• 대한산업공학회지
• /
• 제30권3호
• /
• pp.175-180
• /
• 2004

The machine-part group formation is to group the sets of parts having similar processing requirements into part families, and the sets of machines needed to process a particular part family into machine cells using grid computing. It forms machine cells from the machine-part incidence matrix by means of Self-Organizing Maps(SOM) whose output layer is one-dimension and the number of output nodes is the twice as many as the number of input nodes in order to spread out the machine vectors. It generates machine-part group which are assigned to machine cells by means of the number of bottleneck machine with processing part. The proposed algorithm was tested on well-known machine-part grouping problems. The results of this computational study demonstrate the superiority of the proposed algorithm.

• 대한산업공학회지
• /
• 제27권3호
• /
• pp.281-288
• /
• 2001

We consider in this article a multi-objective part-machine grouping problem in which parts have alternative process plans and expected annual demand of each part is known. This problem is characterized as optimally determining part sets and corresponding machine cells such that total sum of distance (or dissimilarity) between parts and total sum of load differences between machines are simultaneously minimized. Two heuristic algorithms are proposed, and examples are given to compare the performance of the algorithms.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2006년도 춘계공동학술대회 논문집
• /
• pp.354-358
• /
• 2006

This study is concerned with the part-machine grouping(PMG) based on the non-binary part-machine incidence matrix in which real manufacturing Factors such as the operation sequences with multiple visits to the same machine and production volumes of parts are incorporated and each entry represents actual moves due to different operation sequences. The proposed approach adopts Fuzzy ART neural network to quickly create the initial part families and their associated machine cells. To enhance the poor solution due to category proliferation inherent to most artificial neural networks, a supplementary procedure reassigning parts and machines is added. To show effectiveness of the proposed approach to large-size PMG problems, a psuedo-replicated clustering procedure is designed and implemented.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2006년도 추계학술대회
• /
• pp.656-658
• /
• 2006

This study deals with the part-machine grouping (PMG) that considers realistic manufacturing factors, such as the machine duplication, operation sequences with multiple visits to the same machine, and production volumes of parts. Basically, this study is an extension of Won(2006) that has adopted fuzzy ART neural network to group parts and machines. The proposed fuzzy ART neural network algorithm is implemented with an ancillary procedure to enhance the block diagonal solution by rearranging the order of input presentation. Computational experiments applied to large-size PMG data sets with a psuedo-replicated clustering procedure show effectiveness of the proposed approach.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회/대한산업공학회 2003년도 춘계공동학술대회
• /
• pp.375-380
• /
• 2003

This study is concerned with the machine part grouping m cellular manufacturing. To group machines into the set of machine cells and parts into the set of part families, new p-median model considering the production data such as the operation sequences and production volumes for parts is proposed. Unlike existing p-median models relying on the classical binary part-machine incidence matrix which does not reflect the real production factors which seriously impact on machine-part grouping, the proposed p-median model reflects the production factors by adopting the new similarity coefficient based on the production data-based part-machine incidence matrix of which each non-binary entry indicates actual intra-cell or inter-cell flows to or from machines by parts. Computation test compares the proposed p median model favorably.

• 대한산업공학회지
• /
• 제17권1호
• /
• pp.75-82
• /
• 1991

This research is concerned with Machine-Part Group Formation(MPGF) methodology for Flexible Manufacturing Systems(FMS). The purpose of the research is to develop a new heuristic algorithm for effectively solving MPGF problem. The new algorithm is proposed and evaluated by 100 machine-part incidence matrices generated. The performance measures are (1) grouping ability of mutually exclusive block-diagonal form. (2) number of unit group and exceptional elements, and (3) grouping time. The new heuristic algorithm has the following characteristics to effectively conduct MPGF : (a) The mathematical model is presented for rapid forming the proper number of unit groups and grouping mutually exclusive block-diagonal form, (b) The simple and effective mathematical analysis method of Rank Order Clustering(ROC) algorithm is applied to minimize intra-group journeys in each group and exceptional elements in the whole group. The results are compared with those from Expert System(ES) algorithm and ROC algorithm. The results show that the new algorithm always gives the group of mutually exclusive block-diagonal form and better results(85%) than ES algorithm and ROC algorithm.

• 산업경영시스템학회지
• /
• 제21권48호
• /
• pp.123-132
• /
• 1998

The group formation problem of the machine and part is a very important issue in the planning stage of cellular manufacturing systems. This paper investigates Self-Organizing Map(SOM) neural networks approach to machine-part grouping problem. We present a two-phase algorithm based on SOM for grouping parts and machines. SOM can learn from complex, multi-dimensional data and transform them into visually decipherable clusters. Output layer in SOM network is one-dimensional structure and the number of output node has been increased sufficiently to spread out the input vectors in the order of similarity. The proposed algorithm performs remarkably well in comparison with many other algorithms for the well-known problems shown in previous papers.

• 대한산업공학회지
• /
• 제31권1호
• /
• pp.1-9
• /
• 2005

This paper proposes the heuristic algorithm for the generalized GT problems to consider the restrictions which are given the number of machine cell and maximum number of machines in machine cell as well as minimum number of machines in machine cell. This approach is split into two phase. In the first phase, we use the similarity coefficient which proposes and calculates the similarity values about each pair of all machines and sort these values descending order. If we have a machine pair which has the largest similarity coefficient and adheres strictly to the constraint about birds of a different feather (BODF) in a machine cell, then we assign the machine to the machine cell. In the second phase, we assign parts into machine cell with the smallest number of exceptional elements. The results give a machine-part grouping. The proposed algorithm is compared to the Modified p-median model for machine-part grouping.

• 산업공학
• /
• 제12권4호
• /
• pp.532-539
• /
• 1999

The first step to design a cellular manufacturing system is to make part-families and machine-cells. This process is called the machine-part grouping. This paper considers a machine-cell size when grouping machine-cells. By considering a machine-cell size, an unrealistically big size of machine-cell which may be caused by the chaining effect can be avoid. A heuristic algorithm which minimizes the number of inter-cell movement of parts considering a machine-cell size is presented. The effectiveness and performance of the proposed heuristic algorithm are compared with those of several heuristic algorithms previously reported. The comparison shows that the proposed heuristic algorithm is efficient and reliable in minimizing the number of inter-cell movement of parts and also prevents the chaining effect.

• 대한산업공학회지
• /
• 제18권2호
• /
• pp.123-130
• /
• 1992

The group formation problem of the machine and part in Flexible Manufacturing System (FMS) is a very important issue in planning stage of FMS. This paper discusses the problem of machine-part group formation. The purpose of the study is to develop a heuristic algorithm, which can handle more realistic machine-part group formation problem by considering manufacturing factors. A new similarity coeffecient has been developed to solve more realistic machine-part group formation problem. For the purpose of illustations, a numerical example is presented.