Group Technology And Cellular Manufacturing

Methodologies and Applications

Rasaratnam Logendran, Ali Κ. Kamrani

• Publisher: Routledge
• Year: 2020
• eBook ISBN: 9781000150865
• Topic: Social Sciences
• Subtopic: Sociology
• Index: Social Sciences

This book provides the latest up-to-date documentation on the scope of research in Group Technology (GT) and Cellular Manufacturing (CM). It is a comprehensive listing of the methodologies, techniques, algorithms and tools used for practical implementation of the concepts of GT and CM.

I DESIGN, MODELING AND ANALYSIS

CHAPTER 1

THE GROUP TECHNOLOGY PROBLEM WITH BOTTLENECK PARTS

David W. He and Andrew Kusiak

1. INTRODUCTION

This chapter deals with the group technology problem with bottleneck parts. To achieve full benefits of cellular manufacturing, it is desired that machines are grouped into independent cells. However, due to the presence of bottleneck parts, totally independent machine cells cannot be generated. One solution to the problem is to modify process plans of the bottleneck parts or redesign the bottleneck parts so that they can be processed in the existing machine cells. It is assumed that parts are processed in lots in the machine cells and the set up and inventory costs are of a major concern. The group technology problem with bottleneck parts is to assign the bottleneck parts to the machine cells so that the total average setup and inventory cost is minimized. A mathematical programming formulation of the problem is presented. Heuristic algorithms for solving the problem are proposed and their performance is tested. In cellular manufacturing, machines are grouped into cells and parts are grouped into families. It is desired that each part be processed in one machine cell only. The flow of parts among different cells can be eliminated, and as a consequence the material handling cost is reduced. In addition, since parts with similar processing requirements are processed in a cell, the setup time, work-in-process inventory, flow time, and production lead time are reduced. The group technology problem can be represented with a binary machine-part incidence matrix [a_ij], where an entry 1 (0) indicates that machine i is used (not used) to process part j. Typically, when an initial machine-part incidence matrix [a_ij] is constructed, clusters of machines and parts are not visible. The machine-part incidence matrix has to be transformed into a more structured (possibly block diagonal) form.

Clustering a binary incidence matrix may result in the following two categories of clusters (see Figures 1 and 2):

(1) Mutually separable clusters

(2) Partially separable clusters

FIGURE 1. Mutually separable clusters

FIGURE 2. Partially separable clusters

The mutually separable clusters represent totally independent machine cells and the partially separable clusters correspond to partially independent machine cells. The partially separable clusters are due to bottleneck parts, i.e., parts that cannot be included in one part family only. To deal with a bottleneck part, one of the following actions can be taken (Kusiak¹):

AC1. It can be machined in one machine cell and transferred to the other machine cell by a material handling carrier.

AC2. It can be machined in a functional facility

AC3. It can be subcontracted

AC4. Process plan of the bottleneck part might be modified, so that the bottleneck part could be machined in one of the existing machine cells

AC5. Design of the bottleneck part might be modified, so that the resulting process plans could fit the existing machine cell.

Action AC1 is applicable only when the flow of parts among different machine cells is relatively low. Besides the material handling costs, sharing bottleneck parts among different machine cells can cause the following difficulties:

(1) Complex scheduling problems for parts and routing problems for AGVs

(2) Managing material flow

Action AC2 can be expensive due to the necessity of using additional manufacturing facilities. The cost associated with using the additional functional facility to process the bottleneck parts is discussed later in this chapter.

Vannelli and Kumar² solved the machine grouping problem to minimize the number of bottleneck machine cells. A machine cell is considered as bottleneck if it shares more t...

Table of contents

• Cover
• Half Title
• Volume
• Title Page
• Copyright Page
• Dedication Page
• Table of Contents
• Series Preface
• Preface
• Acknowledgments
• List of Contributors
• I DESIGN, MODELING AND ANALYSIS
• II ARTIFICIAL INTELLIGENCE, NEURAL NETWORKS, GENETIC ALGORITHMS AND COMPUTER TOOLS
• Index