL3,···, SLn
Others
MEMORIZE PLC PLC1, PLC2, PLC3,···, PLCn
Others
repository system for system component with corre-
sponding system action.
Selection of the system component is base on individ-
ual system action acquired in the prior stage. An easy
relationship between System Model (SM), System Ac-
tion (SA) and System Component (SC) can be concluded.
In every single system model, there can be more than one
similar system action (a repetition of system action). The
following symbolic relationship can be used to illustrate
the process.
While in every repetition of system action in each final
system model, the corresponding system component can
be of a similar or different component.
In a nutshell, the model which consist of combination
of system action (SA) with corresponding system com-
ponents (SC) can be rewritten in a function form of


SM SA,SC
3.3.3. S y stem Actions and Component s S election
Selection/combination of the system action is based on
the process type acquired from the user/system require-
ments. On top of that, the selection of the component is
also based on the condition (c) extracted from the user
requirement. The following Figure 9 shows an example
of the selection of system components with correspond-
ing system actions.
For sorting process, Table 2 is an example of condi-
tions, c (see Table 2).
In our case, the condition chosen is weight and the
following assigned identification of system component in
corresponds to the system action is shown in Table 3
(see Table 3).
For Assembly process, Table 4 shows few example of
assembly condition, c (see Table 4).
In our case, the condition is inserting part B onto part
A and the following assigned identification of system
component in corresponds to the system action is shown
in Table 5 (see Table 5).
Selection and/or combination of the system models are
based on the number of iteration acquired from the user
requirements commencing from 1 until nth number. This
nth number will therefore depend on the iteration of the
process in the system.
4. Implementing System Hardware
4.1. Space Utilization
The next steps towards the implementation stages are to
finalize the system actions and system components se-
lection from the database. Once the components are
selected, the approximate size of the system can be ob-
tained for initial prediction of the space required for
lying down the system. Table 6 shows the relationship
to select the suitable components for each the actions
for sorting.
On top of the listed components, accessories to run the

Configuration Model for Automating Work System Design 123
Figure 9. Selection of system components with corresponding system actions for sorting process.
Table 2. Example of sorting condition.
CONDITION IDENTIFICATION
Weight 1
Colour 2
Height 3
Shape 4
Table 1. Identification assignment for weight sorting.
CONDITION IDENTIFICATION SYSTEM ACTIONIDENTIFICATIONSYSTEM COMPONENT
SET/RESET 0.1 PLC
RECOGNIZE1 1.1 LOAD CELL SENSOR
RECOGNIZE2 1.2 THROUGH BEAM
MOVE 2.1 CONVEYOR
HOLD 3.1 CYLINDER1
Weight 1
EJECT 3.2 CYLINDER2
Table 2. Example of assembly condition.
CONDITION IDENTIFICATION
Insert from top 1
Insert from side 2
Insert from bottom 3
Table 3. Identification assignment for assembly by inserting subassembly from top.
CONDITION IDENTIFICATION SYSTEM ACTIONIDENTIFICATIONSYSTEM COMPONENT
SET/RESET 0.1 PLC
RECOGNIZE1 1.1 PROXIMITY SENSOR
RECOGNIZE2 1.2 LIMIT SWITCH
RECOGNIZE3 1.3 THROUGH BEAM
MOVE 2.1 CONVEYOR
HOLD 3.1 CYLINDER
Insert from top 1
INSERT 4.1 FEEDER

Configuration Model for Automating Work System Design 124
system may be required but not included in this discus-
sion. From the individual space information, the total
required space for the complete system can be calculated
as follow:
12345
System space utilizationSU
AA2AAAAA
i


6
The approximation of system space utilization layout
for the configured sorting system in this work is shown
in Figure 10.
The next Table 7 shows the component selection for
assembly process.
Similar to sorting process, other accessories to run the
system may be required but not included in this discus-
sion. From the individual space information, the total
required space for the complete system can be calculated
as follow:
123456
System spaceutilization= SU
AAAAAAA
i
 

The approximation of system space utilization layout
for the configured assembly system in this work is shown
in Figure 11.
Table 4. Components selection for the configured sorting process.
ACTIONS COMPONENTS QUANTITY SPACE REQUIRED
SET/RESET PLC 1 A3
RECOGNIZE1 LOAD CELL 1 A1
RECOGNIZE2 THROUGH BEAM 2 A2
MOVE CONVEYOR 1 A4
HOLD CYLINDER1 1 A5
INSERT CYLINDER2 1 A6
Figure 10. Space approximation for the configured sorting system.
Table 7. Components selection for the configured assembly process.
ACTIONS COMPONENTS QUANTITY SPACE REQUIRED
SET/RESET PLC 1 A3
RECOGNIZE1 PROXIMITY SENSOR 1 A1
RECOGNIZE2 LIMIT SWITCH 1 A2
RECOGNIZE3 THROUGH BEAM 1 A4
MOVE CONVEYOR 1 A5
HOLD CYLINDER 1 A6
INSERT FEEDER 1 A7
Figure 11. Space approximation for the configured assembly system.

Configuration Model for Automating Work System Design 125
The layout is not a final layout but is more on the siz-
ing of the proposed new configured system. At this stage
it does not indicate specific orientation of the system.
However the information is useful during layout orienta-
tion stage. The information gathered from this section
will give valuable information in term of system space
utilization.
4.2. Proposal Mode: Implementation
Once the system action and system components are
available, the work system layout needs to be developed.
In term of facilitating the layout of the system, various
methods can be considered for automating the process.
These includes genetic algorithm by Kar Yan [28], Peters
[29] and other method suggested by Robin S. [30]. How-
ever, at this stage, the process is done manually by taking
into consideration all information from the system model.
The process will closely follow the information from the
approximation of space required for necessary compo-
nent. The initiation of the hardware implementation has
been taken place using the modular automation system.
This implementation was resulted from propose system
model and space utilization study conducted for simple
sorting process which can be seen in the following Fig-
ure 12.
The process involves laying out the components
manually base on the system action and system compo-
nent flow. Starting with the initial layout in Figure 11,
the layout has gone through several iteration and adjust-
ments at the final implementation stage. The next Figure
13 shows the orientation for the manual process.
The outcome of the study was implemented using the
proposed model. An example of automation work system
development for sorting process for two boxes of differ-
ent weight is shown in Figure 14.
The implemented system operates using conveyor as
the transfer system. Once the product is placed on the
weighing station, the conveyor will transfer the product
from the current spot until it reaches the decision area. At
the decision area, the pneumatic cylinder will either push
the product onto the first slider or let the product through
to the second slider. This decision making process is
done by the Programmable Logic Controller (PLC).
5. Conclusion
At this stage, the initial model to extract and manipulate
the user requirements has been developed. The outcome
of this work is the division of user requirements into
process type, item count for the process and condition of
process. The outcome will later provide with the general
Figure 12. Implementation of component based on the system actions and system components.
Figure 13. Orientation of the layout.

Configuration Model for Automating Work System Design 126
Figure 14. Hardware implementation for automation work system.
idea for laying out the system. This work proved that the
common instructions, in this case the user requirements,
can be generalized in configuring automation work sys-
tem structure. In the future, this research work will bene-
fit the industry through reducing human involvement
while trying to optimize the current system and at the
same time minimizing the risk of future investment in
simple sorting and assembly. More work is currently un-
derway to improvise the model to be used for both config-
uring and reconfiguring various complex type of system.
6. Acknowledgements
The present works was raised from the collaborative part-
nership between RMIT University, Australia and SAGE
Didactic, Australia an established automation education
facility supporting multi-level learning requirements. The
researcher was financially supported by the Universiti
Teknikal Malaysia Melaka (UTeM), Malaysia and Min-
istry of Higher Education (MOHE) Malaysia.
REFERENCES
[1] J. P. T. Mo, P. Dawson and M. A. A. Rahman, “Active
Learning Approach in Developing Engineering Design
Skill through Open Ended System Specification,” 20th
Australasian Association for Engineering Education Con-
ference, Adelaide, 6-9 December 2009, pp. 61-66.
[2] H. ElMaraghy, et al., “Managing Variations in Products,
Processes and Manufacturing Systems,” CIRP Annals—
Manufacturing Technology, Vol. 58, No. 1, 2009, pp.
441-446.
[3] U. Karmarkar and S. Kekre, “Manufacturing Configura-
tion, Capacity and Mix Decisions Considering Opera-
tional Costs,” Journal of Manufacturing Systems, Vol. 6,
No. 4, 1987, pp. 315-324.
doi:10.1016/0278-6125(87)90007-0
[4] A. Matta, et al., “An Integrated Approach for the Con-
figuration of Automated Manufacturing Systems,” Ro-
botics and Computer-Integrated Manufacturing, Vol. 17,
No. 1-2, 2001, pp. 19-26.
doi:10.1016/S0736-5845(00)00033-8
[5] R. P. Monfared and R. H. Weston, “A Method to Develop
Semi-Generic Information Models of Change-Capable
Cell Control Systems,” Computers in Industry, Vol. 41,
No. 3, 2000, pp. 279-294.
doi:10.1016/S0166-3615(99)00050-0
[6] C. Stoian and G. Frumusanu, “Reconfigurable Manufac-
turing Systems Design Principles,” The Annals Dunarea
de Jos of Galati, Fascicle V, Technolologies in Mecanical
Engineering, Vol. 2007, No. 1, pp. 62-65.
[7] H. P. Wiendahl, et al., “Changeable Manufacturing—
Classification, Design and Operation,” CIRP Annals—Man-
ufacturing Technology, Vol. 56, No. 2, 2007, pp. 783-
809.
[8] E. Travaini, et al., “Methodological Approach and Recon-
figuration Tool for Assembly Systems,” CIRP Annals—
Manufacturing Technology, Vol. 51, No. 1, 2002, pp. 9-
13.
[9] H. A. ElMaraghy, O. Kuzgunkaya and R. J. Urbanic,
“Manufacturing Systems Configuration Complexity,” CIRP
Annals—Manufacturing Technology, Vol. 54. No. 1, 2005,
pp. 445-450.
[10] W. Covanich and D. McFarlane, “Assessing Ease of Re-
configuration of Conventional and Holonic Manufactur-

Configuration Model for Automating Work System Design 127
ing Systems: Approach and Case Study,” Engineering Ap-
plications of Artificial Intelligence, Vol. 22, No. 7, 2009,
pp. 1015-1024. doi:10.1016/j.engappai.2009.01.001
[11] A. Ferscha, et al., “Building Flexible Manufacturing Sys-
tems Based on Peer-Its,” EURASIP Journal on Embedded
Systems, Vol. 2008, 2008. doi:10.1155/2008/267560
[12] A. D. Toni and S. Tonchia, “Manufacturing Fexibility: A
Literature Review,” International Journal of Production
Research, Vol. 36, No. 6, 1998, pp. 1587-1617.
doi:10.1080/002075498193183
[13] M. A. A. Rahman and J. P. T. Mo, “Development of
Theoretical Reconfiguration Structure for Manufacturing
Automation Systems,” International Journal of Agile Sys-
tems and Management, Vol. 5, No. 2, 2012, pp. 132-150.
[14] W. Lepuschitz, et al., “Toward Self-Reconfiguration of Manu-
facturing Systems Using Automation Agents,” IEEE Trans-
actions on Systems, Man, and Cybernetics, Part C: Ap-
plications and Reviews, Vol. 41, No. 1, 2011, pp. 52-69.
[15] Y. Huang, et al., “Manufacturing Execution System Ar-
chitecture Supporting Cross-Grain Reconfiguration,” Jisu-
anji Jicheng Zhizao Xitong/Computer In tegrated Manufac-
turing Systems, Vol. 17, No. 4, 2011, pp. 747-759.
[16] N. Fuller, “The House of Quality,” Supply Management,
Vol. 3, No. 3, 1998, p. 44.
[17] J.-C. Jiang, M.-L. Shiu and M.-H. Tu, “Quality Function
Deployment (QFD) Technology Designed for Contract
Manufacturing,” TQM Journal, Vol. 19, No. 4, 2007, p.
291. doi:10.1108/09544780710756205
[18] Z. Ocak, “Applying Quality Function Deployment in the
Manufacturing Industry: A Review & Case Study in Pro-
duction,” The Business Review, Vol. 9, No. 2, 2008, p. 4.
[19] R. K. Singh, C. C. Elrod and E. A. Cudney, “Comparative
Analysis of Quality Function Deployment Methodologies:
A Case Study Analysis,” The Quality Management Jour-
nal, Vol. 19, No. 1, 2012, p. 7.
[20] Y. Z. Mehrjerdi, “Applications and Extensions of Quality
Function Deployment,” Assembly Automation, Vol. 30, No.
4, 2010, p. 388. doi:10.1108/01445151011075843
[21] R. Mahanti, “The Application of Quality Function De-
ployment to User Interface Design,” The Quality Man-
agement Journal, Vol. 16, No. 1, 2009, p. 29.
[22] C.-H. Liu, “A Group Decision-Making Method with
Fuzzy Set Theory and Genetic Algorithms in Quality
Function Deployment,” Quality and Quantity, Vol. 44,
No. 6, 2010, p. 1175. doi:10.1007/s11135-009-9304-1
[23] A. Olewnik and L. Kemper, “Limitations of the House of
Quality to Provide Quantitative Design Information,” The
International Journal of Quality & Reliability Manage-
ment, Vol. 25, No. 2, 2008, p. 125.
doi:10.1108/02656710810846916
[24] J.-S. Shin, K.-J. Kim and M. J. Chandra, “Consistency
Check of a House of Quality Chart,” The International
Journal of Quality & Reliability Management, Vol. 19,
No. 4, 2002, p. 471. doi:10.1108/02656710210421535
[25] S. Ratchev, H. Hirani and M. Bonney, “Knowledge Based
Formation of Re-Configurable Assembly Cells,” Journal
of Intelligent Manufacturing, Vol. 18, No. 3, 2007, pp. 401-
409. doi:10.1007/s10845-007-0031-y
[26] H. F. Manesh, D. Schaefer and M. Hashemipour, “Infor-
mation Requirements Analysis for Holonic Manufactur-
ing Systems in a Virtual Environment,” The International
Journal of Advanced Manufacturing Technology, Vol. 53,
No. 1-4, 2011, pp. 385-398.
doi:10.1007/s00170-010-2822-0
[27] T. K. Yan, “Genetic Algorithms, Function Optimization,
and Facility Layout Design,” European Journal of Op-
erational Research, Vol. 63, No. 2, 1992, pp. 322-346.
doi:10.1016/0377-2217(92)90034-7
[28] B. A. Peters and M. Rajasekharan, “A Genetic Algorithm
for Determining Facility Design and Configuration of
Single-Stage Flexible Electronic Assembly Systems,” Jour-
nal of Manufacturing Systems, Vol. 15, No. 5, 1996, pp.
316-324. doi:10.1016/0278-6125(96)84194-X
[29] S. L. Robin, “Automated Facilities Layout: Past, Present
and Future,” Automation in Construction, Vol. 9, No. 2,
2000, pp. 197-215. doi:10.1016/S0926-5805(99)00005-9
[30] S. L. Robin, “Automated Facilities Layout: Past, Present
and Future,” Automation in Construction. Vol. 9, No. 2,
2000, pp. 197-215.