A Learning Strategy for Disassembly Process Planning

Author

Grochowski, David E. ; Tang, Ying

Author_Institution

Rowan Univ., Glassboro

fYear

2007

fDate

15-17 April 2007

Firstpage

489

Lastpage

494

Abstract

To aid in the disassembly process of various obsolete products, expert systems can be used to model the process and provide valuable insight pertaining to the decisions made within the process. This paper discusses such a model that integrates a Disassembly Petri Net (DPN) with a Hybrid Bayesian Network (HBN). The rationale for this framework as well as its construction is presented in detail. With the incorporation of human factors and condition of disassembled units, this model proves to be more applicable to real industry setting. The suggestions for parameter learning are also discussed, allowing for the BN to give better results when many products have been disassembled.

Keywords

Petri nets; belief networks; expert systems; learning (artificial intelligence); process planning; disassembled units; disassembly Petri net; disassembly process planning; expert systems; hybrid Bayesian network; learning strategy; parameter learning; Bayesian methods; Construction industry; Equations; Expert systems; High level languages; Human factors; Machine learning; Parameter estimation; Process planning; Uncertainty;

fLanguage

English

Publisher

ieee

Conference_Titel

Networking, Sensing and Control, 2007 IEEE International Conference on

Conference_Location

London

Print_ISBN

1-4244-1076-2

Electronic_ISBN

1-4244-1076-2

Type

conf

DOI

10.1109/ICNSC.2007.372827

Filename

4239040