Research Article  Open Access
Reza Raminfar, Norzima Zulkifli, Mohammadreza Vasili, Tang Sai Hong, "An Integrated Model for Production Planning and Cell Formation in Cellular Manufacturing Systems", Journal of Applied Mathematics, vol. 2013, Article ID 487694, 10 pages, 2013. https://doi.org/10.1155/2013/487694
An Integrated Model for Production Planning and Cell Formation in Cellular Manufacturing Systems
Abstract
Cellular manufacturing (CM) is a production approach directed towards reducing costs, as well as increasing system's flexibility in today's smalltomedium lot production environment. Many structural and operational issues should be considered for a successful CM design and implementation such as cell formation (CF), production planning, and facility layout. Most researchers have addressed these issues sequentially or independently, instead of jointly optimizing a combination of these issues. In order to attain better results to ensure that the system will be capable of remaining efficient in unknown future situations, these issues should be addressed simultaneously. In this paper, a mathematical model is developed using an integrated approach for production planning and cell formation problems in a CM. A set of numerical examples are provided from existing the literature in order to test and illustrate the proposed model. In order to evaluate and verify the performance of the proposed model, it is compared with a wellknown cell formation methods (rank order clustering and direct clustering analysis), using group capability index (GCI) measure. The results and comparisons indicate that the proposed model has a significantly higher and satisfactory performance and it is reliable for the design and the analysis of CM systems.
1. Introduction
Cellular manufacturing (CM) is a production system that involves processing a collection of similar parts (part families) on dedicated cluster of machines or manufacturing processes (cells) [1]. CM is an application of group technology (GT) which offers the advantages of both job shops (flexibility in producing a wide variety of products) and flow lines (efficient flow and high production rate) [2]. The advantages of CM include simplified material flows and reduced material handling cost, reduced production time, reduction in setup time, reduced production cost, reduction in inventory and workinprocess (WIP) inventories, reduction in scrap and waste, decentralization of responsibilities, and saving manufacturing space [3–6]. The design and implementation of an effective CM system involves many issues such as machinepart cell formation (CF), production planning, layout design, and scheduling.
In the design of a CM system, similar parts are grouped into families and associated machines into groups so that one or more part families can be processed within a single machine group [7]. The process of determining part families and machine groups in order to form the manufacturing cells is referred to as the CF problem. This problem may be solved by using different methods and considering different manufacturing features. The CF problems can be classified into binary and comprehensive problems depending on whether or not processing times and the machine capacities are considered.
The binary problem arises if the part demands are unknown when the CM system is being developed [8–10]. If the part demand can be accurately predicted, processing time and machine capacities have to be included in the analysis. This gives rise to comprehensive problems [11]. In addition, the comprehensive models can be classified considering the production requirements. They are static versus dynamic production requirements and deterministic versus stochastic production requirements [12]. A static production requirement implies a single period when designing a CM system. It assumes that product mix and part demand are constant for the entire planning horizon. The product mix refers to a set of parttypes to be produced at each period, and the part demand is the quantity of each parttype to be produced [13]. Product mix and demand in such cases can be deterministic or stochastic. For static and deterministic production requirements, there is only one possible set of product mix and demand which are known [14, 15]. In contrast, static and stochastic production requirements have a set of possible product mixes and demands to occur; each has its probability of occurrence [16]. However, dynamic production requirements imply multiple periods when designing a CM system. In the dynamic environment, the entire planning horizon is divided into multiple smaller periods, and each period has different product mix and part demand. Meanwhile, as mentioned earlier, production requirements can be deterministic or stochastic in each period. For dynamic and deterministic production requirements, product mix and demand in each period are known [17, 18]. For dynamic and stochastic production requirements, the possible product mixes and demands in each period are known with certain probabilities [19, 20].
In today’s world of rapidly changing product demand, small internet orders, tight delivery schedules, high competition, and high service level requirements, it will be increasingly difficult to maintain good operational performance over the time. In such a case, managing the production resources and balancing them between successive time periods with the aim of minimizing the production costs is known as “production planning” [21]. Based on the literature, numerous mathematical models and solution methods have been developed to solve the existing problems in general manufacturing or service industries [22, 23]. However, in the recent researches, critical manufacturing features such as production flexibility and manufacturing cell formations have received considerable attention in developing production planning models. Furthermore, in order to achieve more practical results, the production planning features such as machine capacity, machine cost, operations sequence, inventories holding, backorders, and subcontracting can be considered in order to form the manufacturing cells. In this regard, integrated approaches should be pursued in manufacturing system analysis, since different aspects of the system are interrelated in many ways [24]. Integrated system approaches can minimize the possibility of certain important aspects of the system being neglected, while other issues are being studied [25]. In general, integrating the concepts of the CM system design and production planning is a fundamental requirement for modeling and simulating of the real manufacturing systems. This integrated concept has been discussed, for example, in Chen [25], Safaei and TavakkoliMoghaddam [21], Ah kioon et al. [26], and Mahdavi et al. [27].
Although each of these researches has covered some of the features problem, or utilize special method to solve problems and their efficiencies are different too. For instance, Chen [25] solved problems with a decompositionbased heuristic algorithm or he considered unit cost to move for every parttype in batches between cells; however, it seems not practical enough. Other example is a Safaei and TavakkoliMoghaddam [21], although this is a very comprehensive dynamic integrated model and utilizes mathematical linear programming to solve problems, but for material handling cost feature, this research does not take into account different travel distance between cells. Ah kioon et al., [26] research is also a very wellintegrated model too, but it does not take into consideration some issues that are addressed within this research paper such as machine setup cost (with respect to the machine setup cost for each operation) or they consider intercell material handling cost feature, but they do not take into account different travel distance between cells too. Mahdavi et al., [27] or other called researches proposed their integrated model and solved numerical examples by their models, but they do not try to evaluate and verify performance of their models, although we try to evaluate and verify the CF section performance of the proposed model within this research.
This study aims at introducing a comprehensive dynamic deterministic integrated model to offer an optimal solution for grouping the part families and machine cells, as well as an optimal production plan for minimizing inventory and machine setup costs. The aim of proposed model is to minimize machine operating cost, intercell material handling cost (with respect to the different travel distances between cells), machine operating cost, finishedgoods inventory cost, and machine setup cost (with respect to the machine set up cost for each operation). Our model considers cell size limitation, machine duplication in one period time, and operation sequence features too.
The rest of this paper is organized as follows. Detailed description of the problem and the proposed model are presented in Section 2. Some numerical examples are presented in Section 3, in order to illustrate the proposed model. Discussions to verify the model are presented in Section 4. Summary and conclusions of the research are discussed in Section 5.
2. Mathematical Model Development
2.1. Problem Description
In this section, a mixed integer nonlinear programming model is developed to simultaneously solve the cell formation and production planning problems in CM systems. The objective function of this model is to minimize intercell material handling cost, machine operating cost, production setup cost for every operation in every part, and part inventory cost. A manufacturing system which includes a number of machines to process different parttypes is considered. Each parttype has a number of operations, and that must be processed as numbered respectively, in order to take the sequence of operations into account. The manufacturing system is considered in a number of time periods , where , with . One time period could be a day, a week, or a month. Demands for different parttypes are assumed to be known and deterministic. Demands may be satisfied from production in the same time period or from inventory. Back orders and shortage cost are not considered in this study. This model assumes that there is a single process plan for each parttype. Since both the quantity of different parttypes to be processed by the machines and the formation of the machine cells are decision variables, nonlinear terms are presented in the objective function of the mixed integer programming model.
2.2. Notations
Indices:?: time period index: .?: parttype index: .?: index of operations of parttype .?: cell index: .?: machine index: .
Coefficients and Parameters:?: unit inventory holding cost of parttype for time period .?: known demand of parttype for time period .?: unit machine operating cost for machinetype in time period .?: setup cost to performing operation on machine from parttype .?: cost of moving a unit of parttype from cell to cell .?: minimum number of machines in cell .?: maximum number of machines in cell .
Binary Decision Variables: Subscripts of variable indicates that machine is required to process operation of parttype . This information is known from the given part process plan.
Continuous Decision Variables:?: amount of parttype to be processed in time period ,?: amount of parttype in inventory at the end of time period .
2.3. Mathematical Models Development
With the given notations, the proposed model is formulated as follows: Subject to: The objective function of the proposed model has been shown in (2), and it consists of four terms. The first term of the objective function is the machine operating cost. It is assumed that the machines can be included when they are needed and can be removed from the system when they are not required. The second term of the objective function is the intercell material handling cost. This cost function is similar to those in Atmani et al. [28] and Chen [25]. In a typical cell formation problem, the number of parttype to be produced in time , , is usually considered constant. In that case, the material handling cost function will be linear. However, in this model, the term of the material handling cost is nonlinear, because it has been assumed that the distances between each pair of cells are different (parttype after completion of its operation by machine in cell moves to machine for its next operation, , in cell ). It is further assumed that the specifications of different parttypes (e.g., size or volume of different parttypes) do not influence the material handling cost. The third term is the finished goods inventory cost, and the last term of the objective function is the system setup cost with consideration of setup cost for each operation in each parttype. The first two cost items in the objective function are related to forming manufacturing cells, while the latter two cost items are related to production and inventory control.
Constraints of the model consist of equations and inequalities (3) to (10). Equation (3) is to ensure that if operation of parttype will be processed by machine in one of the cells in time period , then the corresponding binary variable for system setup must be 1. Inequality (4) ensures that once machine is assigned to cell in time period , then the operations of parttypes may be assigned to that machine. Inequality (5) ensures that sufficient machine capacity is assigned to each cell in time period t. Inequality (6) consists of two constraints for the upper and lower limits to the number of machines in each cell. The upper limit for the number of machines is due to the limitations of the physical space. Furthermore, there should be at least one machine in each cell, otherwise the cell will disappear. Equation (7) provides the relationship of storage levels at different time periods. In fact, in planning the production, demand of parttype at time should be deducted from the finished parts in storage at time . Equation (8) implicates that the production in the entire planning horizon definitely meets the total demand. Equation (9) shows the relationship between the setup variable and the part processing quantity . Equation (10) imposes nonnegativity and integrality, respectively.
2.4. Model Linearization
The proposed model is a nonlinear mixedinteger programming model because of the existing two nonlinear parts including the second term of objective function (see (2)) and the seventh constraint (see (9)). In order to find a global solution, the nonlinear model should be transformed into the linear form. Hence, the linearization phase is implemented for these two parts. Consider the second term of the objective function: It can be modified as follows: In order to linearize the previous expression, let us assume that These variables imply that Finally, the second term of objective function can be replaced by the following linear expression: Furthermore, the following constraints should be added to this model: Inequalities (17) and (18) imply that is equal to 1, if one unit of parttype is moved to machine in cell for operation after performing operation on machine in cell . Inequalities (19) and (20) enforce to be when is equal to 1. The last constraint (21) enforces to be 0, when is equal to 0.
For (9) which should be linearized, the conditional 01 requirement for variable can be simply converted to the following set of constraints: where are 01 variables and is a large positive number. One can easily verify that (22) serve the same purpose as (9) in the model.
2.5. A Single Model for Cell Formation Problem
The proposed model (see (2) to (10)) can be simplified through some modifications in the formulation, in order to solve the cell formation problem, without considering the production planning issues. In this model, the variable is equal to . After linearization, the model can be expressed as follows: Subject to
3. Numerical Examples
Two numerical examples with different structures from the existing literature are presented in this section. The examples have been solved using LINGO 12.0, a commercially available optimization software, on a personal computer with Intel Core2 Duo T6400 @ 2.00?GHz processor and 4 GB RAM.
3.1. Example 1
The data used in this example has been adapted from Chen [25] with slight modifications. In this example, 3 cells, 3 timeperiods, 10 parttypes, and 7 machinetypes are considered. The minimum and maximum numbers of machines in each cell are 1 and 5, respectively. Detailed production demand and unit inventory cost to hold each parttype are presented in Table 1. Table 2 shows machine operating costs, partmachine requirements, and refixturing costs. Machine operating costs are presented in the second column of this table. For example, it will cost 15 units to operate 1 unit of machine 1 in the system 10 units for machine 2, and so on. The 6th column of Table 2 shows that there are 2 operations for processing part 4. It also indicates that machine 4 and 7 are required to perform operations 2 and 1, respectively, for part 4. Meanwhile, intercell material handling costs are shown in Table 3.



It is assumed that the specifications of time periods do not influence the productionrelated costs which were presented through Tables 1 to 3 (e.g, there is no inflation). Considering the partoperation requirements in Table 2, in order to reduce the number of variables and constraints, the variables which can be fixed to zero were removed from the model using sparse set membership filtering technique of LINGO [29]. After fixing these variables, some constraints became redundant and were subsequently removed. LINGO solver defined the model of Example 1 as an integer linear problem (MILP) and used the branch and bound (BandB) method to solve it. The resulting formulation has a total of 1089 variables and 1526 constraints. The solution was achieved after 102 minutes of the solver running. The total cost of this problem (which appears as objective value in LINGO solution report) was 997.40 units. Table 4 shows the solution matrix for the cell formation problem (machines and parttypes groups) in timeperiod 1. In addition, Table 5 presents machine cells for different time periods. The entries in this table are the machine numbers.


From Tables 4 and 5, it can be observed that multiple units of the same machine can be used in different cells. However, the usage of more than one unit of each machinetype in the same cell in neither necessary nor economical, since there is no machine capacity constraint in the model. For example, when , there are 2 units of machinetype 1 in cells 1 and 2, and 2 units of machinetype 4 in cells 1 and 3. Table 6 presents the solution of production planning problem. It shows the times and amounts of production to satisfy demands of all time periods. The sequence of parts in Table 6 the is same as that in the solution matrix for (Table 4). From Tables 2, 5 and 6, it can be observed that, for example, 22 units of parttype 7 processed in will flow from cell 3 to cell 1 to be processed by machines 7 and 4, respectively.

3.2. Example 2
The data set of this example have been used from Mungwattana [12] with little modifications. In this example, 3 cells, 3 time periods, 19 parttypes, and 12 machinetypes are considered. The minimum and maximum numbers of machines in each cell are 3 and 9, respectively. Production demands for 19 parttypes in 3 timeperiods and unit inventory cost to hold each parttype are shown in Table 7. Table 8 shows machine operating costs, partmachine requirements, and refixturing costs. In addition, intercell material handling costs are presented in Table 9.



The linear model of Example 2 consists of 3015 variables (including 849 integer variables) and 4421 constraints. The solver was interrupted by the authors, since it was unable to obtain the results after almost 70 hours of running. It is due to increasing the number of variables and constraints in comparison with Example 1.
However, this example can be solved by the single cell formation model which was presented in Section 2.5. In this case, under the single model for the cell formation and with one time period, the model includes 663 variables and 727 constraints. The solver obtained the results after almost 6 minutes of running and objective function value was 19335. Table 10 shows the solution matrix of Example 2 (for one time period).

4. Performance Comparison
Several different CF problemsolving methodologies have been proposed in the literature. Likewise, various performance measures have been suggested in order to evaluate and compare the efficiency of these methodologies. Hsu [30] proposed the group capability index (GCI) as a measure of goodness and claimed that this measure maybe is more consistent in predicting the suitability of a manufacturing system for CM. GCI simultaneously considers production volume and processing time of operations factors and excludes voids (“zero” entries) from the calculation of goodness. GCI is defined by where is the number of exceptional elements in the machinepart matrix and is the total number of “one” entries in the machinecomponent matrix.
Therefore, in this section, the CF results of the proposed model are compared with the results of rank order clustering (ROC) method [31] and direct clustering analysis (DCA) [32] methods, by means of GCI measure. Table 11 shows this comparison based on the results presented in Tables 4 and 10 and also the results of ROC and DCA methods. For the sake of concise presentation, those steps required to solve the examples with ROC and DCA methods are not described in this paper. From Table 11, it is observed that the proposed model represents significant improvements in GCI, in comparison with the ROC and DCA methods for Examples 1 and 2.

5. Conclusions
In this paper, a comprehensive dynamic deterministic integrated mathematical model was developed to simultaneously solve the production planning and cell formation problems in CM systems. The overall objective function of the proposed model minimizes the intercell material handlings (with respect to the different travel distances between cells), machine operating cost finished goods inventory, and machine setup costs (considering the machine set up cost for each operation) for CM systems. The constraints of proposed model define the relationships between variables, such as the relationship of storage levels at different time periods, and upper machine and minimum machine limits for each cell. Some numerical examples with different sizes from the existing literature were considered to test and illustrate the proposed model. The examples were solved by means of LINGO optimization software. In order to evaluate the performance of the proposed model, it was compared with the rank order clustering (ROC) method and direct clustering analysis (DCA) method, using group capability index (GCI) measure. It was shown that the proposed CF model has a higher and satisfactory performance.
The proposed mathematical model offers the advantage of solving CF problem with high performance, while it simultaneously considers the production planning issues with sequence data. However, it can be observed that the proposed model is not suitable for solving the large scale problems. Hence, the use of heuristic methods to deal with such problems deserves further study.
References
 N. Singh, “Design of cellular manufacturing systems: an invited review,” European Journal of Operational Research, vol. 69, no. 3, pp. 284–291, 1993. View at: Google Scholar
 A. Tariq, I. Hussain, and A. Ghafoor, “A hybrid genetic algorithm for machinepart grouping,” Computers & Industrial Engineering, vol. 56, no. 1, pp. 347–356, 2009. View at: Google Scholar
 U. . Wemmerlöv and N. L. Hyer, “Cellular manufacturing in the U.S. Industry: a survey of users,” International Journal of Production Research, vol. 27, no. 9, pp. 1511–1530, 1989. View at: Google Scholar
 S. Kalpakjian and S. R. Schmid, Manufaturing Enginnering and Technology, Prentice Hall, 6th edition, 2010.
 S. S. . Heragu, “Group technology and cellular manufacturing,” IEEE Transactions on Systems Man and Cybernetics, vol. 24, no. 2, pp. 203–215, 1994. View at: Google Scholar
 R. . Shanker and P. Vrat, “Some design issues in cellular manufacturing using the fuzzy programming approach,” International Journal of Production Research, vol. 37, no. 11, pp. 2545–2563, 1999. View at: Google Scholar
 M. Murugan and V. Selladurai, “Optimization and implementation of cellular manufacturing system in a pump industry using three cell formation algorithms,” International Journal of Advanced Manufacturing Technology, vol. 35, no. 1, pp. 135–149, 2007. View at: Google Scholar
 J. McAuley, “Machine grouping for efficient production,” Production Engineering, vol. 51, no. 2, pp. 53–57, 1972. View at: Google Scholar
 R. G. Askin et al., “A Hamiltonian path approach to reordering the partmachine matrix for cellular manufacturing,” International Journal of Production Research, vol. 29, no. 6, pp. 1081–1100, 1991. View at: Google Scholar
 S. J. . Chen and C. S. Cheng, “A neural networkbased cell formation algorithm in cellular manufacturing,” International Journal of Production Research, vol. 33, no. 2, pp. 293–318, 1995. View at: Google Scholar
 S. Zolfaghari, “Design and plannng for cellular manufacturing: application of neural networks and advanced search techniques,” in Mechanical Engineering, Universty of Ottawa, Ottawa, Canada, 1997. View at: Google Scholar
 A. Mungwattana, Design of Cellular Manufacturing Systems for Dynamic and Uncertain Production Requirements with Presence of Routing Flexibility, Virginia Polytechnic Institute and State University, Blacksburg, Va, USA, 2000.
 R. TavakkoliMoghaddam, M. B. Aryanezhad, N. Safaei, and A. Azaron, “Solving a dynamic cell formation problem using metaheuristics,” Applied Mathematics and Computation, vol. 170, no. 2, pp. 761–780, 2005. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 R. Logendran, “A binary integer programming approach for simultaneous machinepart grouping in cellular manufacturing systems,” Computers & Industrial Engineering, vol. 24, no. 3, pp. 329–336, 1993. View at: Google Scholar
 S. . Zolfaghari and M. Liang, “Comprehensive machine cell/part family formation using genetic algorithms,” International Journal of Manufacturing Technology and Management, vol. 15, no. 6, pp. 433–444, 2004. View at: Google Scholar
 H. Seifoddini, “A probabilistic model for machine cell formation,” Journal of Manufacturing Systems, vol. 9, no. 1, pp. 69–75, 1990. View at: Google Scholar
 M. Chen, “A mathematical programming model for system reconfiguration in a dynamic cellular manufacturing environment,” Annals of Operations Research, vol. 77, no. 0, pp. 109–128, 1998. View at: Google Scholar
 E. M. Wicks, Designing Cellular Manufacturing Systems with Time Varying Product Mix and Resource Availability, Virginia Polytechnic Institute and State University, Blacksburg, Va, USA, 1995.
 G. Harhalakis, R. Nagi, and J. M. Proth, “An efficient heuristic in manufacturing cell formation for group technology applications,” International Journal of Production Research, vol. 28, no. 1, pp. 185–198, 1990. View at: Google Scholar
 G. Harhalakis, G. Ioannou, I. Minis, R. Nagi et al., “Manufacturing cell formation under random product demand,” International Journal of Production Research, vol. 32, no. 1, pp. 47–64, 1994. View at: Google Scholar
 N. Safaei and R. TavakkoliMoghaddam, “Integrated multiperiod cell formation and subcontracting production planning in dynamic cellular manufacturing systems,” International Journal of Production Economics, vol. 120, no. 2, pp. 301–314, 2009. View at: Google Scholar
 Yu. N. Sotskov and T.C. Lai, “Minimizing total weighted flow time under uncertainty using dominance and a stability box,” Computers & Operations Research, vol. 39, no. 6, pp. 1271–1289, 2012. View at: Publisher Site  Google Scholar
 N. M. Matsveichuk, Yu. N. Sotskov, N. G. Egorova, and T.C. Lai, “Schedule execution for twomachine flowshop with interval processing times,” Mathematical and Computer Modelling, vol. 49, no. 56, pp. 991–1011, 2009. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 B. . Arvindh and S. A. Irani, “Cell formation the need for an integrated solution of the subproblems,” International Journal of Production Research, vol. 32, no. 5, pp. 1197–1218, 1994. View at: Google Scholar
 M. Chen, “A model for integrated production planning in cellular manufacturing systems,” Integrated Manufacturing System, vol. 12, no. 4, pp. 275–284, 2001. View at: Google Scholar
 S. Ah kioon, A. A. Bulgak, and T. Bektas, “Integrated cellular manufacturing systems design with production planning and dynamic system reconfiguration,” European Journal of Operational Research, vol. 192, no. 2, pp. 414–428, 2009. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 I. Mahdavi, A. Aalaei, M. M. Paydar, and M. Solimanpur, “Designing a mathematical model for dynamic cellular manufacturing systems considering production planning and worker assignment,” Computers & Mathematics with Applications, vol. 60, no. 4, pp. 1014–1025, 2010. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 A. Atmani, R. S. Lashkari, and R. J. Caron, “A mathematical programming approach to joint cell formation and operation allocation in cellular manufacturing,” International Journal of Production Research, vol. 33, no. 1, pp. 1–15, 1995. View at: Google Scholar
 LINDO Systems Inc, LINGO User's Guide, LINDO System, Chicago, Ill, USA, 2010.
 C. P. Hsu, The Similarity Coefficient Approaches of MachineComponent Cell Formation in Cellular Manufacturing: A Comparative Study, in Department of Industrial and Systems Engineering, of WisconsinMilwaukee, Milwaukee, Wis, USA, 1990.
 M. P. Groover, Automation, Production Systems, and ComputerIntegrated Manufacturing, Prentice Hall Press, 2007.
 U. Wemmerlov and N. Hyer, “Comments on direct clustering algorithm for group formation in cellular manufacture,” Journal of Manufacturing Systems, vol. 3, no. 1, 1984. View at: Google Scholar
Copyright
Copyright © 2013 Reza Raminfar et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.