In this issue, we once again bring 10 regular papers to our readers covering several interesting problems in performability engineering. To start with, there are two papers from Professors Hausken of Stavanger University, and Gregory Levitin of Israel. The first paper from them highlights the how resources should optimally be allocated by a defender between a passive defence by employing false targets and to use strikes against the attacker preventively. There has been an exponential growth of papers in the area of attack and defence strategies of systems considered important by the national interest and a variety of models have been presented in the recent past on the subject of system’s defence and attack strategies. In fact, the authors of the first paper together and individually have made tremendous contributions to the literature in this area. Therefore we felt, it is time that we invited them to contribute a review paper on the topic which can be used as a bench mark by the new researchers to the area in order to apprise them of what has been already published and in which direction the future research can be oriented fruitfully. We do hope the paper on Review of Systems Defence and Attack Models will be found very useful by the future researchers in the area. The authors of this paper have done a good job of review by including some 129 papers and categorizing them according to the models, system configuration and various strategies.

The third paper is again a review paper on the subject of signatures and D-spectra of coherent systems. The approach using signature and D-spectra helps in computing system reliability, particularly, the characteristics such as probabilistic resilience, component importance, the system failure probability as a function of component failure probability q, and approximations to system reliability, and so also the bounds on system reliability, if we know only the bounds on q. However, the computations become unwieldy with large number of components. So authors suggest the use of Monte Carlo methods with this approach.

The fourth paper of the issue presents failure risk analysis of a water distribution system which must be treated as continuous operating system and tries to identify the factors responsible for the failure of a water distribution system. The author claims the originality of the approach and may provide authorities in decision making support with regard to the performance of water distribution system.
In the fifth paper of this issue, the authors present a data processing method, fitted to the actual measurements using the Generalized Weibull-FR function. This, they claim, is likely to remove the external noise from the data before it is used as input to the PHM. Two case studies using real-world vibration monitoring data are used to demonstrate the approach. The approach is validated to be effective and likely to save the total average maintenance cost by increasing the average replacement interval and by making better use of remaining useful life.

The authors of the sixth paper present a model to predict the number of residual defects before testing phase of a software. They claim that this information would help the developers if this information is available a priori for optimal testing resource planning and quality assessment of the software being developed. In the early stages of software development life cycle (SDLC), software residual defects are affected by both product and process characteristics of a project. The paper uses software size metric and three metrics of requirement analysis phase for predicting the residual defects during testing or operational phase using fuzzy logic. The predictive capability is demonstrated using data of twenty software projects.

The seventh paper of the issue presents an inspection strategy for a multi-state system. The system being considered can be in any of three states: nominal operating state, degraded state or failure state. A system state is known only after inspection. A maintenance action is undertaken if at a predetermined instant an inspection reveals that the system is in degraded or failure state. The maintenance action is likely to restore the system to its nominal operating state with a certain probability. The paper considers a periodic inspection policy.

In the eighth paper, the author presents a methodology that helps computing reliability of a cold standby system with constituent components having time dependent hazard rates. To keep generality in mind, the author computes reliability of k-out-of-n cold standby system with components having Erlang life time distribution. The paper also considers the failure of decision switch. The use of Erlang distribution allows constant as well as increasing hazard rates for components besides permitting approximate life time distributions for components. This also allows one to have a closed form expression. The author demonstrates efficiency of his approach using some illustrative examples.

The ninth paper of the issue presents an analytical method for evaluating the throughput of a production line composed of two reconfigurable machines separated by a finite capacity buffer. Each machine is assumed to be composed of essential and non essential equipment. The failure of any essential equipment induces the shutdown of the entire machine. The failure of the non essential equipment implies the continuity of the machine service with a reduced level of functionality. Simulation and numerical experiments have been conducted to demonstrate the efficacy of the approach. The proposed model can be used as the building block for the performance evaluation of a long production line using aggregation techniques.

The tenth paper of this issue proposes a Software Reliability Qualification Model (SRQM) model to predict the reliability of a frozen software packages based on actual usage testing. Also the paper presents the cell based modeling of an application. Using this cell based modeling, the developer can point out which part of the application needs reliability improvement to meet the user requirements and also set criteria to qualify a product for the reliability requirements.

The next three papers are short communications which are also refereed and revised just as the regular papers are processed. In the first short communication, the authors propose a novel importance measure for multi-state systems with the consideration of external factors, such as external environment, common cause failures and human factors etc. The external environment can be either contamination of fluid systems, or corrosion, and electrical noise, dirt, dust, humidity and temperature. The proposed importance analysis allows quantifying the effect of the external factor on the component and system performance. Importance measures as such have been widely used for identifying the weakest component and supporting system reliability and performance improvement activities.

The next short communication, reliability analysis of tracking, telemetry and command (TT&C) and communication systems is discussed. Needless to say that its reliability is essential for the success of launching, motioning and landing of spacecraft. In this paper, a formal description of TT&C and communication task is given to facilitate the reliability modeling of such systems. A continuous-time Markov chain (CTMC) model is built for an idle task arc and a model for TT&C and communication tasks in consecutive flight cycles is proposed.

In the last short communication, the authors present the concept of an unconstrained, economically designed, theoretically appropriate monitor for the Poisson process. The authors have presented in this short communication is an alternative approach which is theoretically appropriate in that its assumptions do not deviate from the true distribution of what is being monitored.

We do hope that the papers published in this issue will invigorate further research in the respective areas.

A defender allocates its resource between defending an object passively by deploying false targets, and striking preventively against an attacker seeking to destroy the object. With no preventive strike the defender allocates its entire resource to deploying false targets, which the attacker cannot distinguish from the genuine object. If the defender strikes preventively, the attacker’s vulnerability depends on its protection and on the defender's resource allocated to the strike. If the attacker survives, the object’s vulnerability depends on the attacker’s revenge attack resource allocated to the attacked object. An optimization model is presented for making a decision about the efficiency of the preventive strike based on the estimated attack probability, dependent on a variety of model parameters. The optimal number of false targets to deploy and the optimal subset of targets to attack are determined.

This review paper classifies 129 published papers according to the system structure, defense measures, and attack tactics and circumstances. System structure is further divided into single element, series systems, parallel systems, series-parallel systems, networks, multiple elements, interdependent systems, and other types of systems. Defense measures are divided into separation of system elements, redundancy, protection, multilevel defense, false targets deployment and preventive strike. Attack tactics and circumstances are divided into attack against single element, attack against multiple elements, consecutive attacks, random attack, combination of intentional and unintentional impacts, incomplete information, and variable resources. The classification is intended to give an overview of the field and implicitly suggest future areas of research.

Coherent systems with binary components have important structural parameters known in literature as signatures and D-spectra. The knowledge of these parameters allows to obtain the probabilistic description of coherent systems behavior in the process of their component failures, and such system characteristics as probabilistic resilience, component importance, system failure probability as a function of component failure probability q, approximations to system reliability, and the bounds on system reliability if we know only the bounds on q.
When the system has many components, the exact calculation of signatures or D-spectra becomes a very complicated issue. We suggest the use for their approximation efficient Monte Carlo procedures. All relevant calculations are illustrated by examples of networks.
This paper provides a nonformal review of relevant literature and the methodology of reliability calculations, based mostly on recently published books and research papers.

A water distribution system (WDS) ought to be high reliable continuous operating system. Failure factors in WDS should be identified and prioritised, for example, the causing factors in the most frequent failures in water-pipe network. In this paper, the failure risk analysis of the WDS is presented, and accordingly, a new method consisting the failures index (FI) and the evaluation of risk of failure within the relevant area, based on the assumed categories (tolerable, controlled and unacceptable risk). It is expecting that the methodology for the WDS performance risk analysis would provide the city leadership for decision making support.

In condition based maintenance (CBM) using proportional hazards model (PHM), fitting PHM is a very important step because it has a great influence on the effectiveness of the optimal maintenance policy. Previously actual condition monitoring measurements are directly used to fit the PHM. However this may introduce external noise and the optimal maintenance policy obtained based on this model may not be really optimal. To resolve this problem, a data processing method, which is fitting the actual measurements using the Generalized Weibull-FR function, is proposed to remove the external noise and fit the data before using it as input to the PHM. Two case studies using real-world vibration monitoring data are used to demonstrate the proposed approach. The proposed approach is validated to be effective and will save the total average maintenance cost by increasing the average replacement interval and making better use of remaining useful life.

This paper presents a model to predict the number of residual defects before testing phase. The developers need this information a priori for optimal testing resource planning and quality assessment of the software being developed. In the early stages of software development life cycle (SDLC), software residual defects are affected by both product and process characteristics of the project. These product and process characteristics are embedded in software metrics which have subjective assessments in the early stages of SDLC. Therefore, software metrics are considered for developing the model for early software defects prediction. This paper uses the software size metric and three metrics of requirement analysis phase for predicting the residual defects that are likely to be found during testing or operational usage using fuzzy logic. The predictive capability of the proposed approach is examined using qualitative data of requirement metrics of twenty real software projects and results are compared with the existing model.

In this paper, an inspection strategy for a multi-state system is proposed. This system can be in any of three states: nominal operating state, degraded state or failure state. The system state is known only after inspection. A maintenance action is undertaken when at a predetermined instant an inspection reveals that the system is in degraded or failure state. The maintenance action restores the system to its nominal operating mode with a certain probability. For this study, a periodic type inspection strategy is used. It aims at maximizing the productivity of the considered multi-state system. Analytical and numerical results are presented.

Many fielded systems use cold standby redundancy as an effective system design strategy. However, methods for analyzing the reliability of k-out-of-n cold standby systems, particularly with components having age-dependent hazard rates, are limited. In this paper, using the concepts of counting processes, we propose an efficient method to evaluate the reliability of k-out-of-n cold standby systems. This proposed method considers Erlang distributions for component lives and the effects of switch failures on system reliability. The main advantage of this counting process-based method is that it reduces a complex problem involving multiple integrals into an equivalent simple problem involving one-dimensional convolution integrals. We consider the Erlang distribution for three reasons: (1) it can be used to model either constant or increasing hazard rates, (2) it can be used to approximate several component failure time distributions, and (3) it has well established closed-form expressions for calculating the convolutions that are used in the counting process-based method. We show that all steps involved in finding the reliability of k-out-of-n cold standby system using the proposed method are simple. We demonstrate the proposed method and its computational efficiency using a numerical example.

This paper presents an analytical method for evaluating the throughput of a production line composed of two reconfigurable machines separated by a finite capacity buffer. This means that each machine is composed of essential and non essential equipment. The failure of any essential equipment induces the shutdown of the entire machine. The failure of the non essential equipment implies the continuity of the machine service with a reduced level of functionality. To assess the accuracy of the proposed method, simulation and numerical experiments have been conducted. The proposed model can be used as the building block for performance evaluation of longer production lines using either decomposition or aggregation techniques.

Present software development process provides the developer, flexibility through functional programming technique to embed the third party objects into their software product in order to cut down the unnecessary duplication of coding. The existing models for estimation of software reliability are based on past failure data and whenever the software developer aims to estimate the reliability of the software product, it is compounded with the lack of past failure data of the third party objects, resulting in the loss of accuracy on the estimation process results incomplete estimation. The SRQM model proposed is used to predict the reliability of the frozen software packages based on actual usage testing. Also SRQM introduced the cell based modeling of an application. Using this cell based modeling the developer can point out which part of the application needs reliability improvement to meet the user requirements and also set criteria to qualify a product for the reliability requirements.

Importance measures and analysis have been used to identify weak components to prioritize system upgrading activities, maintenance activities, etc. Traditionally, importance measures do not consider the possible effect due to external environment and phenomena, which however can be causes of system failures and therefore should be taken into consideration. This paper proposes a novel importance measure for multi-state systems with the consideration of external factors. And the proposed importance analysis can effectively quantify the effect of the state of the external factor on the component and system performance.

For the reliability analysis of tracking, telemetry and command (TT&C) and communication systems, most existing modeling methods can only deal with general TT&C and communication tasks. In this paper, a formal description of TT&C and communication task is given to facilitate the reliability modeling of such systems. A continuous-time Markov chain (CTMC) model is built for an idle task arc. A model for TT&C and communication tasks in consecutive flight cycles is proposed, in which the tasks are combined to a new complicated one. Examples with numerical results show the effectiveness of the proposed approach.

Because the probability of Type I error is not evenly distributed beyond upper and lower three-sigma limits the c chart is theoretically inappropriate for a monitor of Poisson distributed phenomena. Furthermore the normal approximation to the Poisson is of little use when c is small. These practical and theoretical concerns should motivate the computation of true error rates associated with individuals control assuming the Poisson distribution.