Data mining algorithm for manufacturing process control |
| |
Authors: | Hovhannes Sadoyan Armen Zakarian Pravansu Mohanty |
| |
Affiliation: | (1) Department of Industrial and Manufacturing Systems Engineering, University of Michigan – Dearborn, USA;(2) Department of Mechanical Engineering, University of Michigan – Dearborn, Dearborn, MI, 48128, U.S.A. |
| |
Abstract: | In this paper, a new data mining algorithm based on the rough sets theory is presented for manufacturing process control. The algorithm extracts useful knowledge from large data sets obtained from manufacturing processes and represents this knowledge using “if/then” decision rules. Application of the data mining algorithm developed in this paper is illustrated with an industrial example of rapid tool making (RTM). RTM is a technology that adopts rapid prototyping (RP) techniques, such as spray forming, and applies them to tool and die making. A detailed discussion on how to control the output of the manufacturing process using the results obtained from the data mining algorithm is also presented. Compared to other data mining methods, such decision trees and neural networks, the advantage of the proposed approach is its accuracy, computational efficiency, and ease of use. |
| |
Keywords: | Data mining Decision rules Manufacturing process control Rough sets |
本文献已被 SpringerLink 等数据库收录! |
|