IJPAM: Volume 66, No. 1 (2011)


Soma Roychowdhury$^1$, Debasis Bhattacharya$^2$
$^1$Indian Institute of Social Welfare and Business Management
College Square West, Kolkata, 700 073, INDIA
e-mail: srcdb@yahoo.com
$^2$Visva-Bharati University
Santiniketan, INDIA
e-mail: debases_us@yahoo.com

Abstract.We generally think about engineering system whose component life times are independent and consider its life length, probably the most meaningful effectiveness measure, to measure the system's quality. When there is a question of selection of the best system from a collection of $N$ systems, placed into service, it is quite common to collect the data on their life lengths (or times-to-failure) for the purpose of assessing the overall system performance. In the life-testing experiment, prior to deployment of the system or sale of the system, such data are very useful and critically analyzed. However, the assumption of independence of component lifetimes in the system is a very restrictive assumption, and the components may be dependent (in many ways) rather than being independent. The systems become more complex under the dependent set-up and the dependence of components is difficult to describe even for simple systems. It is essential to study the effect of dependence on system performance for better reliability design and its analysis. It is possible that we might underestimate or overestimate the system performance if we ignore the effect of dependence. In this paper we consider different multi-component coherent systems, and  make an attempt to explore the effect of dependence on the system performance.

Received: October 30, 2010

AMS Subject Classification: 90B25, 62P30

Key Words and Phrases: coherent systems, minimal cut sets, positive dependence, positive lower orthant dependence (PLOD), positive upper orthant dependence (PUOD), system life, system reliability

Source: International Journal of Pure and Applied Mathematics
ISSN: 1311-8080
Year: 2011
Volume: 66
Issue: 1