A REVIEW ON CELLULAR MANUFACTURING SYSTEM

1
VIKRANT SHARMA, 2JINESH JAIN, 3ANKUR PAREEK
1,2,3
Department of Mechanical Engineering, GEC Ajmer
Email: [email protected]

Abstract - A big challenge of today’s batch production industries is to offer fast delivery of a variety of products to the
customer. To meet this challenge, manufacturing industries are turning to cellular manufacturing system (CMS). CMS is
based on group technology approach where workstations are arranged in form of manufacturing cells and part families are
assigned to these cells. Each manufacturing cell performs all or most of the operations required by part families. Primary
purpose of CMS is to reduce setup time, cycle time, and improve machining utilization. The important steps in CMS are
identifying part families, formation of manufacturing cells and assigning part families to the cells. This paper reviews
literature on cellular manufacturing system. Different approaches such as heuristic, meta-heuristics, mathematical
programming methods etc. have been discussed.

Keywords - Cellular manufacturing system, group technology, manufacturing cells, part families.

I. INTRODUCTION grouped into part families. Part Family is a collection

of parts that posses similarities in shape and size. Part
Cellular Manufacturing (CM) is a approach for families are a very important features of GT.
improving operations in batch and job shop
production. In cellular manufacturing, Group Mitrofanov was the first who introduced the concept
Technology (GT) is used to form part families based of GT philosophy and the machine grouping problem
on similar processing required on the job. Machines in the late 1950’s. The other early pioneers in the
and parts are then grouped together based on in order field of GT are Burbidge and Ham (Heragu, 1994).
or immediate techniques (Arora, Haleem, & Singh, Early techniques developed for solving GT and CM
2011). problems are as follow:

Group Technology is defined as “A manufacturing Visual inspection is the technique where parts are
philosophy in which similar parts are identified and classified into appropriate families, based on the parts
grouped together to take advantage of their or photos of the parts. But this technique is not
similarities in design and production”. Group suitable on large scale.

Technology (GT) is a method of manufacturing parts Part classification and coding is the technique where
by the classification of these parts into groups and parts are classified into appropriate families, based on
subsequently applying similar technological number of attributes, such as dimensions of part;
operations to each group (Mitrofanov, 1966). shape of part; material of part; tolerance requirement.
Dividing the manufacturing facility into small groups Usually, each part is assigned a code and each code
or cells of machines are known as “cellular representing an attribute of the part.
manufacturing”.
Production flow analysis (PFA) is the technique
There are significant advantages that can be achieved where parts are classified into appropriate families
as a result of implementing a cellular manufacturing using information contained on route sheets. This
system (CMS). technique also identifies machine cells using input as
a complete list of machines and machine capabilities.
 Reduced handling of materials
 Reduced tooling and equipments II. CELL FORMATION TECHNIQUES
 Reduced set-up time
 Reduced work -in-process inventory In last few decades numbers of researchers have
 Reduced part make span developed methods for formation of cells in CM.
 Improved operator capability
1. Mathematical Programming:
The main aim in the design of a CMS is to identify Mathematical programming approaches are
part families, formation of machine cells, and commonly in use in cellular manufacturing systems.
allocation of part families to machine cells such that Mathematical programming methods such as linear
the movement of parts between cells is minimized. programming, goal programming, and dynamic
GT and CM is suitable for the plant which currently programming are proposed by researcher.
uses batch production and a process type layout. The Mathematical programming methods can easily
required condition to apply GT is that the parts can be include a number of design logics in their objectives

Proceedings of International Conference on Recent Innovations in Engineering and Technology, Jaipur, India, 18th - 19th Feb’2017, ISBN: 978-93-86291-63-9
162
 A Review on Cellular Manufacturing System
and constraint functions. Purcheck was among the that these techniques have more real world
first researchers to apply linear programming to applications. Different clustering algorithms that have
group technology in (1975). Kusiak in (1987) been proposed are Genetic Algorithm (GA),
proposed the p-median model to identify part simulated annealing, tabu search, particle swarm
families. Srinivasan in (1990) proposed an optimization, ant colony optimization.
assignment model for the part families and machine Genetic Algorithm (GA) is flexible and effective
grouping problem. Sankaran in (1990) presented a optimization tool that can provide near optimal
mathematical model of goal programming. Adil et al. solutions. Joines et al. developed a GA for solving
proposed nonlinear mathematical model for integer programming formulations of cell. Mak et al.
formation of machine cell and part families (2000) proposed an adaptive genetic approach as an
simultaneously. Kulkarni P C et. al. [2007] presented effective means of providing the optimal solution to
a mathematical formulation for layout in cellular the CF problem. Wu X et. al. [2007] developed a
manufacturing system. hierarchical genetic algorithm to simultaneously form
manufacturing cells and determine the group layout
2. Cluster Analysis Techniques: of a CMS. Arkat J et. al. [2012] proposed a multi-
Cluster analysis is composed of many diverse objective genetic algorithm (MOGA).
techniques for recognizing structure in a complex Ant Colony Optimization (ACO) is one of the newest
data set. Incidence matrix of machines is used in meta-heuristics used for cell formation. ACO was
array based clustering. This matrix represents first introduced by Dorigo et al. (1996). ACO was
processing requirements of components on machines. inspired by the foraging behavior of real ants and in
Different clustering algorithms that have been particular, how ants can find the shortest route path
proposed are Bond Energy Analysis (BEA) by from their colony to a food source (Deneubourg et al.,
McCormick (1972), Rank Order Clustering (ROC) by 1999). Li D et. al. [2014] developed a combinational
King (1980), ROC2 by King and Nakornchai (1982), ant colony optimization (CACO) approach, which
Modified Rank Order Clustering (MODROC) by designs two structures for the single-processing
Chandrasekharan and Rajagopalan (1986), Direct machines and the batch-processing machine. Islier
Clustering Algorithm (DCA) by Chan and Milner (2005) proposed artificial ant system to optimize cell
(1982) . formation problems
In Hierarchical clustering techniques machine cells Simulated Annealing is the oldest among meta-
are formed employing a similarity or distance heuristics. Souilah (1995) presents the general
function between machines or parts and then Simulated Annealing algorithm. Vakharia and Chang
machines or parts are separated into a few broad cells, (1997) developed two heuristic methods for the cell
each of which is further divided into smaller groups. formation problem both based on simulated annealing
Common Hierarchical clustering techniques that have and tabu search algorithms.
been proposed is Single Linkage Clustering (SLC) Tabu search (TS) is held to be one of the most fruitful
algorithm by McAuley (1972). meta-heuristic techniques. Logendran et al. (1994)
In Non Hierarchical clustering techniques iterations used TS in CMS design. Adenso-Diaz et al. (2001)
are used and they begin with an initial partition of the developed a TS based methodology to solve cell
data set or the choice of a few seed points. Non - formation with an emphasis on different machine
hierarchical techniques have been developed by grouping problems. Tavakkoli-Moghaddam et al.
Lemoine and Mutel, (1983), Chandrasekharan and (2005) proposed TS, SA and GA methods for
Rajagopalan, (1986), and Srinivasan and Narendran, dynamic condition of cell formation problems.
(1991). Particle Swarm Optimization (PSO) was first
proposed by Kennedy and Eberhart (1995). Andres
3. Heuristics: and Lozano (2006) developed first PSO technique for
Heuristics techniques provide an alternative basis for cell formation problem. Ming and Ponnambalam
solving cell formation problem. Although a heuristics (2008) proposed a hybrid PSO approach combined
technique does not assure for optimum solution but with GA for cell formation problem.
they are very useful in providing feasible solution.
Mukattash et al. (2002) proposed three heuristic 5. Neural Network:
procedures. Chan et al. (2003) developed a heuristic Neural networks techniques commonly applied in cell
algorithm considering the intra-cell materials flow. formation due to their robust and adaptive nature.
Kim et al. (2004) proposed a method of cell Kaparthi and Suresh (1992) have applied artificial
formation with multi-objective machine formation. neural networks for classification and coding of
rotational parts. Rao and GU proposed artificial
4. Meta-heuristics: neural network to cell formation problem. Liang and
Meta-heuristics technique are used solve a varied Zolfaghari (1999) presented a neural network
range of optimization problems especially approach to the machine cell/part-family formation
combinatorial problems. These techniques provide an problem considering processing time, lot size,
approximate solution. Jones et al. (2002) conclude machine capacity, and machine duplication.

Proceedings of International Conference on Recent Innovations in Engineering and Technology, Jaipur, India, 18th - 19th Feb’2017, ISBN: 978-93-86291-63-9
163
 A Review on Cellular Manufacturing System
Soleymanpour et al. (2002) proposed a transiently thoughtfulness towards production planning and
chaotic neural network algorithm. control activities of cellular manufacturing system.
Also the technique does not discuss the optimal size
6. Fuzzy Logic: of the cell.
Xu and Wang (1989) applied fuzzy mathematics to
the cell formation problem where part features are Cell formulation solution methods are shown in Fig 1
transformed by membership functions into fuzzy
numbers. Chu and Hayya (1991) applied a fuzzy c-
means clustering algorithm to production data. Torkul
et al. (2006) employed fuzzy logic to study the design
of part families and machine cells simultaneously.
Ravichandran and Rao (2001) proposed a new fuzzy
clustering algorithm. Won and Currie (2007)
proposed a more comprehensive Fuzzy ART
approach for part-machine grouping. Safaei et al.
(2008) developed a fuzzy programming-based
approach to solve an extended mixed-integer
programming model for a dynamic cell formation
problem.

3. Literature Summary:
As understood from the previous section, there are a
large number of methods which have been developed
for formulation of manufacturing cell. The stress
Figure 1
seems to have been on developing "new" techniques.
Cell formation problems can be formulated using
single objective or multiple objectives, such as Journal wise papers selected for literature survey are
intercell or intracell part movement, within cell load shown in Fig 2.
variation, machine utilization etc. Majority of the
published works is on cell formulation but very little

Figure 2

Proceedings of International Conference on Recent Innovations in Engineering and Technology, Jaipur, India, 18th - 19th Feb’2017, ISBN: 978-93-86291-63-9
164
 A Review on Cellular Manufacturing System
CONCLUSION [19] Forghani, K & Mohammadi, M.,(2016). Designing cellular
manufacturing systems considering S-shaped layout.
Computers and Industrial Engineering, 98(May), 221–236.
This paper presented brief study and salient features [20] Gholipour-Kanani Y, Tavakkoli-Moghaddam R, Khorrami A.
of various techniques and methodology of cell Solving a multi-criteria group scheduling problem for a
formation. A Good number of research papers was cellular manufacturing system by scatter search. J Chinese
Inst Ind Eng. 2011;28(3):192-205.
identified on different methodologies employed such [21] Hazarika, M., & Laha, D. (2016). A Heuristic Approach for
as mathematical programming, cluster analysis, Machine Cell Formation Problems with Alternative Routings.
heuristics, meta-heuristics, neural networks and fuzzy Procedia Computer Science, 89, 228–242.
logic. Genetic Algorithm has been the most adaptable [22] HACHICHA W, MASMOUDI F, HADDAR M. A
algorithm for the researcher as a solution method of correlation analysis approach of cell formation in cellular
manufacturing system with incorporated production data. Int
cell formation. J Manuf Res. 2006;1(3):332-353..
[23] Hachicha W, Masmoudi F, Haddar M. Formation of machine
REFERENCES groups and part families in cellular manufacturing systems
using a correlation analysis approach. Int J Adv Manuf
Technol. 2008;36(11-12):1157-1169.
[1] Akturk MS. Joint cell loading and scheduling approach to
[24] Houshyar, A. N., Leman, Z. Bin, Ariffin, M. K. A. M.,
cellular manufacturing systems. Int J Prod Res.
Ismail, N., Moghadam, H. P., & Iranmanesh, H. (2016).
2011;49(21):6321-6341.
Proposed linear-mathematical model for configuring cell and
[2] Alhourani F. Clustering algorithm for solving group
designing unequal-area facility layout in dynamic cellular
technology problem with multiple process routings. Comput
manufacturing system. International Journal of Industrial and
Ind Eng. 2013;66(4):781-790.
Systems Engineering, 22(3), 332.
[3] Arkat J, Abdollahzadeh H, Ghahve H. A new branch and
[25] Halat K, Bashirzadeh R. Concurrent scheduling of
bound algorithm for cell formation problem. Appl Math
manufacturing cells considering sequence-dependent family
Model. 2012;36(10):5091-5100.
setup times and intercellular transportation times. Int J Adv
[4] Amruthnath, N., & Gupta, T. (2016). Modified Rank Order
Manuf Technol. 2015;77(9-12):1907-1915.
Clustering Algorithm Approach by Including Manufacturing
[26] Haraguchi H, Kaihara T, Fujii N, Kokuryo D. A study on
Data. IFAC-PapersOnLine, 49(5), 138–142..
operator allocation and scheduling method considering
[5] Arkat J, Ghahve H. Scheduling of virtual manufacturing cells
demand fluctuations in cell manufacturing systems. 2015
with outsourcing allowed. Int J Comput Integr Manuf.
54th Annu Conf Soc Instrum Control Eng Japan, SICE 2015.
2014;27(12):1079-1089.
2015:1007-1012.
[6] Aryanezhad MB, Aliabadi J, Tavakkoli-Moghaddam R. A
[27] Hassan Zadeh A. AA., Afshari H. HH., Ramazani Khorshid-
new approach for cell formation and scheduling with
Doust R. R. Integration of process planning and production
assembly operations and product structure. Int J Ind Eng
planning and control in cellular manufacturing. Prod Plan
Comput. 2011;2(3):533-546.
Control. 2013;25(10):840-857.
[7] Askin RG. Contributions to the design and analysis of
[28] Heragu SS. Group technology and cellular manufacturing.
cellular manufacturing systems. Int J Prod Res. 2013;51(23-
IEEE Trans Syst Man Cybern. 1994;24(2):203-215.
24):6778..
[8] Bychkov, I., & Batsyn, M. (2017). An efficient exact model
[29] Kamaruddin S, Khan Z a., Siddiquee AN, Wong Y-S. The
for the cell formation problem with a variable number of
impact of variety of orders and different number of workers
production cells.
on production scheduling performance. J Manuf Technol
[9] Chang CC, Wu TH, Wu CW. An efficient approach to
Manag. 2013;24(8):1123-1142.
determine cell formation, cell layout and intracellular
[30] Karthikeyan S, Saravanan M, Ganesh K. GT machine cell
machine sequence in cellular manufacturing systems. Comput
formation problem in scheduling for cellular manufacturing
Ind Eng. 2013;66(2):438-450..
system using meta-heuristic method. Procedia Eng.
[10] Chaudhry IA, Khan AA. A research survey: Review of
2012;38:2537-2547.
flexible job shop scheduling techniques. Int Trans Oper Res.
[31] Kesen SE, Das SK, Gungor Z. A mixed integer programming
2016;23(3):551-591.
formulation for scheduling of virtual manufacturing cells
[11] Costa A, Cappadonna FA, Fichera S. Joint optimization of a
(VMCs). Int J Adv Manuf Technol. 2010;47(5-8):665-678..
flow-shop group scheduling with sequence dependent set-up
[32] Kulkarni PC, Shanker K. A genetic algorithm for layout
times and skilled workforce assignment. Int J Prod Res.
problems in cellular manufacturing systems. IEEM 2007
2014;52(9):2696-2728.
2007 IEEE Int Conf Ind Eng Eng Manag. 2007:694-698.
[12] Costa A, Cappadonna FA, Fichera S. A hybrid genetic
[33] Li D, Meng X, Li M, Tian Y. An ACO-based intercell
algorithm for minimizing makespan in a flow-shop sequence-
scheduling approach for job shop cells with multiple single
dependent group scheduling problem. J Intell Manuf. 2015..
processing machines and one batch processing machine. J
[13] Das SR, Khumawala BM. Group Technology and Cellular
Intell Manuf. 2016;27(2):283-296.
Manufacturing: An Overview. J Inf Optim Sci.
[34] Li Y, Li X, Gupta JND. Solving the multi-objective flowline
1986;7(2):197-207.
manufacturing cell scheduling problem by hybrid harmony
[14] Dimopoulos C. Explicit consideration of multiple
search. Expert Syst Appl. 2015;42(3):1409-1417.
objectives in cellular manufacturing. Eng Optim.
[35] Liu C, Wang J, Leung JY-T, Li K. Solving cell formation and
2007;39(5):551-565.
task scheduling in cellular manufacturing system by discrete
[15] Ebrahimi M, Fatemi Ghomi SMT, Karimi B. Hybrid flow
bacteria foraging algorithm. Int J Prod Res. 2016;54(3):923-
shop scheduling with sequence dependent family setup time
944.
and uncertain due dates. Appl Math Model. 2014;38(9-
[36] Mak KL, Peng P, Wang XX, Lau TL. An ant colony
10):2490-2504.
optimization algorithm for scheduling virtual cellular
[16] Elmi A, Solimanpur M, Topaloglu S, Elmi A.
manufacturing systems. Int J Comput Integr Manuf.
[17] A simulated annealing algorithm for the job shop cell
2007;20(March 2012):524-537.
scheduling problem with intercellular moves and reentrant
[37] Megala N, Rajendran C. An improved ant-colony algorithm
parts. Comput Ind Eng. 2011;61(1):171-178.
for the grouping of machine-cells and part-families in
[18] França PM, Gupta JND, Mendes AS, Moscato P, Veltink KJ.
cellular manufacturing systems. Int J Oper Res.
Evolutionary algorithms for scheduling a flowshop
2013;17(3):345-373.
manufacturing cell with sequence dependent family setups.
[38] Neufeld JS, Gupta JND, Buscher U. Minimising makespan in
Comput Ind Eng. 2005;48(3):491-506.
flowshop group scheduling with sequence-dependent family

Proceedings of International Conference on Recent Innovations in Engineering and Technology, Jaipur, India, 18th - 19th Feb’2017, ISBN: 978-93-86291-63-9
165
 A Review on Cellular Manufacturing System
set-up times using inserted idle times. Int J Prod Res. using an ant colony system approach. Int J Adv Manuf
2014;(ahead-of-print):1-16. Technol. 2005;25(9-10):1013-1019.
[39] Neufeld JS, Gupta JND, Buscher U. A comprehensive review [43] Rafiee Parsa N, Karimi B, Moattar Husseini SM. Minimizing
of flowshop group scheduling literature. Comput Oper Res. total flow time on a batch processing machine using a hybrid
2016;70:56-74. max???min ant system. Comput Ind Eng. 2016;99:372-381.
[40] Papaioannou G, Wilson JM. The evolution of cell formation [44] Reisman a., Kumar a., Motwani J, Cheng CH. Cellular
problem methodologies based on recent studies (1997-2008): Manufacturing: A Statistical Review of the Literature (1965-
Review and directions for future research. Eur J Oper Res. 1995). Oper Res. 1997;45(4):508-520.
2010;206(3):509-521. [45] Salmasi N, Logendran R, Skandari MR. Makespan
[41] Pillai VM, Subbarao K. A robust cellular manufacturing minimization of a flowshop sequence-dependent group
system design for dynamic part population using a genetic scheduling problem. Int J Adv Manuf Technol. 2011;56(5-
algorithm. Int J Prod Res. 2008;46(18):5191-5210. 8):699-710.
[42] Prabhaharan G, Asokan P, Girish BS, Muruganandam A. [46] Saraç T, Ozcelik F. A genetic algorithm with proper
Machine cell formation for cellular manufacturing systems parameters for manufacturing cell formation problems. J
Intell Manuf. 2012;23(4):1047-1061.



Proceedings of International Conference on Recent Innovations in Engineering and Technology, Jaipur, India, 18th - 19th Feb’2017, ISBN: 978-93-86291-63-9
166