Cell Formation Algorithm Research Articles

An analytical-iterative clustering algorithm for cell formation in cellular manufacturing systems with ordinal-level and ratio-level data

In this paper, the problem of clustering machines into cells and components into part-families with the consideration of ratio-level and ordinal-level data is dealt with. The ratio-level data is characterized by the use of workload information obtained both from per-unitprocess times and production quantity of components, and from machine capacity. In the case of ordinal-level data, we consider the sequence of operations for every component. These data sets are used in place of conventional binary data for arriving at clusters of cells and part-families. We propose a new approach to cell formation by viewing machines, and subsequently components, as 'points' in multi-dimensional space, with their coordinates defined by the corresponding elements in a Machine-Component Incidence Matrix (MCIM). An iterative algorithm that improves upon the seed solution is developed. The seed solution is obtained by formulating the given clustering problem as a Traveling Salesman Problem (TSP). The solutions yielded by the proposed clustering algorithm are found to be good and comparable to those reported in the literature.

Configuration of cellular manufacturing systems using association rule induction

A cell-formation approach based on association rule induction is developed to find the effective configurations for cellular manufacturing systems. To gain the benefits of flexibility and efficiency, the manufacturing system is decomposed into several manageable subsystems by categorizing similar parts into part families and disparate machines into cells. It is advantageous to find the important associations among machines such that the occurrence of some machines in a machine cell will cause the occurrence of other machines in the same cell. Relationships among machines can be found from the process database by inducting association rules. By applying association rules to cell-formation problems, certain sets of machines (machine groups) that frequently process some parts together can be inducted. A data-mining technique referred to as association rule induction is used herein to find the association rules among machines from the process database. Seventeen data sets of various size and complexity were used to evaluate the effectiveness of the proposed cell-formation algorithm based on association rule induction. The performance of the proposed approach is compared with several existing techniques. From the computational results, the proposed approach shows its ability to find quality solutions.

Note on a comparative evaluation of nine well-known algorithms for solving the cell formation problem in group technology

Over the last three decades, numerous algorithms have been proposed to solve the work-cell formation problem. For practicing manufacturing managers it would be nice to know as to which algorithm would be most effective and efficient for their specific situation. While several studies have attempted to fulfill this need, most have not resulted in any definitive recommendations and a better methodology of evaluation of cell formation algorithms is urgently needed. Prima facie, the methodology underlying Miltenburg and Zhang's (M&Z) (1991) evaluation of nine well-known cell formation algorithms seems very promising. The primary performance measure proposed by M&Z effectively captures the objectives of a good solution to a cell formation problem and is worthy of use in future studies. Unfortunately, a critical review of M&Z's methodology also reveals certain important flaws in M&Z's methodology. For example, M&Z may not have duplicated each algorithm precisely as the developer(s) of that algorithm intended. Second, M&Z's misrepresent Chandrasekharan and Rajagopalan's [C&R's] (1986) grouping efficiency measure. Third, M&Z's secondary performance measures lead them to unnecessarily ambivalent results. Fourth, several of M&Z's empirical conclusions can be theoretically deduced. It is hoped that future evaluations of cell formation algorithms will benefit from both the strengths and weaknesses of M&Z's work.

A new fuzzy-c-means and assignment technique-based cell formation algorithm to perform part- type clusters and machine-type clusters separately

The incorporation of fuzziness in the cell formation problem by Chu and Hayya (1991 International Journal of Production Research, 29, 1475-1487) is a notable contribution, in which non-binary classification logic is used. However, despite this development, numerical illustrations performed in this research demonstrate that the Chu and Hayya approach can result in solutions with empty party-type cluster(s) and/or empty machine-type cluster(s). Further, it is noted that solutions based on the Chu and Hayya approach can contain non-empty part-type cluster(s) being assigned to empty machine-type cluster(s) and vice versa. Three strategies are offered in this research to overcome these inadequacies: the separate formation of part-type and machine-type clusters; modification of the stopping criterion; and the adoption of the assignment technique in the formation of the final manufacturing cell solutions. A new algorithm has resulted from these modifications and has been rigorously compared to the performance of the Chu and Hayya approach. In general, the new algorithm demonstrates superiorities in global efficiency and in generating feasible party-type clusters, machine-type clusters and manufacturing cells.

A cell formation algorithm: Hypergraph approximation - Cut tree

A new cell formation technique is presented and analyzed in this paper. The cell formation problem is defined using the hypergraph representation of the manufacturing systems. The proposed method approximates the hypergraph model by graphs so that the cuts are less affected by the approximation. Consequently, a Gomory-Hu cut tree of the graph approximation is obtained. The minimum cuts between all pairs of vertices are calculated easily by means of this tree, and a partition tree is produced. Our cell formation algorithm successively cuts the partition tree. The algorithm is subjected to an experimentation of randomly generated manufacturing situations. The algorithm is compared with other cell formation techniques as well.

A TSP-based heuristic for forming machine groups and part families

Cellular manufacturing has been proposed as a layout approach to improve manufacturing efficiency and productivity. In implementing cellular manufacturing, parts are grouped into part families based on their similarity in manufacturing, and machines are grouped into machine cells to reduce intercellular movement of parts. To model the cellular manufacturing problem, a machinepart incidence matrix is often used. A cell formation algorithm must produce machine cells and associated part families to minimize intercellular movement of parts. In this paper, the cell formation problem is formulated as a travelling salesman problem (TSP) and a solution methodology based on genetic algorithms (GAs) is proposed to solve the TSP-cell formation problem. The proposed algorithm is compared very favourably to a well-known algorithm available in the literature.

CASE: A clustering algorithm for cell formation with sequence data

Cellular manufacturing is a well-known strategy for reducing lead times in batch production systems. Most of the methods of cell formation are based on machinecomponent incidence alone. However, other factors such as production sequence and product volumes, if incorporated, can enhance the quality of the solutions. This study uses sequence data for cell-formation. A new similarity measure is defined for this purpose and appropriate measures of performance for evaluating solutions are introduced. A clustering approach of the non-hierarchical type is proposed. With new seeding techniques, the proposed algorithm clusters machines and components on the basis of sequence data. The algorithm gives encouraging results when applied to sample problems.

Cell formation using tabu search

We present a tabu search-based cell formation algorithm in this paper. The cell formation problem studied here is to group machines into cells such that the material movement between cells is minimized. By using a weighted graph to represent the material movement between machines, we model the cell formation problem as a graph partition problem. The algorithm solves this problem through improving a cell configuration using the tabu search technique. A binary-tree data structure and a look-ahead scheme are developed for the algorithm to increase its efficiency and effectiveness. Our computational results show that this algorithm is capable of generating very good cell configurations in moderate computational time.

A neural network-based cell formation algorithm in cellular manufacturing

The Adaptive Resonance Theory (ART) neural network is a novel method for the cell formation problem in group technology (GT). The advantages of using an ART network over other conventional methods are its fast computation and the outstanding ability to handle large scale industrial problems. One weakness of this approach is that the quality of a grouping solution is highly dependent on the initial disposition of the machine-part incidence matrix especially in the presence of bottleneck machines and/or bottleneck parts. The effort of this paper has been aimed at alleviating the above mentioned problem by the introduction of a set of supplementary procedures. The advantages of the supplementary procedures are demonstrated by 40 examples from the literature. The results clearly demonstrate that our algorithm is more reliable and efficient in cases of ill-structured data.

A general search algorithm for cell formation in group technology

Abstract 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 clustering algorithm for machine cell formation in group technology using minimum spanning trees

Abstract I address the machine cell part family formation problem in group technology. The minimum spanning tree (MST) For machines is constructed from which seeds to cluster components are generated. Seeds to cluster machines are obtained from component clusters. The process of alternate seed generation and clustering is continued until feasible solutions are obtained. Edges are removed from the MST to identify alternate starting seeds for clustering. The algorithm is tested with matrices available in the literature. The results compare favourably with existing methods.