Random Initial Imperfection (II)

  • Kiyohiro Ikeda
  • Kazuo Murota
Part of the Applied Mathematical Sciences book series (AMS, volume 149)


Cylindrical shells may be quoted as one of the most famous D n -symmetric (or, to be precise, D-symmetric) systems subject to strength variation due to initial imperfections. The initial imperfection data bank has been developed by measuring the initial imperfections of shells (Singer, Abramovich, and Yaffe, 1978 [168]; Arbocz and Abramovich, 1979 [4]). The effect of random axisymmetric imperfections on the buckling of circular cylindrical shells under axial compression was investigated by Elishakoff and Arbocz, 1982 [45]. The statistical properties of the shells were evaluated based on the measured data, and the reliability function of the shells was computed by means of the Monte Carlo method. Later on, on the basis of an assumption that the initial imperfections are represented by normally distributed random variables, the first-order second-moment method was employed to replace the Monte Carlo method and, in turn, to greatly reduce computational costs (e.g., Elishakoff et al., 1987 [49]; Arbocz and Hol, 1991 [5]).


Probability Density Function Critical Load Multivariate Normal Distribution Reliability Function Initial Imperfection 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Science+Business Media New York 2002

Authors and Affiliations

  • Kiyohiro Ikeda
    • 1
  • Kazuo Murota
    • 2
  1. 1.Department of Civil EngineeringTohoku UniversityAoba SendaiJapan
  2. 2.Department of Mathematical InformaticsUniversity of TokyoTokyoJapan

Personalised recommendations