International Journal of Reliability, Quality and Safety Engineering

European Railway Traffic Management System/European Train Control System (ERTMS/ETCS) is a recent standard aimed at improving performance, safety and inter-operability of modern railways. In order to be compliant to ERTMS/ETCS, a railway signalling system must meet strict nonfunctional requirements on system level failure modes. In this paper, a multiformalism model is employed to perform an availability analysis of an ERTMS/ETCS reference architecture at early phases of its development cycle. At this aim, a bottom-up analysis is performed from subsystem failure models (expressed by means of Generalized Stochastic Petri Nets, Fault Trees and Repairable Fault Trees) up to the overall system model. The modular approach, here used, allows to evaluate the influence of basic design parameters on the probability of system-level failure modes and demonstrates that system availability is within the bound required by the ERTMS/ETCS specification. The results show that the multiformalism modeling approach helps to cope with complexity, eases the verification of availability requirements and can be successfully applied to the analysis of complex critical systems.

Maintenance contracts received significant attentions in the past 15 to 20 years as it has huge potential to reduce the upfront investments in maintenance infrastructure, specialized maintenance facilities, and risks to the owners through expert services provided by the original equipment manufacturers and/or specialist maintenance providers. To make maintenance contracts more effective, there is a need to develop mathematical models and understand future costs that could be built into the contract price. In this paper, a conceptual model is developed for estimating cost of outsourcing maintenance of complex and critical asset/equipment taking into account both corrective and preventive maintenance as servicing strategies and risks and penalty costs associated with such contract. The developed cost models are analyzed using illustrated numerical examples.

Inspection threshold is one of effective methods to reduce causes of misjudgement under imperfect inspection. Based on the Bayesian theory, the threshold is defined as the boundary between inspection and noninspection execution. However, the formulation of the true threshold is difficult under a general complex system. This paper proposes approximation methods to derive the maximum and the minimum inspection thresholds for a general system. A binary reliability theory is used for the derivation of the thresholds. Structure functions, minimal path and cut sets are used for the approximation. Finally, some numerical examples are given.

Software quality prediction models are useful tools for creating high quality software products. The general process is that practitioners use software metrics and defect data along with various data mining techniques to build classification models for identifying potentially faulty program modules, thereby enabling effective project resource allocation. The predictive accuracy of these classification models is often affected by the quality of input data. Two main problems which can affect the quality of input data are high dimensionality (too many independent attributes in a dataset) and class imbalance (many more members of one class than the other class in a binary classification problem). To resolve both of these problems, we present an iterative feature selection approach which repeatedly applies data sampling (to overcome class imbalance) followed by feature selection (to overcome high dimensionality), and finally combines the ranked feature lists from the separate iterations of sampling. After feature selection, models are built either using a plain learner or by using a boosting algorithm which incorporates sampling. In order to assess the impact of various balancing, filter, and learning techniques in the feature selection and model-building process on software quality prediction, we employ two sampling techniques, random undersampling (RUS) and synthetic minority oversampling technique (SMOTE), and two ensemble boosting approaches, RUSBoost and SMOTEBoost (in which RUS and SMOTE, respectively, are integrated into a boosting technique), as well as six feature ranking techniques. We apply the proposed techniques to several groups of datasets from two real-world software systems and use two learners to build classification models. The experimental results demonstrate that RUS results in better prediction than SMOTE, and also that boosting is more effective in improving classification performance than not using boosting. In addition, some feature ranking techniques, like chi-squared and information gain, exhibit better and more stable classification behavior than other rankers.

In this paper, we study a decision-making problem related to software creation using commercial-off-the-shelf (COTS) products in a modular software system. The optimal selection of COTS products is difficult due to the variations in various critical parameters such as cost, reliability, execution time, and delivery time. Further, it is difficult to estimate precisely the values of these parameters since sufficient data may not be available and also there could be measurement errors. We present a fuzzy 0–1 optimization model of the multiobjective COTS products selection problem using exponential membership functions that simultaneously minimize the total cost, size, execution time and delivery time and maximize the reliability of a modular software system subject to many realistic constraints. The fuzzy goals are defined for each selection criterion as per the preferences of the decision maker and are aggregated using product operator to obtain an equivalent optimization model for optimal COTS selection. A real-world case study is discussed to demonstrate the effectiveness of the proposed model and the solution methodology.