Comparison of the results of factorial experiments, fractional factorial experiments, regression trees and MARS for fuel consumption data

Kan B., YAZICI B.

WSEAS Transactions on Mathematics, vol.9, no.2, pp.110-119, 2010 (Scopus) identifier

  • Publication Type: Article / Article
  • Volume: 9 Issue: 2
  • Publication Date: 2010
  • Journal Name: WSEAS Transactions on Mathematics
  • Journal Indexes: Scopus
  • Page Numbers: pp.110-119
  • Keywords: Aid (automatic interaction detection), Basis function, Factorial experiments, Fractional factorial experiments, GCV (generalized cross validation), MARS, Recursive partitioning, Regression trees
  • Anadolu University Affiliated: Yes

Abstract

In this study effects that are significant on the fuel consumption of F-4 aircrafts are analyzed by using different statistical methods and the results of those different methods are compared. The response variable of every possible combination of factors and interactions are analyzed to provide information about main effect(s) and interaction effect(s). MARS is applied in order to find the relationship and a tree model in order to summarize the relationship of main effects and the interactions on a plot. Also a 2 5 factorial experiment with 4 replications is performed to determine the significant main effects and interactions. The analysis is repeated for one-half fractional factorial experiment. A comparison of performance of the methods is made in order to investigate their applicability.