The development of a cellular manufacturing system for automotive parts

The development of a cellular manufacturing system for automotive parts

Computers ind. Engng Vol. 33, Nos 1-2, pp. 243-247, 1997 © 1997 Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0360-8...

262KB Sizes 2 Downloads 40 Views

Computers ind. Engng Vol. 33, Nos 1-2, pp. 243-247, 1997 © 1997 Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0360-8352197 $17.00 + 0.00

Pergamon Plh S0360-8352(97)00084-3

The Development of a Cellular Manufacturing System For Automotive Parts

Adelina Castillo, 1Hamid Seifoddini,2 Jeffrey Abell, Ph.D., 3 ~Delphi Interior and Lighting Systems, P.O. Box 9009, Warren, MI 48092-5905 2Department of Industrial & Manufacturing Engineering, University of Wisconsin-Milwaukee P.O. Box 784, Milwaukee, WI 53201 3Delphi Interior and Lighting Systems, 1401 Crooks Road, Troy, MI 48084-7155

ABSTRACT The similarity of coefficient method, simulation modeling, and computer-aided layout design are employed to develop, implement, and evaluate a cellular manufacturing system for the production of automotive parts for car seats. © 1997Publishedby ElsevierScienceLid

AREAS OF INTEREST Cellular Manufacturing, Group Technology

INTRODUCTION The survival in a highly competitive manufacturing environment of today calls for the development and implementation of very efficient production systems. Many innovative techniques have been developed to improve the efficiency and flexibility of manufacturing systems. Among these techniques, cellular manufacturing is credited with overcoming some of the major problems of traditional batch-type manufacturing includingfrequent setups, long throughput times, and excessive work-in-process inventories. Cellular manufacturing is based on the identification of part-families and formation of machine cells for the processing of the part-families(machine cell formation). The data for machine cell formation is organized in a machine-part matrix which represents the manufacturing requirements of parts in the production system. The machine-part matrix is input to machine cell formation algorithms which are used to identify the machine-component groups for the development of cellular manufacturing systems. There are many different methods of machine cell formation involving a large number of machinecomponent grouping algorithms. While machine cells can be formed with relative ease, especially, in the absence of a large number of exceptional parts, the development of a fully operational cellular manufacturing system involves many challenges including the development of plant layout of manufacturing cells, the distribution of workload among machine cells, assignment of parts to machines, and production scheduling within machine cells.

243

244

21st International Conference on Computers and Industrial Engineering

group layout, and bottleneck machines were examined to make sure that the cellular manufacturing system is performing according to the expectation. Major steps for the development of the cellular manufacturing system can be summarized as follows: • Data for the machine-part matrix was extracted from part routings, flow process charts, and layouts. • Daily volume numbers reflect a percentage of actual volumes. • Manufacturing processes of car seat components were given alphabetical codes. • Seat components were given numerical codes. • Parts and processes were organized in the machine-component chart. • Machine-component groups were formed by the single linkage (SLINK) clustering algorithm. • Intercellular materials flow was examined and machines generating them were identified as bottleneck machines. Some bottleneck machines were duplicated (Seifoddini, 1991) and some others were assigned to a separate machine cell to minimize the intercellular material flow. • Machine loads, cell capacities and manpower requirements were studied to form machine cells with maximum efficiency. • The plan for the development of cellular manufacturing system was shared with employees in the plant for their inputs and constraints. • A computer-based plant layout algorithm was used to determine the suboptimal arrangement of machines within cells and cells within the plant. To test the plan for the development of the cellular manufacturing system, extensive simulation analysis is required to maximize machine utilization, to minimize intercellular moves, to enhance the efficiency of scheduling, and to improve materials flow,

RESULTS The machine-part matrix for the manufacturing system of car seat components consists of 93 machines and 65 components (Castillo, 1996). After the application of the SLINK algorithm, seven machine-component groups are formed. These machine-component groups are further modified by machine duplication to eliminate the exceptional parts (parts having operations on bottleneck machines). The final result is a block diagonal machine-part matrix composed of 9 machinecomponent groups. A total of 21 machines were duplicated. In addition, 11 machines were assigned to a separate machine cell. One of the CNC machines which did not fit well in any of the existing machine cells was designated as an independent cell (Castillo, 1996). When the machine cells were formed and part-families were identified, the routings of parts were determined and along the production volume for each part are summarized in a matrix which will be used for the scheduling purpose.

CONCLUSIONS In this paper, a case study about the development of a cellular manufacturing system for car seat components was reviewed. Major steps in the development process including the preparation of the initial machine-part matrix, the formation of machine-component groups, the evaluation of machine cells for satisfying the real world constraints, and the analysis of material handling for the purpose of layout design were briefly discussed.

21st International Conference on Computers and Industrial Engineering

245

In this paper, the results of a case study involving the development and implementation of a cellular manufacturing system for the production of car seat components will be presented. The similarity coefficient method is employed to form the machine component groups. Major challenges in the transformation of a traditional job shop manufacturing system to a cellular manufacturing system in automotive industry will be discussed. This paper provides more insight into the practical aspect of the development and implementation of cellular manufacturing systems for the production of car seat components. In this study, the design of operational cells subject to real world constraints such as the nature of machines, layout of the plant, employee satisfaction, workload balancing, and so on are examined and the effectiveness of the similarity coefficient method in forming machine cells for a real world application is tested. The case study will serve as an example of the application of group technology to batch-type production in the specific case of car seat components in automotive industries.

DEFINITION OF THE PROBLEM

The introduction of an existing seat component into a different plant provided the opportunity for the evaluation of the manufacturing system for the implementation of cellular manufacturing. The diversity of components in car seats and the variation of manufacturing operations make a strong case for the application of cellular manufacturing. The first step in the planning and implementation of cellular manufacturing is the identification of part-families and formation of machine cells. Well defined part-families 1 with independent machine cells represent a manufacturing situation with a great potential for the implementation of cellular manufacturing (Burbidge, 1992; Gettleman, 1971). The data for the identification of part-families and formation of machine cells (machine-component grouping) are extracted from routings and are organized in a machine-part matrix which represents the manufacturing requirements of different parts. The structure of the machine-part matrix, to a great extent, determines the effectiveness of the corresponding cellular manufacturing system. There are many efficiency measures that can be used to evaluate the suitability of a particular manufacturing system for conversion to a cellular manufacturing system (Seifoddini and Djassemi, 1995). Though efficiency measures are helpful in identifying manufacturing situations with high potential or with low potential for the application of cellular manufacturing, generally an extensive evaluation of a cellular manufacturing system is necessary prior to commitment to the development of such a system. In this case study, the machine-part matrix for the manufacturing operations of car seats components is developed based on careful examination of the routings of the car seats components. Then a clustering algorithm is employed to form the machine components group. Finally, the plant layout is evaluated, the machine loads are examined, and material handling is studied to develop an efficient cellular manufacturing system subject to the operational, financial, and physical constraints.

MAJOR STEPS IN THE DEVELOPMENT OF CM SYSTEM

To determine the feasibility of developing a cellular manufacturing system for the processing of car seats components at Delphi Interior and Lighting Systems, relevant data on parts, manufacturing processes and material flow was collected and organized in a machine-part matrix. Then a clustering algorithm based on the similarity coefficient method (Seifoddini and Wolfe, 1986) was used to form machine-component groups. Finally, machine cells, part families, materials flow,

21st International Conference on Computers and Industrial Engineering

246

A sample of machine cells and part routings for the manufacturing system is as follows.

Table 1. Machine Cell No. 1

Equipment CELL 1 3 Air Sonder 4 Rivetor 4 Rivetor 4 Manual 4 Manual 2 Rivetor 2 Rivetor 1 Rivetor 4 Rivetor 1 Rivetor

Code

Vol. Mach.

A B C D E F G I H J

2700 2700 2700 2700 2700 1800 1200 600 300 300

Total Time (Hr.) Per Equip.

Man Req. (Shift)

8 8 8 8 8 8 7 7 1 4

3 4 4 4 4 3 2 1 4 1

Total Time(Hr.) Per Equip.

Man Req. (Shift)

Table 2. Machine Cell No. 2

Equipment

Code

1 Rivetor 1 Rivetor 1 Air Gun GUN1 Air Gun 1 Staker/Air 1 Manual 1 Air Gun 1 Air Scrw 1 Air Gun 1 Manual 1 Air Gun 1 Manual 1 Welder 1 Manual 1 Staker/Air 1 Air Scrw 1 Rivetor 1 Rivetor 1 Rivetor

AD mE AF AG AH AQ A1 AJ AL AM AN AO AP O AC AK Z AA All

Vol. Mach. 600 600 600 600 600 600 600 600 600 600 600 600 600 600 300 300 600 600 600

8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

21st International Conference on Computers and Industrial Engineering

247

Seifoddini, H. and Wolfe, P., (1986) "Application of The Similarity Coefficient Method in Group Technology," Institute of Industrial Engineers Transactions, 271-279. Seifoddini, H. and Djassemi, M., (1995) "Selection of Machine-Component Charts for Cellular Manufacturing Based on Quality Index," Proceedings of 4th Industrial Engineering Research

Conference. Seifoddini, H., (1991) "Duplication Process in Machine Cells Formation in Group Technology," Institute of Industrial Engineers Transactions. Tompkins, J. A. and White, J. A., (1984)"Computer-Aided Layout Facilities Planning," John Wiley & Sons, Inc..

CAIE 33:1/2-J