Fuzzy ART/RRR-RSS: a two-phase neural network algorithm for part-machine grouping in cellular manufacturing

In this paper an efficient methodology adopting Fuzzy ART neural network is presented to solve the comprehensive part-machine grouping (PMG) problem in cellular manufacturing (CM). Our Fuzzy ART/RRR-RSS (Fuzzy ART/ReaRRangement-ReaSSignment) algorithm can effectively handle the real-world manufacturing factors such as the operation sequences with multiple visits to the same machine, production volumes of parts, and multiple copies of machines. Our approach is based on the non-binary production data-based part-machine incidence matrix (PMIM) where the operation sequences with multiple visits to the same machine, production volumes of parts, and multiple identical machines are incorporated simultaneously. A new measure to evaluate the goodness of the non-binary block diagonal solution is proposed and compared with conventional performance measures. The comparison result shows that our performance measure has more powerful discriminating capability than conventional ones. The Fuzzy ART/RRR-RSS algorithm adopts two phase approach to find the proper block diagonal solution in which all the parts and machines are assigned to their most preferred part families and machine cells for minimisation of inter-cell part moves and maximisation of within-cell machine utilisation. Phase 1 (clustering phase) attempts to find part families and machines cells quickly with Fuzzy ART neural network algorithm which is implemented with an ancillary procedure to enhance the block diagonal solution by rearranging the order of input presentation. Phase 2 (reassignment phase) seeks to find the best proper block diagonal solution by reassigning exceptional parts and machines and duplicating multiple identical machines to cells with the purpose of minimising inter-cell part moves and maximising within-cell machine utilisation. To show the robustness and recoverability of the Fuzzy ART/RRR-RSS algorithm to large-size data sets, a modified procedure of replicated clustering which starts with the near-best solution and rigorous qualifications on the number of cells and duplicated machines has been developed. Experimental results from the modified replicated clustering show that the proposed Fuzzy ART/RRR-RSS algorithm has robustness and recoverability to large-size ill-structured data sets by producing highly independent block diagonal solution close to the near-best one.     

Network flow procedures for the analysis of cellular manufacturing systems

A three-phase network-flow-based procedure is developed for minimizing intercellular part moves in machine-part grouping problems. The unique feature of this methodology is its consideration of several variations related to the number of cells, the number of machines in each cell, and the part family size. The first phase computes a functional relationship between machines on the basis of either a machine-part matrix or actual operation sequences for the parts being considered. The final purpose of this phase is a network modeling of the problem. The second phase partitions the network according to mutually exclusive sets of nodes that represent manufacturing cells. A 0-1 integer programming model and a 0-1 quadratic programming model are discussed and network-flow-based solution procedures are developed. Finally, the third phase identifies the part families. A 0-1 integer programming model is formulated and the solution of this model is again performed through a network approach that allows the identification of a feasible assignment of parts to machine cells. Computational results indicate that the proposed approach is appropriate for solving large-scale industrial problems efficiently.     

Application of simulated annealing to a linear model forthe formulation of machine cells ingroup technology

The central issue in group technology is the cell formation problem, which involves the grouping of parts into families and machines into cells, so that parts with similar manufacturing (and design) attributes are identified and processed by dedicated cells of machines. In the present work, the cell formation problem is modelled as a linear integer programming problem with the objective of minimizing the number of intercellular moves subject to cell-size constraints and taking into account the machine operation sequence of each part. An interesting feature of the proposed formulation is that there is no need of specifying a priori the number of cells to be used, which is automatically adjusted within the solution procedure. A very efficient random search heuristic algorithm, based on the simulated annealing method, is adopted for its solution. The heuristic is tested on a number of problems and its performance is evaluated. Subsequently, a straight forward model is presented to identify the families of parts which are to be processed by the corresponding machine cells.     

Assignment allocation and simulated annealing algorithms for cell formation

In this paper a nonlinear mathematical programming model is developed for cell formation that identifies part families and machine groups simultaneously with no manual intervention or subjective judgement. The objective of the model is minimization of the weighted sum of the voids and the exceptional elements. Changing weights for void and exceptional elements aids the designer with a systematic generation of different solutions, i.e., forming large loose cells or small tight cells. An assignment allocation algorithm (AAA) and a simulated annealing algorithm (SAA) are developed to solve the model. AAA and SAA compare favorably with many well-known procedures for the problems tested. AAA is less computer-intensive and hence large problems with 400 parts and 240 machines were solved with AAA in less than a minute on Sun Sparc station. However, AAA is sensitive to the initial machine grouping solution input to the algorithm. SAA gives consistent results but requires more computational time.     

Models for machine-part grouping in cellular manufacturing

For cellular manufacturing strategies to succeed, the productive system first has to be divided into highly independent cells. This means that a partition of the machines into machine groups, a partition of the parts into part families, and a matching between the machine groups and the part families have to be simultaneously determined. Mathematically, this question can be expressed as the problem of finding a near block diagonal permutation of the machine-part incidence matrix. Research on such grouping problems has primarily concentrated on the design of heuristics. Different grouping efficiency criteria have been proposed to express the quality of the groupings proposed by these heuristics. This paper is concerned with mathematical programming approaches to the formation of production cells. Existing models are reviewed and their features are briefly discussed. An alternative model is proposed, which allows for the formulation of various constraints and grouping efficiency criteria. Finally, some test problems are used to support the claim that this model may be adequate for the solution to optimality of the cell formation problem.     

Simultaneous formation of machine and human cells in group technology: a multiple objective approach

Since the advent of group technology (GT) as a primary manufacturing tool for reducing setup times and improving production efficiencies, its central theme has been the grouping of similar parts into part families and machines into machine cells. Although the formation of machine-part manufacturing cells is the essence of GT, its full benefits cannot be gained without forming ‘human’ cells in such a way that machine operators with similar expertise and skills are brought together to produce similar part families. Nevertheless, much of the existing GT literature overlooks the behavioural issues associated with a group of workers in the machine cell. This paper addresses such issues by simultaneously forming both machine and compatible human cells. In so doing, we develop a multiple objective model that enables us to analyse the tradeoff between economic and behavioural benefits.     

Design and scheduling of hybridmulti-cell flexible manufacturing systems

The objective of this paper is to minimize machine duplication by increasing its utilization, minimize intercell moves, simplify the scheduling problem and increase the flexibility of the manufacturing system. An integrated approach of design and scheduling alternative hybrid multi-cell flexible manufacturing systems (MCFMSs) in four steps will be presented in this paper. The first step is the implementation of branch and bound techniques which provide tools to design group technology (GT) cells. The second step is balancing the inter-cell workload of GT cells which leads to a hybrid MCFMS with better utilization of the machines. The problem of the exception machines and their utilization and workload balance will be solved within the MCFMScentre. Thus the performance of GT cells can be improved by transferring workloads from a congested (bottleneck) machine in one cell to an alternative one, a less congested (exception) machine in another cell within a group of GT cells forming a MCFMS centre. The third step is the group scheduling; a proposed heuristic method will be used for the scheduling of a family of parts with the objective of minimizing the maximum completion time of each part. The problem of scheduling under MCFMS can be reduced by considering the scheduling of each family of parts. Finally, the flexibility of the system will be enhanced by selecting appropriate machine tools and flexible material handling equipments. This approach is both effective and efficient-it has generated a hybrid MCFMS centre which includes several alternatives, for some benchmark problems in much shorter time than algorithms previously reported in the literature. In addition, the method is conceptually simple and easy to implement.     

PRODUCTION PLANNING DECISIONS IN FLEXIBLE MANUFACTURING SYSTEMS WITH RANDOM MATERIAL FLOWS

In this paper, we consider the FMS planning problem of determining optimal machine workload assignments in order to rninimize mean part flow time. We decompose this problem into the subproblems of first forming machine groups and next assigning operations to these groups. Three types of grouping configurations—no grouping, partial grouping and total grouping—are considered. In both no grouping and partial grouping, each machine is tooled differently. While each operation is assigned to only one machine in no grouping, partial grouping permits multiple operation assignments. On the other hand, total grouping partitions the machines into groups of identically-tooled machines; each machine within a group is capable of performing the same set of operations. Within this grouping framework, we consider three machine loading objectives—minimizing the total deviation from the optimal group utilization levels, minimizing part travel and maximizing routing flexibility, for generating a variety of system configurations.

A queueing network model of an FMS is used to determine the optimal configurations and machine workload assignments for the no grouping and total grouping cases. It is shown that under total grouping, the configuration of M machines into G groups that minimizes flow time is one in which the sizes of the machine groups are maximally unbalanced and the workload per machine in the larger groups is higher. This extends previous results on the optimality of unbalancing both machine group sizes and machine workload to the mean flow time criterion.

A simulation experiment is next conducted to evaluate the alternative machine configurations to understand how their relative performance depends upon the underlying system characteristics, such as system utilization level and variation among operation processing times. We also investigate the robustness of these configurations against disruptions, such as machine unreliability and variation in processing batch sizes. While different configurations minimize mean flow time under different parameter values, partial grouping with state-dependent part routing performs well across a wide range of these values. Experimental results also show that the impact of disruptions can be reduced by several means, such as aggregating operations of a part to be performed at the same machine, in addition to providing routing flexibility.     

Improved design of CMS by considering operators decision-making styles

Cell formation is one of the oldest problems in cellular manufacturing systems (CMS) including assigning parts, machines and operators to cells. Cell manufacturing contains a number of cells where each cell is responsible for processing the family of similar parts. Another important aspect of cell formation is worker assignment to cells. Since operators work together in long periods, it is suggested to consider operators’ personal characteristics to increase their satisfaction and the productivity of system. This paper considers decision-making styles of operators (as an index of operator’s personal characteristics) and presents a new mathematical programming model for clustering parts, machines and workers simultaneously. The model includes two objectives; (1) minimization of intracellular movements and cell establishment costs, (2) minimization of decision-making style inconsistency among operators in each cell. The paper applies ε-constraint method for solving the problem and gathering non-dominated solutions such as Pareto optimal solutions. Furthermore, this paper uses common weighted multicriteria decision analysis (MCDA)-data envelopment analysis method to choose the best solution from the candidate Pareto optimal solutions that have been achieved by solving the mathematical model. A real case study is investigated to show the capability of the proposed model to design CMS in the assembly unit. The proposed design assists decision-makers to develop cellular systems with more operators’ satisfaction and productivity.     

Cell design and multi-period machine loading in cellular reconfigurable manufacturing systems with alternative routing

This paper deals with the design and loading of Cellular Reconfigurable Manufacturing Systems in the presence of alternative routing and multiple time periods. These systems consist of multiple reconfigurable machining cells, each of which has Reconfigurable Machine Tools and Computer Numerical Control (CNC) machines. Each reconfigurable machine has a library of feasible auxiliary machine modules for achieving particular operational capabilities, while each CNC machine has an automatic tool changer and a tool magazine of a limited capacity. The proposed approach consists of two phases: the machine cell design phase which involves the grouping of machines into machine cells, and the cell loading phase that determines the routing mix and the tool and module allocation. In this paper, the cell design problem is modelled as an Integer Linear Programming formulation, considering the multiple process plans of each part type as if they were separate part types. Once the manufacturing cells are formed, a Mixed Integer Linear Programming model is developed for the cell loading problem, considering multi-period demands for the part types, and minimising transportation and holding costs while keeping the machine and cell utilisations in each period, and the system utilisation across periods, approximately balanced. An illustrative problem and experimental results are presented.     

The cost of eliminating exceptional elements in group technology cell formation

Many researchers have suggested methods for the formation of machine cells/part families in group technology. However, few of these methods have addressed the possible existence of exceptional elements (EE) in a reasonable manner. EE can be the result of bottleneck machines whose processing is needed by parts assigned to more than one part family. They can also be caused by parts that require processing on machines assigned to more than one machine cell. The existence of EE in cell formation solutions is a nontrivial problem that requires interaction between machine cells intended to be independent for production efficiency. This paper presents a systematic method for identifying opportunities for reducing the number of intercell transfers caused by the existence of EE. The method recognizes how each EE in a cell formation solution may be involved in the creation of intercell transfers. The sequence of operation in each part routeing is also considered. The method then analyses the costs associated with alternative actions for the removal of the EE. The result is a prioritized list (based on relative cost-effectiveness) of the EE-removal actions. The method recognizes that interdependencies exist among EE: actions taken to eliminate one EE may have an effect on others as well. The process is demonstrated with an example.     

Cell formation: the need for an integrated solution of the subproblems

Cellular manufacturing is a viable option in many manufacturing systems. There are various subproblems in the design of a cellular manufacturing system. These are machine group and part family formation, machine duplication, intracell layout and intercell layout. The only comprehensive design strategy that attempts to address all of these is production flow analysis. However, this technique is a sequential strategy where the subproblems mentioned above are assumed nested within each other and are solved in a forward pass with no feedback. This is a satisfactory approach only in cases where the part families are relatively disjoint and machine groups are formed without constraints on machine duplication to eliminate intercell flow. The presence of bottleneck machines and parts renders the problem considerably more complex, as the subproblems influence each other substantially. This paper presents an integrated framework for solving these subproblems by generating a limited set of feasible alternative solutions.     

Performance evaluation of family-based dispatching in small manufacturing cells

In many practical instances, the choice of whether to apply family-based dispatching or not can be decided per machine. The present paper explores the impact of the location of family-based dispatching, load variations between machines and routing of jobs on the flow time effect of family-based dispatching. These factors are explored in small manufacturing cells with and without labour constraints. An industrial case motivates the study. A simulation study is performed to assess the impact of these effects. The results show that shop-floor characteristics such as routing and load variation impact the decision where to locate family-based dispatching in manufacturing cells without labour constraints. By contrast, the effect of family-based dispatching is much less vulnerable to shop-floor characteristics in cells with labour constraints. Since workers are the bottleneck in these cells, it becomes less important at what machine the set-up time involving a worker is reduced. In general, there seems to be a trade-off between the positive effect of applying family-based dispatching at a (bottleneck) machine and the possible negative effect of the more irregular job arrivals at subsequent machines. The results further indicate that family-based dispatching is more advantageous in cells with labour constraints than in cells without labour constraints, when both types of manufacturing cells have comparable machine utilizations.     

Machine grouping problem in cellular manufacturing systems— an integer programming approach

In this paper a methodology is proposed to group the machines in cellular manufacturing systems based on the tooling requirements of the parts, toolings available on the machines and the processing times. Two 0-1 integer programming formulations are proposed. These formulations assume that the part families are known. The first formulation groups the machines based on the compatibility of parts with machines. The second formulation groups the machines in order to minimise the cost of allocating the machines and the cost of intercell movement. These formulations take into account the limitations on the number of machines in a group and the number of machines available of a particular type. The application of these formulations is illustrated using an example.     

A methodology for designing flexible cellular manufacturing systems

Cell formation in cellular manufacturing deals with the identification of machines that can be grouped to create manufacturing cells and the identification of part families to be processed within each cell. Dynamic and random variations in part demands can negatively impact cell performance by creating unstable machine utilizations. The purpose of this paper is to introduce and illustrate an interactive cell formation method that can be used to design 'flexible' cells. Flexibility in this context refers to routing flexibility (i.e., the ability for the cellular system to process parts within multiple cells) and demand flexibility (i.e., the ability of the cell system to respond quickly to changes in part demand and part mix). Through an experimental analysis using multiple data sets, we also validate the procedure and provide guidelines for parameter settings depending upon the type of flexibility of interest to the user. Finally, trade-offs and interdependences between alternative types of flexibility in the context of cellular systems are illustrated.     

Integrated group technology,cell formation,process planning,and production planning with application to the emergency room

In this paper an integrated approach for the formation of parts and machine families in group technology is developed. The integrated approach is used to solve cell formation, process planning, and production planning simultaneously. The given information is part processing sequence, part production volume, part alternative processing plans, and part processing times. The approach is used to determine the machine-part cells and part processing plans, while the total intercell part flow is minimized. Also, the convergence of the algorithm is investigated. The approach goes across and beyond the group technology methods by considering sequencing, production planning, process planning, and part-machine cellular information simultaneously. Two methods are investigated: exact (optimal) and heuristic. The approach first solves an integer programming problem to find processing plans and then uses a procedure to form the machine-part cells. The proposed approach solves the problem iteratively until a set of plans for machine-part cell formations is obtained with minimal intercell part flow or interflow cost. An example is presented to explain the developed approach. Experimental results are also provided. An extension of the approach for solving the operations planning of an emergency room is also covered. In this extension of the approach, the application of cell formation provides a solution to efficiently managing patients and utilizing resources. By grouping patients by their needed medical procedures, time and resource efficiency is accomplished. An application to ER of University Hospitals of Case Western Reserve University is given.     

A methodology for cellular manufacturing design

A two-phase methodology is presented as an aid to organizing job shop production in a cellular manufacturing system. The first phase (selection/assignment phase) selects the machines to be kept on the shop floor and assigns parts to the machines retained. The second phase (partition/reassignment phase) establishes a partition of the set of parts and corresponding cells of machines and reassigns some of the operations with a view to eliminating some intercell material movements. This phase is repeated until a partition meeting the operator's requirements is obtained. The results obtained with this method on several examples found in the literature are consistently equivalent to or even better than those hitherto proposed, in terms of intercell moves.     

A cut-tree-based approach for clustering machine cells in the bidirectional linear flow layout

This paper considers the joint cell clustering-layout problem where machine cells are to be located along the popular bidirectional linear material flow layout. The joint problem seeks to minimize the actual intercell flow cost instead of the typical measure that minimizes the number of intercell movements when the layout problem is excluded from the clustering process. Owing to the computational difficulty, a three-phase approach is proposed using the cut-tree-network model to solve this joint problem. The cell clustering and layout problem is transformed into a multi-terminal network flow model. A cut tree is constructed and partitioned into a number of subgraphs via the selected primary path. Each subgraph is a clustered cell and their locations are assigned to the layout sequence by comparing the cut capacities. Thus, the proposed approach concurrently determines the machine cells and their relative sequences in the bidirectional linear flow layout. Computational procedures are illustrated and additional experiments, with data adapted from the literature, are performed to demonstrate the viability of the approach.     

Selection of a Threshold Value Based on Material Handling Cost in Machine-Component Grouping

Machine-component grouping is a basic step in the application of group technology to manufacturing. It is the process of finding families of similar parts (part-families) and forming the associated machine cells such that one or more part-families can be processed within a single cell. Among the algorithms used to form the machine cells, those based on the Similarity Coefficient Method (SCM) are more flexible in incorporating the manufacturing data into the machine-component grouping process. SCM is the application of clustering techniques in forming the machine cells. One of the major problems with SCM is that it generates a set of alternative solutions rather than a unique solution. The number and size of machine cells in a given solution depends upon the similarity coefficient (threshold value) at which machine cells/machines are joined. This paper discusses the problem of selecting a proper threshold value and presents a solution to it.     

Managing flexible manufacturing technology: Must parts in an FMS always move in a job-shop manner?

The literature typically assumes that an FMS consists of a number of machines of different processing capability. Furthermore, in view of the inherent flexibility of an FMS, many researchers believe that 'flexible' implies that a wide variety of parts can be, and should be, processed simultaneously in an FMS. To do otherwise will inevitably mean some machines are idle and hence the FMS may become a tremendous waste. This paper is an attempt to study this issue. Unlike a job shop, which is typically an eclectic collection of machines of different process capabilities, most flexible manufacturing systems consist of only one or two types of versatile programmable machines. We contend that the FMS machines are analogous to the multi-skilled workers in lean production. In lean production, workers work in teams. It is common for each team to work on one job at a time, from start to finish. Jobs are not passed around from worker to worker. Likewise, machines in an FMS can be assigned into 'teams', where each 'team' is dedicated to process a single part type (or family) at a time. Parts do not necessarily have to move from machine to machine in a job shop manner.