چكيده لاتين :
Seueral traditiounl stopping criteria in Genetic Alqoritluns (GAs) are
applied to the optimization process of a typical laminaied composite plate. The
results show that neither- the cr-iteria of the type of statistical paיmmeterיs, nOT
those of the kinds of theoreticalniod-I» perjotm satisjactoriu; in dctcrtuiuino the
intelT1lption point fOT the GA procese. HeTCo consideTLng the ronfigיllmtion of
the histor-y curve of the marimum objective 0י(י fitness nalue with geneTation
number. a Loqaritlnnic Performance Criterion (LPC) is introduced as a
stopping criterion forי optimization of composite plates. The results highlight
the advantages of LPC in perjormiru; sufficiently smooth (no noise], requirinq
reasonoble number of yenem,tions, less parameter dependency, less need [or
cotisernaiine assumption, on-line contTOliability, wide scope (/pplicabihty, and
reasonably easy appliration in enqineerinq decision-making problems.