An application of PCA to rank problem parts

  • Authors:
  • K. M. George

  • Affiliations:
  • Oklahoma State University, Stillwater, OK

  • Venue:
  • Proceedings of the Second International Conference on Computational Science, Engineering and Information Technology
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

A model is developed and tested by implementing a program to identify problem parts. The model computes sustainability ranking number for parts of a system such as an aircraft. The ranking is defined as a function on a set of critical variables spanning a multidimensional space. Test values of the variables are extracted from different databases. Principal Component Analysis is used to reduce the data dimension. The ranking is obtained as a linear combination of the original variables. The sustainability ranking computed by the program is useful as a forecasting tool and critical part identification for part managers as a component of a decision support system.