Date of Award

2011

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Industrial and Manufacturing Systems Engineering

First Advisor

Wang, Hunglin (Industrial and Manufacturing Systems Engineering)

Keywords

Engineering, Industrial.

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

Although there exists several ways of solving the cellular manufacturing problem, including several ant-based algorithms, many of these algorithms focus on obtaining the best possible answer instead of efficiency. An existing artificial-ant based algorithm AntClass, was modified so that it is easier to manipulate. AntClass uses Euclidean vectors to measure the similarity between parts, because similarity is used to group parts together instead of distances, the modified version uses similarity coefficients. The concept of heaping clusters was also introduced to ant algorithms for cellular manufacturing. Instead of using Euclidean vectors to measure the distance to the center of a heap, as in the AntClass algorithm, an average similarity was introduced to measure the similarity between a part and a heap. The algorithm was tested on five common similarity coefficients to determine the similarity coefficient which gives the better quality solution and the most efficient process.

Share

COinS