Optimal sensor distribution in multi-station assembly processes for maximal variance detection capability Academic Article

abstract

• Recent advances in sensor technology now allow manufacturers to distribute multiple sensors in multi-station assembly processes. A distributed sensor system enables the continual monitoring of manufactured products and greatly facilitates the determination of the underlying process variation sources that cause product quality defects. This paper addresses the problem of optimally distributing sensors in a multi-station assembly process to achieve a maximal variance detection capability. A sensitivity index is proposed for characterizing the detection ability of process variance components and the optimization problem for sensor distribution is formulated for a multi-station assembly process. A data-mining-guided evolutionary method is devised to solve this non-linear optimization problem. The data-mining-guided method demonstrates a considerable improvement compared to the existing alternatives. Guidance on practical issues such as the interpretation of the rules generated by the data mining method and how many sensors are required are also provided.

authors

• Ding, Yu

published proceedings

• IIE TRANSACTIONS

author list (cited authors)

• Ren, Y., & Ding, Y. u.

citation count

• 6

complete list of authors

• Ren, Yuan||Ding, Yu

publication date

• January 2009

publisher

• Taylor & Francis  Publisher

published in

• IIE Transactions  Journal

keywords

• Data Mining
• Evolutionary Algorithms
• Multi-station Assembly Process
• Sensitivity

Digital Object Identifier (DOI)

• 10.1080/07408170902789050

start page

• 804

end page

• 818

volume

• 41

issue

• 9