, Volume 17, Issue 3, pp 313–317 | Cite as

Computerized statistical analysis of melting and pouring variables on the production of high quality steel castings

  • J. K. Sprinkle
  • M. H. Davison
  • J. Keverian
22nd Electric Furnace Conference


Statistical analysis of production data relating mechanical properties to process variables resulted in the separation of variables into three major classes:
  1. 1)

    Important, but not statistically-significant variables. These variables have sufficiently tight control in production to guarantee a small deviation in mechanical properties. For example, percent carbon and heat treatment factors are known to be important but were not statistically significant in this study.

  2. 2)

    Important and statistically significant variables. These variables will be brought under control as required. For example, rate of rise of molten metal in the mold was an unsuspected variable that was statistically significant in this study. This variable is believed to be important because it influences the as-cast grain size.

  3. 3)

    Not important and not statistically significant variables. Some variables had quite a wide range and yet were not effective in changing mechanical properties. No new measures of control are required for this vast host of variables



Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    K. M. Smith: Foundry, Aug. 1964.Google Scholar
  2. 2.
    M. H. Davison, J. K. Sprinkle, and J. Keverian: Mod. Castings, Sept. 1964.Google Scholar
  3. 3.
    M. H. Davison, J. K. Sprinkle, and J. Keverian: The Application of a Digital Computer for Identifying Critical Process Variables in Casting Production, to be published.Google Scholar
  4. 4.
    Staff Article: Computer Solves Casting Mysteries, Iron Age, Nov. 5, 1964.Google Scholar
  5. 5.
    J. K. Sprinkle: DF63-SDF-10, May 16, 1963.Google Scholar
  6. 6.
    M. J. Moroney: Facts From Figures, Penguin Books.Google Scholar
  7. 7.
    P. G. Hoel: Introduction to Mathematical Statistics, Wiley.Google Scholar
  8. 8.
    W. J. Dixon and F. J. Massey, Jr.: Introduction to Statistical Analysis, McGraw-Hill.Google Scholar
  9. 9.
    W. Volk: Applied Statistics for Engineers, McGraw-Hill.Google Scholar
  10. 10.
    G. W. Snedecor: Statistical Methods, Iowa State College Press.Google Scholar
  11. 11.
    M. L. Conneally and K. E. Fritz: Application of Multiple Regression Analysis to Metallurgical Data, presented at 1963 ASM Metals Congress, Cleveland, Ohio.Google Scholar
  12. 12.
    J. N. Lynch, G. F. Kent, M. P. Berardi, F. J. Mantegani, and J. Zotos: Mod. Castings, Aug. 1963.Google Scholar
  13. 13.
    P. A. Cooper, G. F. Kent, and J. Zotos: Development of Improved Mathematical Models for Prediction of Steel Castings Mechanical Properties, presented at 1964 AFS Castings Congress, Atlantic City, N. J.Google Scholar

Copyright information

© The Minerals, Metals & Materials Society 1965

Authors and Affiliations

  • J. K. Sprinkle
    • 1
  • M. H. Davison
    • 1
  • J. Keverian
    • 1
  1. 1.Foundry dept.General Electric Co.SchenectadyUSA

Personalised recommendations