Distortion and copula functions represent powerful tools in the description of the reliability of some complex systems as functions of their components' reliability. On this aim we study several pairs of reliability systems with one or more shared components, in the case when their lifetimes are independent and identically distributed or independent but not identically distributed. We focus on the dependence that arises from sharing components, often described by Marshall-Olkin copulas, making use of some distance measures related to the Gini's mean difference and its new recent generalizations. A special role is played by a new distortion function related to the ROC curve.

Mean Distances and Dependence Structures for Lifetimes of Systems With Shared Components / Capaldo, Marco; Di Crescenzo, Antonio; Pellerey, Franco. - In: APPLIED STOCHASTIC MODELS IN BUSINESS AND INDUSTRY. - ISSN 1524-1904. - ELETTRONICO. - 41:2(2025), pp. 1-13. [10.1002/asmb.70002]

Mean Distances and Dependence Structures for Lifetimes of Systems With Shared Components

Pellerey, Franco
2025

Abstract

Distortion and copula functions represent powerful tools in the description of the reliability of some complex systems as functions of their components' reliability. On this aim we study several pairs of reliability systems with one or more shared components, in the case when their lifetimes are independent and identically distributed or independent but not identically distributed. We focus on the dependence that arises from sharing components, often described by Marshall-Olkin copulas, making use of some distance measures related to the Gini's mean difference and its new recent generalizations. A special role is played by a new distortion function related to the ROC curve.
File in questo prodotto:
File Dimensione Formato  
Appl Stoch Models Bus Ind - 2025 - Capaldo - Mean Distances and Dependence Structures for Lifetimes of Systems With.pdf

accesso aperto

Descrizione: Open version
Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Creative commons
Dimensione 572.47 kB
Formato Adobe PDF
572.47 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2998191