A hybrid of Particle Swarm Optimization and Population Based Incremental Learning in Cell Formation
碩士 === 國立臺灣科技大學 === 工業管理系 === 102 === The development of meta-heuristics has received great interest in recent years. Several methods have been implemented in cell formation. In this thesis book, the problem of grouping parts into families; and machines into cells are considered with the objective o...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2014
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/uwtsjt |
| Summary: | 碩士 === 國立臺灣科技大學 === 工業管理系 === 102 === The development of meta-heuristics has received great interest in recent years. Several methods have been implemented in cell formation. In this thesis book, the problem of grouping parts into families; and machines into cells are considered with the objective of maximizing the grouping efficacy. A hybrid of particle swarm optimization and population-based incremental learning (PBILSO) is proposed and used for solving this problem. The proposed algorithm uses a modification of permutation with separator encoding scheme and similarity measure is used to evaluate similarity between parts as an index for assigning parts to cells. A set of 10 test problems with various sizes from literatures are used to test the performance of both algorithms. The Taguchi method is used for determining the initialization parameters. Numerical experiments have proved that the proposed method can increase the grouping efficacy for 60% cases and 10% near optimal than other algorithms in several references. It is also proved that the hybrid can generate more convergence result so that the number of hits increasing and able to group the machine and part into cells simultaneously.
|
|---|