The Study of Neural Networks for Product End-of-Life Strategies and Design Improvement Guidelines for Product Reuse

碩士 === 國立成功大學 === 機械工程學系碩博士班 === 90 === Abstract Concerning the limits of natural resource and the effect on environmental impact, products had better consider the whole life cycle from the design, manufacture, sale, use and recycle in order to reduce the cost of products and reduce impact in the en...

Full description

Bibliographic Details
Main Authors: Chun-Nan Wu, 吳俊男
Other Authors: Jahau-Lewis Chen
Format: Others
Language:zh-TW
Published: 2002
Online Access:http://ndltd.ncl.edu.tw/handle/et56gz
id ndltd-TW-090NCKU5490045
record_format oai_dc
spelling ndltd-TW-090NCKU54900452018-06-25T06:05:42Z http://ndltd.ncl.edu.tw/handle/et56gz The Study of Neural Networks for Product End-of-Life Strategies and Design Improvement Guidelines for Product Reuse 類神經網路於產品壽命終了時之處理策略分析及產品再使用設計改善之研究 Chun-Nan Wu 吳俊男 碩士 國立成功大學 機械工程學系碩博士班 90 Abstract Concerning the limits of natural resource and the effect on environmental impact, products had better consider the whole life cycle from the design, manufacture, sale, use and recycle in order to reduce the cost of products and reduce impact in the environment. In all kinds of ways, the remanufacture of the products is the most practicable strategy to arrive above goals. The main structure of this thesis includes two parts. First, it analyzes the end of life strategies that includes the reuse, service, remanufacture, recycle with disassembly, recycle without disassembly and disposal of products by the back propagation neural network. Another part is planning sub-modular of products with the way of product modularity to enhance the remanufactureability. Neural network has the advantages of easily to use and feasible to non-linear problems with learning capability. The theory of End-of-Life Design Advisor (ELDA) is selected as the basic structure of back-propagation neural network to determine the useful strategy. Furthermore, self organize map neural network was selected to analyze the relation between each strategies. Hence, the trained neural networks can simulate the analysis mode of ELDA rapidly and offer the designer with an easy operation method in the relative research domain. Product modularity plays a quite important role in designing procedure, especially when the products have to be up-graded. General speaking, product modularity is divided from the functions. In order to enhance the possibilities of remanufacture and decrease the quantity of parts, we divide the sub-modular of products based on the point of view of material, life and up-grade. Finally, this thesis provides the designer the design principles and suggestions in product remanufacture design tasks. Jahau-Lewis Chen 陳家豪 2002 學位論文 ; thesis 103 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立成功大學 === 機械工程學系碩博士班 === 90 === Abstract Concerning the limits of natural resource and the effect on environmental impact, products had better consider the whole life cycle from the design, manufacture, sale, use and recycle in order to reduce the cost of products and reduce impact in the environment. In all kinds of ways, the remanufacture of the products is the most practicable strategy to arrive above goals. The main structure of this thesis includes two parts. First, it analyzes the end of life strategies that includes the reuse, service, remanufacture, recycle with disassembly, recycle without disassembly and disposal of products by the back propagation neural network. Another part is planning sub-modular of products with the way of product modularity to enhance the remanufactureability. Neural network has the advantages of easily to use and feasible to non-linear problems with learning capability. The theory of End-of-Life Design Advisor (ELDA) is selected as the basic structure of back-propagation neural network to determine the useful strategy. Furthermore, self organize map neural network was selected to analyze the relation between each strategies. Hence, the trained neural networks can simulate the analysis mode of ELDA rapidly and offer the designer with an easy operation method in the relative research domain. Product modularity plays a quite important role in designing procedure, especially when the products have to be up-graded. General speaking, product modularity is divided from the functions. In order to enhance the possibilities of remanufacture and decrease the quantity of parts, we divide the sub-modular of products based on the point of view of material, life and up-grade. Finally, this thesis provides the designer the design principles and suggestions in product remanufacture design tasks.
author2 Jahau-Lewis Chen
author_facet Jahau-Lewis Chen
Chun-Nan Wu
吳俊男
author Chun-Nan Wu
吳俊男
spellingShingle Chun-Nan Wu
吳俊男
The Study of Neural Networks for Product End-of-Life Strategies and Design Improvement Guidelines for Product Reuse
author_sort Chun-Nan Wu
title The Study of Neural Networks for Product End-of-Life Strategies and Design Improvement Guidelines for Product Reuse
title_short The Study of Neural Networks for Product End-of-Life Strategies and Design Improvement Guidelines for Product Reuse
title_full The Study of Neural Networks for Product End-of-Life Strategies and Design Improvement Guidelines for Product Reuse
title_fullStr The Study of Neural Networks for Product End-of-Life Strategies and Design Improvement Guidelines for Product Reuse
title_full_unstemmed The Study of Neural Networks for Product End-of-Life Strategies and Design Improvement Guidelines for Product Reuse
title_sort study of neural networks for product end-of-life strategies and design improvement guidelines for product reuse
publishDate 2002
url http://ndltd.ncl.edu.tw/handle/et56gz
work_keys_str_mv AT chunnanwu thestudyofneuralnetworksforproductendoflifestrategiesanddesignimprovementguidelinesforproductreuse
AT wújùnnán thestudyofneuralnetworksforproductendoflifestrategiesanddesignimprovementguidelinesforproductreuse
AT chunnanwu lèishénjīngwǎnglùyúchǎnpǐnshòumìngzhōngleshízhīchùlǐcèlüèfēnxījíchǎnpǐnzàishǐyòngshèjìgǎishànzhīyánjiū
AT wújùnnán lèishénjīngwǎnglùyúchǎnpǐnshòumìngzhōngleshízhīchùlǐcèlüèfēnxījíchǎnpǐnzàishǐyòngshèjìgǎishànzhīyánjiū
AT chunnanwu studyofneuralnetworksforproductendoflifestrategiesanddesignimprovementguidelinesforproductreuse
AT wújùnnán studyofneuralnetworksforproductendoflifestrategiesanddesignimprovementguidelinesforproductreuse
_version_ 1718705093004492800