A PSO-based approach to cell formation problems with alternative process routings

Group technology (GT) has been extensively applied to cellular manufacturing system (CMS) design for decades due to many benefits such as decreased number of part movements among cells and increased machine utilisation in cells. This paper considers cell formation problems with alternative process routings and proposes a discrete particle swarm optimisation (PSO) approach to minimise the number of exceptional parts outside machine cells. The approach contains two main steps: machine partition and part-routing assignment. Through inheritance and random search, the proposed algorithm can effectively partition machines into different cells with consideration of multiple part process routings. The computational results are compared with those obtained by using simulated annealing (SA)-based and tabu search (TS)-based algorithms. Experimental results demonstrate that the proposed algorithm can find equal or fewer exceptional elements than existing algorithms for most of the test problems selected from the literature. Moreover, the proposed algorithm is further tailed to incorporate various production factors in order to extend its applicability. Four sample cases are tested and the results suggest that the algorithm is capable of solving more practical cell formation problems.     

Evolving knowledge for the solution of clustering problems in cellular manufacturing

Hierarchical clustering has been widely used for the solution of problems in the area of cellular manufacturing. Hierarchical clustering procedures utilize coefficients that quantify the level of similarity between pairs of machines or parts in the plant. An evolutionary methodology is proposed for the construction of new similarity coefficients that can be used by standard hierarchical clustering methodologies for the solution of cell-formation problems. A typical application is presented for the simplest case of the cell-formation problem. However, alternative similarity coefficients can be evolved for advanced formulations of the problem by suitably modifying the set of fitness cases that constitute the environment of the evolutionary process.     

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.     

Intelligent dispatching for flexible manufacturing

A decision rule for real-time dispatching of parts, each of which may have alternative processing possibilities, has been developed and tested in a simulated flexible manufacturing system. A part, upon completion of an operation, is not routed to a specific machine, but is, in effect, sent to a general queue. Thus, a machine has a global option for choosing parts which in turn may be processed on alternative machines. For effective use of the system's routeing flexibility under these circumstances, the machine needs an intelligent part-selection strategy (rather than shallow heuristics represented by the conventional dispatching rules) that takes into account the current state and trends of the system. The proposed intelligent reasoning procedure has been found to achieve better shop performance than some of the popular dispatching rules, the improved performance being due to the ability to respond to changing circumstances.     

Designing multi-product lines: job routing in cellular manufacturing systems

An operation sequence-based method which integrates intra-cell layout cost with cell formation to minimize the total cost of the materials flow and machine investment is developed here for designing a cellular manufacturing system. The method comprises three distinct approaches: part-family formation, cell-formation, and layout configuration. In the first phase, an operation sequence-based similarity coefficient is applied in a p-median model to group the parts to form part families with similar operation sequences. In the second phase, machine assignment to part families is determined where a trade-off between potential inter-cell movement cost due to the. bottleneck machine and the potential benefit of assigning bottleneck machines to certain part-family is considered. In the third phase, intra-cell layout is determined for each cell so as to refine the initial layout of the cell further. Numerical examples are employed to demonstrate the mechanism of the procedure throughout all phases. A comparative study is also performed to support the present method     

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.     

A general search algorithm for cell formation in group technology

Group Technology (GT) is a manufacturing approach, which organizes and uses the information about an item's similarity (parts and/or machines) to enhance efficiency and effectiveness of batch manufacturing systems. The application of group technology to manufacturing requires the identification of part families and formation of associated machine-cells. One approach is the Similarity Coefficient Method (SCM), an effective clustering technique for forming machine cells. SCM involves a hierarchical machine grouping process in accordance with computed ‘similarity coefficients’. While SCM is capable of incorporating manufacturing data into the machine-part grouping process, it is very sensitive to the data to be clustered (Chan and Milner 1982). It has been argued that for SCM to be meaningful, all machines must process approximately the same numbers of parts (Chan and Milner 1982).We present a new approach, based on artificial intelligence principles, to overcome some of these problems by incorporating an evaluation function into the grouping process. Our goal is to provide a method that is both practical and flexible in its use for the process of cell formation. Our method uses the similarity matrix to generate the feasible machine groups. Then an evaluation function is applied to select a machine-cell arrangement through an iterative process. The approach features a graph-based representation (N-tuple) to represent the problem and illustrate the solution strategies. Also, we develop an algorithm to search for the most promising machine groups from the graph. Compared with Single Linkage Clustering and Average Linkage Clustering approaches, our approach attains comparable or better results     

A scatter search approach with dispatching rules for a joint decision of cell formation and parts scheduling in batches

A joint decision of cell formation and parts scheduling is addressed for a cellular manufacturing system where each type of machine and part may have multiple numbers and parts must require processing and transferring in batches. The joint decision problem is not only to assign batches and associated machine groups to cells, but also to sequence the processing of batches on each machine in order to minimise the total tardiness penalty cost. A nonlinear mixed integer programming mathematical model is proposed to formulate the problem. The proposed model, within nonlinear terms and integer variables, is difficult to solve efficiently for real size problems. To solve the model for practical purposes, a scatter search approach with dispatching rules is proposed, which considers two different combination methods and two improvement methods to further expand the conceptual framework and implementation of the scatter search so as to better fit the addressed problem. This scatter search approach interactively uses a combined dispatching rule to solve a scheduling sub-problem corresponding to each integer solution visited in the search process. A computational study is performed on a set of test problems with various dimensions, and computational results demonstrate the effectiveness of the proposed approach.     

A cellular similarity coefficient algorithm for the design of manufacturing cells

The development and implementation of an integrated system for computer-aided process planning and cellular manufacturing design is described. A coding scheme is proposed for classifying parts according to the manufacturing processes required to produce them. The routeing data and the classification codes obtained from computer-aided process planning are then stored in a database. A heuristic algorithm has also been developed for designing manufacturing cells using the classification data stored in the process planning database. This algorithm is based on a new concept developed in this study called cellular similarity coefficient which considers the similarity between machine cells rather than individual machines, as in the case of other similarity coefficients.     

Design of delayed reconfigurable manufacturing system based on part family grouping and machine selection

To improve the convertibility of reconfigurable manufacturing system (RMS), the concept of delayed reconfigurable manufacturing system (D-RMS) was proposed. RMS and D-RMS are both constructed around part family. However, D-RMS may suffer from ultra-long system problem with unacceptable idle machines using generic RMS part families. Besides, considering the complex basic system structure of D-RMS, machine selection of D-RMS should be addressed, including dedicated machine, flexible machine, and reconfigurable machine. Therefore, a system design method for D-RMS based on part family grouping and machine selection is proposed. Firstly, a part family grouping method is proposed for D-RMS that groups the parts with more former common operations into the same part family. The concept of longest relative position common operation subsequence (LPCS) is proposed. The similarity coefficient among the parts is calculated based on LPCS. The reciprocal value of the operation position of LPCS is adopted as the characteristic value. The average linkage clustering (ALC) algorithm is used to cluster the parts. Secondly, a machine selection method is proposed to complete the system design of D-RMS, including machine selection rules and the dividing point decision model. Finally, a case study is given to implement and verify the proposed system design method for D-RMS. The results show that the proposed system design method is effective, which can group parts with more former common operations into the same part family and select appropriate machine types.     

Multi-objective cell formation and production planning in dynamic virtual cellular manufacturing systems

This article presents a fuzzy goal programming-based approach for solving a multi-objective mathematical model of cell formation problem and production planning in a dynamic virtual cellular manufacturing system. In a dynamic environment, the product mix and part demand change over a planning horizon decomposed into several time periods. Thus, the cell formation done for one period may be no longer efficient for subsequent periods and hence reconfiguration of cells is required. Due to the variation of demand and necessity of reconfiguration of cells, the virtual cellular manufacturing (VCM) concept has been proposed by researchers to utilise the benefits of cellular manufacturing without reconfiguration charges. In a VCM system, machines, parts and workers are temporarily grouped for one period during which machines and workers of a group dedicatedly serve the parts of that group. The only difference of VCM with a real CM is that machines of the same group are not necessarily brought to a physical proximity in VCM. The virtual cells are created periodically depending on changes in demand volumes and mix, as new parts accumulate during a planning horizon. The major advantage of the proposed model is the consideration of demand and part mix variation over a multi-period planning horizon with worker flexibility. The aim is to minimise holding cost, backorder cost and exceptional elements in a cubic space of machine–part–worker incidence matrix. To illustrate the applicability of the proposed model, an example has been solved and computational results are presented.     

Multi-channel manufacturing with compatible machines

Multi-channel manufacturing (MCM) is a noteworthy approach offered as a kind of fractal manufacturing. Here cells are organised as channels. Machines are lined up to facilitate streamlined materials flow. Their replication in individual and separate channels is allowed. The machines are grouped in such a way to form channels that any product is prone to be produced at least in one channel. In addition, important parts are subject to production in more channels to prevent congestion. On the other hand, one-to-one association of operations to machines as in classical grouping techniques seems very restrictive. In fact any machine is capable of fulfilling more then one kind of task, and any operation is possibly accomplished in more than one kind of machine. In other words, compatible machines should not be overlooked. In this study, enhancing of MCM with compatible machines is aimed. Numerical experience has proved the superiority of the proposed method to form manufacturing cells.     

Multi-agent-based approach to solve part selection and task allocation problem in flexible manufacturing systems

Agent technology is currently being considered as an important approach for developing intelligent manufacturing systems. It offers a new way of thinking about many of the classical problems in manufacturing engineering. A multi-agent-based approach for solving the part allocation problems in flexible manufacturing systems (FMS) is presented that can easily cope with the dynamic environment. Four agents were involved in carrying out the tasks of allocating parts on different machines: communicator, machine, part and material handling device (MHD). Upon arrival in the manufacturing facility, the part informs the communicator agent about the task requirements. The communicator agent divides the task into subtasks and sends a call-for-bids message to the machine and MHD agents. Each machine responds in accordance with its process capabilities and buffer limit. This response may be for the whole task or for one or more subtasks and it contains the price and cost details for these subtasks along with the performance index and acceptance ratio of the machine. The final allocation is made based on the objective function that includes processing and transportation costs and time. An algorithm is presented that is used by the communicator agent for allocating parts to different machines. An illustrative example is given to solve the task allocation on five machines, with each machine having different performance index and acceptance ratio.     

Using Functional Virtual Population as assistance to learn scheduling knowledge in dynamic manufacturing environments

When a scheduling environment is static and system attributes are deterministic, a manufacturing schedule can be obtained by applying analytical tools such as mathematical modelling technology, dynamic programming, the branch- and-bound method or other developed searching algorithms. Unfortunately, a scheduling environment is usually dynamic in a real manufacturing world. A production system may vary with time and require production managers to change schedule repeatedly. Therefore, the main aim here was to find a scheduling method that could reduce the need for rescheduling. An approach called Functional Virtual Population was proposed as assistance to learn robust scheduling knowledge for manufacturing systems under rationally changing environments. The used techniques include machine learning with artificial neural networks and IF-THEN scheduling rules. To illustrate the study in detail, a simulated flexible manufacturing system consisting of four machines, four parts, one automatic guided vehicle and eight buffers was built as the foundation for learning the concept. Also, Pythia software (a back-propagation-based neural networks) was employed as the learning tool in the learning procedure.     

An agent-based negotiation approach to integrate process planning and scheduling

This paper presents the development of an agent-based negotiation approach to integrate process planning and scheduling (IPPS) in a job shop kind of flexible manufacturing environment. The agent-based system comprises two types of agents, part agents and machine agents, to represent parts and machines respectively. For each part, all feasible manufacturing processes and routings are recorded as alternative process plans. Similarly, alternative machines for an operation are also considered. With regard to the scheduling requirements and the alternative process plans of a part, the proposed agent-based IPPS system aims to specify the process routing and to assign the manufacturing resources effectively. To establish task allocations, the part and machine agents have to engage in bidding. Bids are evaluated in accordance with a currency function which considers an agent's multi-objectives and IPPS parameters. A negotiation protocol is developed for negotiations between the part agents and the machine agents. The protocol is modified from the contract net protocol to cater for the multiple-task and many-to-many negotiations in this paper. An agent-based framework is established to simulate the proposed IPPS approach. Experiments are conducted to evaluate the performance of the proposed system. The performance measures, including makespan and flowtime, are compared with those of a search technique based on a co-evolutionary algorithm.     

Manufacturing cell design with alternative routings in generalized group technology: reducing the complexity of the solution space

In the presence of alternative routings, the manufacturing cell design involves the solution of a large number of (simple) cell-formation problems. We present a bounding scheme which examines all combinations of alternative routings and solves only a few cell-formation problems, thereby limiting the solution space which is searched heuristically. This leads to increased reliability for the solutions obtained. The resulting algorithm is shown to be applicable for problems arising in practice frequently, when the initial design includes relatively few exceptional elements, and machine duplication and/or genuine multiple process plans are used correctively, to eliminate the remaining intercellular traffic.     

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.     

A hybrid Hopfield network-genetic algorithm approach to optimal process plan selection

In the automated manufacturing environment, different sets of alternative process plans can normally be generated to manufacture each part. However, this entails considerable complexities in solving the process plan selection problem because each of these process plans demands specification of their individual and varying manufacturing costs and manufacturing resource requirements, such as machines, fixtures/jigs, and cutting tools. In this paper the problem of selecting exactly one representative from a set of alternative process plans for each part is formulated. The purpose is to minimize, for all the parts to be manufactured, the sum of both the costs of the selected process plans and the dissimilarities in their manufacturing resource requirements. The techniques of Hopfield neural network and genetic algorithm are introduced as possible approaches to solve such a problem. In particular, a hybrid Hopfield network-genetic algorithm approach is also proposed in this paper as an effective near-global optimization technique to provide a good quality solution to the process plan selection problem. The effectiveness of the proposed hybrid approach is illustrated by comparing its performance with that of some published approaches and other optimization techniques, by using several examples currently available in the literature, as well as a few randomly generated examples.     

Optimization-based parts-machines matching in flexible manufacturing systems

A new dynamic scheduling strategy, Parts-Machines Matching (PMM), is developed and tested in simulated flexible manufacturing systems. This strategy is aimed to achieve globally optimal matching between parts and machines by a semi-qualitative optimization algorithm, originally developed for the Stable Marriage Problem. Global and Partial implementations of PMM are presented and compared with other conventional part-flow rules. They are found to achieve better shop performance than conventional rules, in terms of system throughput, robustness against travel time uncertainties, and recovery from machine breakdowns. The prospect of bringing about system-wide optimization-based performance improvements into bidding schemes makes the proposed framework very significant.     

A within-cell utilization based heuristic for designing cellular manufacturing systems

The initial stage in the facilities design of cellular manufacturing systems involves the identification of part families and machine groups and forming cells possessing specific manufacturing capabilities. A new heuristic for the part family/machine group formation (PF/MGF) problem is presented in this paper. The distinguishing feature of this heuristic is its consideration of several practical criteria such as within-cell machine utilization, work load fractions, maximum number of machines that are assigned to a cell, and the percentage of operations of parts completed within a single cell. Computational results, based on several examples from the literature, show that this heuristic performs well with respect to more than one criteria. The heuristic also clarifies the source of various arguments in the literature concerning the ‘goodness’ of the solutions obtained by other researchers. An application of the heuristic to a large sample of industrial data involving 45 workcentres composed of 64 machines, and 305 parts is given, and the usefulness of the heuristic for trading-off several objectives is illustrated.