Authors: S.M. Kannan, R. Sivasubramanian, V. Jayabalan
Addresses: Department of Mechanical Engineering, Kalasalingam University, Krishnankoil – 626 190, India. ' Department of Mechanical Engineering, SACS M.A.V.M.M Engineering College, Kidaripatti – 625 301, Madurai District, South India. ' Department of Manufacturing Engineering Division, Anna University, Chennai – 600 025, South India
Abstract: A product consists of two or more components assembled together is an assembly. The quality of the product depends upon the quality of the assembly. The contributing quality characteristics of the mating parts play a major role. A good amount of research has been carried out to improve the quality of assembly using selective assembly, when the contributing quality characteristics confirms to normal distribution. However, in reality, the contributing quality characteristics of a component will have some skewness, which will make the models proposed by earlier researcher not suitable for practice. In this paper, a new method is proposed with component quality characteristics having skewness and selective assembly can be effectively used to meet the specification requirements without any surplus parts. The proposed method ensures that all the components of the mating part population is used and at the same time there is minimum variation in the assembly even there is skewness in the dimensional distribution of the mating parts. Genetic algorithm (GA) is used to find the number of components in selective group combinations for a given clearance variation.
Keywords: selective assembly; clearance variation; genetic algorithms; GAs; skewed distribution; assembly quality; component quality.
International Journal of Productivity and Quality Management, 2009 Vol.4 No.5/6, pp.569 - 589
Published online: 16 May 2009 *Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article