Gaofeng Da

On computing signatures of coherent systems

It is difficult to compute the signature of a coherent system with a large number of components. This paper derives two basic formulas for computing the signature of a system which can be decomposed into two subsystems (modules). As an immediate application, we obtain the formula for computing the signature of systemwise redundancy in terms of the signatures of the original system and the backup one. The formula for computing the signature of a componentwise redundancy system is also derived. Some examples are given to illustrate the power of the main results.

Further results for parallel systems with two heterogeneous exponential components

In this paper, we investigate ordering properties of lifetimes of parallel systems with two independent heterogeneous exponential components with respect to likelihood ratio and hazard rate orders. Two sufficient conditions are provided for likelihood ratio and hazard rate orders to hold between the lifetimes of two parallel systems, respectively. Moreover, we extend the results from exponential case to the proportional hazard rate models. The results established here strength some of the results known in the literature. Finally, some numerical examples are given to illustrate the theoretical results derived here as well.

Component Level Versus System Level

A well-known principle in engineering is that redundancy at the component level is generally more reliability effective than that at the system level. Here, the redundancy simply means that components are connected in parallel. Motivated by this principle, in this paper, a more general problem of assembling

k

K

-Out-of-

-out-of-

n

N

Assembly Systems

systems optimally is studied. We consider two assembled systems: one is by assembling a

On Bivariate Signatures for Systems with Independent Modules

k

On the Lorenz ordering of order statistics from exponential populations and some applications

-out-of-

On sample ranges from two sets of heterogenous random variables

n