Soma Roychowdhury

Rearrangement of Components in a System Using Component Importance Measures

In this paper we consider the system reliability of coherent systems consisting of components with independently distributed lives. In reliability engineering, increasing the system reliability is an important issue, which can be achieved in various ways, such as, by using high quality components, by reducing the operational load on the components, by implementing better maintenance etc., but selecting the optimal way is a difficult job. Here an attempt has been made to increase the system reliability of coherent systems by rearranging the components pairwise using various component importance measures.

A study on estimation of reliability of a coherent system under field condition

An engineered system is a functionally related group of components, and thus estimation of its reliability depends on the estimation of the reliability of its components. The component reliabilities are estimated usually under a laboratory set up while the system may be installed to work under a different condition, called field condition. The difference may arise due to the change in temperature, humidity, voltage, friction, etc. The cause of the difference may be attributed to a random variable, known as an environmental variable, which affects the functioning of the system. The presence of this variable makes the component failure-times dependent, and hence they behave differently from what they would have behaved under independent set up. Naturally, this change in the behavior of the components affects the estimates of system reliability. Thus, under such situation, if the reliability of a system is not estimated considering the conditions under which it actually works, there will be an upward or downward bias in estimation. Here we make an attempt to study the effect of an environmental variable on the reliability estimation of a complex coherent system, and try to find out the conditions under which the system performance is optimized. A simulation study has been included to see the crossing behavior of the reliability function, which describes the nature of the reliability curves under laboratory and field conditions, at different time points, in particular, when they cross each other.

Comparison of lifetimes of coherent systems using a stochastic ordering criterion

Finding an optimal way to compare the systems of different orders and to order them according to their lives, which are random in nature, is a problem of great concern to the reliability engineers. A direct method of comparing system lives or an indirect method using dominance of the signature vectors of the systems are often used. But these methods have their own limitations. In particular, it has been observed that these methods work under a restrictive setup and for a group of simple systems only. The problem becomes complicated for large and complex systems. In this paper a technique based on stochastic ordering has been proposed to resolve the issue. The method discussed here works under a more general and less restrictive framework. It has been shown to work well for different coherent systems with different component life distributions. Simulation studies have been done to prove the worthiness of the proposed method.

A warehouse selection problem based on random delivery times

This paper presents an approach to selecting the best warehouse, from a number of warehouses that is expected to have the highest possibility to deliver the goods to a retailer under consideration within a specified time. The decision is made at the design phase of the supply chain on the basis of the data on delivery times collected from the warehouses using affordable resources. For determining the necessary sample sizes, available resource is allocated over the warehouses. Here the delivery times are assumed to be random. A method of finding the optimal solution has been developed and applied to different delivery time distributions. A nonparametric method has been adopted in case the delivery time distributions are unknown. Numerical examples are provided for illustrating the method. Simulations have been carried out to show the worthiness of the proposed method. A sensitivity analysis has been done to examine how the optimal allocation changes with the change in the parameter values of the underlying distributions.

On the selection of a best system from a group of systems with mutually dependent component lives

This paper presents a probabilistic optimization problem of selecting the best system from a group of competing coherent systems with mutually dependent components by optimizing the probability that the system will meet a specified target lifetime and ranking the systems from best to worst according to the orders of those probability values from highest to lowest. The problem of comparing different system designs in search for the best configuration is of great concern to the reliability engineers, especially when the systems are of different orders and having mutually dependent components. Earlier works in this direction have discussed the problem for systems with independent components. The problem becomes complicated if the component lives are mutually dependent. The paper attempts to find an approach to solving the issue of comparing coherent systems of different orders with mutually dependent components. Here we select a system by minimizing the probability that the system will fail to meet a specified target lifetime, where the component lifetimes are mutually dependent, using a stochastic ordering criterion. A technique based on stochastic ordering has been proposed to compare systems of different orders. Some complex systems have been used to show that the result derived in this paper can be used effectively to make comparison of the systems with dependent components and to find the best system from a group of competing systems of different orders.