International Journal of Reliability, Quality and Safety Engineering

Predicting the reliability of software systems based on a component-based approach is inherently difficult, in particular due to failure dependencies between software components. One possible way to assess and include dependency aspects in software reliability models is to find upper bounds for probabilities that software components fail simultaneously and then include these into the reliability models. In earlier research, it has been shown that including partial dependency information may give substantial improvements in predicting the reliability of compound software compared to assuming independence between all software components. Furthermore, it has been shown that including dependencies between pairs of data-parallel components may give predictions close to the system's true reliability. In this paper, a Bayesian hypothesis testing approach for finding upper bounds for probabilities that pairs of software components fail simultaneously is described. This approach consists of two main steps: (1) establishing prior probability distributions for probabilities that pairs of software components fail simultaneously and (2) updating these prior probability distributions by performing statistical testing. In this paper, the focus is on the first step in the Bayesian hypothesis testing approach, and two possible procedures for establishing a prior probability distribution for the probability that a pair of software components fails simultaneously are proposed.

Reliability of a software product should be tracked during the software lifecycle right from the architectural phase to its operational phase. Heterogeneous systems consist of several globally distributed components, thus rendering their reliability evaluation more complex with respect to the conventional methods. In this context, reliability prediction of software process oriented systems assumes prime importance. It is important to take into account the proven processes like Rational Unified Process (RUP) to mitigate risks and increase the reliability of systems while building distributed based applications. This paper presents an algorithm using feed-forward neural network for early qualitative software reliability prediction. The inputs for neural networks consist of techno-complexity, practitioner's level, creation effort, size and leakage defects. The number of defects detected in each cycle can be predicted by using Artificial Neural Networks (ANN). Illustrative examples prove the effectiveness of the methodology employed.

In this article, a new profit function based on a surrogate variable of its performance variable is provided to develop an optimum screening procedure for manufacturing. The optimum screening procedure helps producers set up the mean level of manufacturing and the screening limits of a surrogate variable to reach a maximum expected profit per unit. The proposed method is useful when the products in the manufacturing process are classified into different grades and sold in two alternate markets. A cement-packing example is used to illustrate the proposed method, and a numerical study is conducted to evaluate the effects of cost components and distribution parameters on the expected profit per unit. The proposed screening procedure provides a significant improvement over existing methods in term of higher expected profit per unit.

Modeling of software reliability has gained lot of importance in recent years. Use of software-critical applications has led to tremendous increase in amount of work being carried out in software reliability growth modeling. Number of analytic software reliability growth models (SRGM) exists in literature. They are based on some assumptions; however, none of them works well across different environments. The current software reliability literature is inconclusive as to which models and techniques are best, and some researchers believe that each organization needs to try several approaches to determine what works best for them. Data-driven artificial neural-network (ANN) based models, on other side, provide better software reliability estimation. In this paper we present a new dimension to build an ensemble of different ANN to improve the accuracy of estimation for complex software architectures. Model has been validated on two data sets cited from the literature. Results show fair improvement in forecasting software reliability over individual neural-network based models.

In general railway infrastructure operates in a complex and non homogeneous environment where low power electronics has to function in the similar environment as large voltages and currents from the trains. The environment close to the railway tracks is heavily polluted by electromagnetic (EM) noise from the railway systems itself. The reliability of the railway signalling-, communication-, and control system depends on the degree of isolation from EM noise. It is important to observe the real system and its characteristics in real situation. The complexity of the infrastructure is not easy to simulate or calculate, and therefore measurements were performed on real system in operation. A large number of measurements were made on site at signalling systems and installations of Banverket (the Swedish Rail Administration). By studying and analyzing these measurements, the extent of EMI (Electromagnetic interferences) -related faults can be estimated. In situ measurements must be performed on systems that have obvious problems show increasingly fast and high transients in a DC signalling subsystem. The statistics also show an increased activity of transients during a period before a certain circuit in this system gets out of order. The presented results clearly show how the electromagnets interference effects he signalling system in a signalling box. The measurements in this case study have provided new information on a railway subsystem and have revealed frequencies that are difficult to explain at this moment.