Get access

A generalized linear model approach to designing accelerated life test experiments

Authors

  • Eric M. Monroe,

    Corresponding author
    1. Intel Corporation, 5000 W. Chandler Blvd, Mailstop CH2-117, Chandler, AZ 85226, U.S.A.
    • Intel Corporation, Chandler, AZ 85226, U.S.A.
    Search for more papers by this author
    • Six Sigma Blackbelt at Intel Corporation.

  • Rong Pan,

    1. Arizona State University, Tempe, AZ 85287, U.S.A.
    Search for more papers by this author
    • Associate Professor of Industrial Engineering at Arizona State University.

  • Christine M. Anderson-Cook,

    1. Los Alamos National Laboratory, Los Alamos, New Mexico 87545, U.S.A.
    Search for more papers by this author
    • Research Scientist in the Statistical Sciences Group at Los Alamos National Laboratory.

  • Douglas C. Montgomery,

    1. Arizona State University, Tempe, AZ 85287, U.S.A.
    Search for more papers by this author
    • Regent's Professor of Industrial Engineering at Arizona State University.

  • Connie M. Borror

    1. Arizona State University, Glendale, AZ 85306, U.S.A.
    Search for more papers by this author
    • Professor in the Division of Mathematical and Natural Sciences at Arizona State University.


Abstract

Optimal experimental design practices are prominent in many applications. This paper proposes an alternate way of computing the information matrix, a key consideration in planning an accelerated life test. The generalized linear model approach allows optimal designs to be computed using iteratively weighted least-square solutions versus a maximum likelihood method. This approach is demonstrated with an assumed exponential distribution and allows the practitioner to observe the underlying structure of the optimal experimental design matrix and its relationship to important factors such as censoring and a nonlinear response function. Optimality criteria are discussed for both parameter estimation and prediction variance at an intended usage condition, which is typically outside the feasible accelerated test region. Copyright © 2010 John Wiley & Sons, Ltd.

Ancillary