Julkisen oikeuttamisen analyysi sosiologisena tutkimusmenetelmänä

Artikkelissa esitellään uusi julkisen keskustelun tutkimiseen kehitetty analyysimenetelmä, julkisen oikeuttamisen analyysi (JOA). JOA perustuu Luc Boltanskin ja Laurent Thévenot'n oikeuttamisteorian seitsemän "maailman" – inspiraation, kodin, maineen, kansalaisuuden, markkinoiden, teollisuuden ja ekologian – muodostamalle analyysikehikolle. Se tutkii keskusteluissa esiintyvien vaateiden moraalisia oikeutuksia, niiden yhdistelmiä ja tapoja kiistää ja tuomita kiistakumppaneiden oikeutuksia. JOA:n avulla voidaan kvalitatiiviseen tekstianalyysiin yhdistää myös kvantitatiivista luokittelua, jolloin menetelmä soveltuu suurtenkin aineistojen analyysiin. JOA:n käyttöä havainnollistetaan artikkelissa kahden tutkimusesimerkin avulla. Ensimmäinen esimerkki käsittelee Helsingin Sanomissa globalisaatiosta vuosina 1999–2005 käydyn keskustelun osapuolten, erityisesti kansalaisyhteiskunnan sekä taloudellisen ja poliittisen eliitin, argumentteja ja niille annettuja oikeutuksia. Tämän esimerkin kautta kuvataan erilaisten oikeuttamisyhdistelmien ilmaisuvoimaa yhteiskunnallisten kiistakysymysten moraalisten ulottuvuuksien analysoimisessa. Toinen esimerkki keskittyy paikallisiin kiistoihin Suomessa ja Ranskassa tarkastelemalla kansalaisten ja kaupungin edustajien esittämiä oikeutuksia paikallislehdistössä. Tämä esimerkki osoittaa JOA:n vahvuudet vertailevan tutkimuksen työkaluna.

Bayesian Accelerated Life Testing under Competing Weibull Causes of Failure

Consider a J-component series system which is put on Accelerated Life Test (ALT) involving K stress variables. First, a general formulation of ALT is provided for log-location-scale family of distributions. A general stress translation function of location parameter of the component log-lifetime distribution is proposed which can accommodate standard ones like Arrhenius, power-rule, log-linear model, etc., as special cases. Later, the component lives are assumed to be independent Weibull random variables with a common shape parameter. A full Bayesian methodology is then developed by letting only the scale parameters of the Weibull component lives depend on the stress variables through the general stress translation function. Priors on all the parameters, namely the stress coefficients and the Weibull shape parameter, are assumed to be log-concave and independent of each other. This assumption is to facilitate Gibbs sampling from the joint posterior. The samples thus generated from the joint posterior is then used to obtain the Bayesian point and interval estimates of the system reliability at usage condition.

Bayesian Accelerated Life Testing under Competing Weibull Causes of Failure

Consider a J-component series system which is put on Accelerated Life Test (ALT) involving K stress variables. First, a general formulation of ALT is provided for log-location-scale family of distributions. A general stress translation function of location parameter of the component log-lifetime distribution is proposed which can accommodate standard ones like Arrhenius, power-rule, log-linear model, etc., as special cases. Later, the component lives are assumed to be independent Weibull random variables with a common shape parameter. A full Bayesian methodology is then developed by letting only the scale parameters of the Weibull component lives depend on the stress variables through the general stress translation function. Priors on all the parameters, namely the stress coefficients and the Weibull shape parameter, are assumed to be log-concave and independent of each other. This assumption is to facilitate Gibbs sampling from the joint posterior. The samples thus generated from the joint posterior is then used to obtain the Bayesian point and interval estimates of the system reliability at usage condition.

Weibull modulus for diverse strength due to sample-specificity

By sample specificity it is meant that specimens with the same nominal material parameters and tested under the same environmental conditions may exhibit different behavior with diversified strength. Such an effect has been widely observed in the testing of material failure and is usually attributed to the heterogeneity of material at the mesoscopic level. The degree with which mesoscopic heterogeneity affects macroscopic failure is still not clear. Recently, the problem has been examined by making use of statistical ensemble evolution of dynamical system and the mesoscopic stress re-distribution model (SRD). Sample specificity was observed for non-global mean stress field models, such as the duster mean field model, stress concentration at tip of microdamage, etc. Certain heterogeneity of microdamage could be sensitive to particular SRD leading to domino type of coalescence. Such an effect could start from the microdamage heterogeneity and then be magnified to other scale levels. This trans-scale sensitivity is the origin of sample specificity. The sample specificity leads to a failure probability Phi (N) with a transitional region 0 < (N) < 1, so that globally stable and catastrophic modes could co-exist. It is found that the scatter in strength can fit the Weibull distribution very well. Hence, the Weibull modulus is indicative of sample specificity. Numerical results obtained from the SRD for different non-global mean stress fields show that Weibull modulus increases with increasing sample span and influence region of microdamage.

Estudio de fallas en dieléctricos aplicando la distribución estadística de Weibull con datos truncados

Resumen: En el siguiente trabajo se presentan resultados preliminares en la implementación de análisis de fallas aplicando una distribución de Weibull a partir del truncamiento de la población considerada. Se muestra adicionalmente cuáles son los esfuerzos orientados tanto experimentalmente como computacionalmente.

Fracture statistics of brittle materials: Weibull or normal distribution

The fit of fracture strength data of brittle materials (Si3N4, SiC, and ZnO) to the Weibull and normal distributions is compared in terms of the Akaike information criterion. For Si3N4, the Weibull distribution fits the data better than the normal distribution, but for ZnO the result is just the opposite. In the case of SiC, the difference is not large enough to make a clear distinction between the two distributions. There is not sufficient evidence to show that the Weibull distribution is always preferred to other distributions, and the uncritical use of the Weibull distribution for strength data is questioned.

Influence of threshold stress on the estimation of the Weibull statistics

The influence of threshold stress on the estimation of the Weibull statistics is discussed in terms of the Akaike information criterion. Numerical simulations show that, if sample data are limited in number and threshold stress is not too large, the two-parameter Weibull distribution is still a preferred choice. For example, the fit of strength data of glass and ceramics to the two- and three-parameter Weibull distributions is compared.

Bayesian Analysis of Simple Step-stress Model under Weibull Lifetimes

Department of Statistics, Cochin University of Science & Technology, Part of this work has been supported by grants from DST and CSIR, Government of India. 2Department of Mathematics and Statistics, IIT Kanpur