Recently data mining methods have gained importance in addressing network security issues, including network intrusion detection — a challenging task in network security. Intrusion detection systems aim to identify attacks with a high detection rate and a low false alarm rate. Classification-based data mining models for intrusion detection are often ineffective in dealing with dynamic changes in intrusion patterns and characteristics. Consequently, unsupervised learning methods have been given a closer look for network intrusion detection. We investigate multiple centroid-based unsupervised clustering algorithms for intrusion detection, and propose a simple yet effective self-labeling heuristic for detecting attack and normal clusters of network traffic audit data. The clustering algorithms investigated include, k-means, Mixture-Of-Spherical Gaussians, Self-Organizing Map, and Neural-Gas. The network traffic datasets provided by the DARPA 1998 offline intrusion detection project are used in our empirical investigation, which demonstrates the feasibility and promise of unsupervised learning methods for network intrusion detection. In addition, a comparative analysis shows the advantage of clustering-based methods over supervised classification techniques in identifying new or unseen attack types.
For discrete distribution with reliability function R(k), k = 1, 2,…,[R(k - 1) - R(k)]/R(k - 1) has been used as the definition of the failure rate function in the literature. However, this is different from that of the continuous case. This discrete version has the interpretation of a probability while it is known that a failure rate is not a probability in the continuous case. This discrete failure rate is bounded, and hence cannot be convex, e.g., it cannot grow linearly. It is not additive for series system while the additivity for series system is a common understanding in practice. In the paper, another definition of discrete failure rate function as In[R(k - 1)/R(k)] is introduced, and the above-mentioned problems are resolved. On the other hand, it is shown that the two failure rate definitions have the same monotonicity property. That is, if one is increasing/decreasing, the other is also increasing/decreasing. For other aging concepts, the new failure rate definition is more appropriate. The failure rate functions according to this definition are given for a number of useful discrete reliability functions.
Reliable software is mandatory for complex mission-critical systems. Classifying modules as fault-prone, or not, is a valuable technique for guiding development processes, so that resources can be focused on those parts of a system that are most likely to have faults. Logistic regression offers advantages over other classification modeling techniques, such as interpretable coefficients. There are few prior applications of logistic regression to software quality models in the literature, and none that we know of account for prior probabilities and costs of misclassification. A contribution of this paper is the application of prior probabilities and costs of misclassification to a logistic regression-based classification rule for a software quality model. This paper also contributes an integrated method for using logistic regression in software quality modeling, including examples of how to interpret coefficients, how to use prior probabilities, and how to use costs of misclassifications. A case study of a major subsystem of a military, real-time system illustrates the techniques.
Optimal solutions to the redundancy allocation problem are determined for systems designed with multiple k-out-of-n subsystems in series. The objective is to select the components and redundancy levels to maximize system reliability given system-level constraints. The individual subsystems may use either active or cold-standby redundancy, or they may require no redundancy. Previously, optimization methods for this problem either pertained to k-out-of-n systems consisting of a single subsystem or to series–parallel systems (k=1). Additionally, it had generally been assumed that only active redundancy was to be used. In practice design problems can vary appreciably from these restrictions and the design process may consider more complex system configurations. Unfortunately, available optimization algorithms are inadequate for many of these design problems. The methodology presented here is specifically developed to accommodate the case with k-out-of-n subsystems. Optimal solutions to the problem are found by an equivalent problem formulation and integer programming. The methodology is demonstrated on a well-known test problem with interesting results. The availability of this tool fills a void and should result in more reliable and cost-effective engineering designs.
Most complex systems today contain software, and systems failures activated by software faults can provide lessons for software development practices and software quality assurance. This paper presents an analysis of software-related failures of medical devices that caused no death or injury but led to recalls by the manufacturers. The analysis categorizes the failures by their symptoms and faults, and discusses methods of preventing and detecting faults in each category. The nature of the faults provides lessons about the value of generally accepted quality practices for prevention and detection methods applied prior to system release. It also provides some insight into the need for formal requirements specification and for improved testing of complex hardware-software systems.
This paper proposes a method of estimating the lifetime distribution at use condition for constant stress accelerated life tests when an extrinsic failure mode as well as intrinsic one exists. A mixture of two distributions is introduced to describe these failure modes. It is assumed that the log lifetime of each failure mode follows a location-scale distribution and a linear relation exists between the location parameter and the stress. An estimation procedure using the expectation and maximization algorithm is proposed and specific formulas for Weibull distribution are obtained. Simulation studies are performed to investigate the properties of the estimates and the effects of stress level. Numerical comparisons with the masked data model are also performed.
For products sold with warranty, preventive maintenance actions by manufacturers and/or buyers have an impact on the total costs for both parties. This paper develops a framework to study preventive maintenance actions when items are sold under warranty and reviews the models that have appeared in the literature. It then develops a new model and carries out its analysis.
This paper proposes a random and age replacement policy for an operating unit which works at random times. First, the unit is replaced before failure at a planned time T or at the completion of a working time, whichever occurs first. The expected cost rate is obtained. Next, as one extended model, the unit is replaced before failure at a number N of working times or at a planned time T. An optimal policy which minimizes the expected cost rate is discussed analytically, and its numerical example is given. Two modified models, where the unit is replaced at the first completion of the working time over time T or at number N, and it is replaced at time T or number N, whichever occurs last, are considered. Furthermore, we show that this corresponds to a cumulative damage model by replacing shock with work. Finally, one optimization problem of how much time to preset for scheduling the completion of N works is proposed.
Currently corrective maintenance and time-based preventive maintenance strategies are widely used in wind power industry. However, few methods are applied to optimize these strategies. This paper aims to develop opportunistic maintenance approaches for an entire wind farm rather than individual components that most of the existing studies deal with. Furthermore, we consider imperfect actions in the preventive maintenance tasks, which address the issue that preventive maintenance do not always return components to the as-good-as-new status in practice. In this paper we propose three opportunistic maintenance optimization models, where the preventive maintenance is considered as perfect, imperfect and two-level action, respectively. Simulation methods are developed to evaluate the costs of the proposed opportunistic maintenance policies. Numerical examples are provided to demonstrate the advantage of the proposed opportunistic maintenance methods in reducing the maintenance cost. The two-level action method demonstrates to be the most cost-effective in different cost situations, while the imperfect maintenance policy, which is a simpler method, is a close second. The developed methods are expected to bring immediate benefits to wind power industry.
When monitoring the process variability, it is a common practice that a Phase I data set is used to estimate the unknown in-control process standard deviation σ0 or variance [Formula: see text] to set up the control limits, then monitoring proceeds. Once the process is considered to be in-control, the estimated control limits are assumed as fixed. This practice ignores the effect of estimating the unknown in-control process variance [Formula: see text]. In this paper, we derive the exact run length distribution of the S2 control chart when the in-control process variance [Formula: see text] is estimated and find that m = 200 or more Phase I samples are needed to neglect the effect of using estimated control limits. New control limits when m is small are also derived.
This paper, proposes the Multi-Criteria Decision Making (MCDM) methodology for selection of a maintenance strategy to assure the consistency and effectiveness of maintenance decisions. The methodology is based on an AHP-enhanced TOPSIS, VIKOR and benefit-cost ratio, in which the importance of the effectiveness appraisal criteria of a maintenance strategy is determined by the use of AHP. Furthermore, in the proposed methodology the different maintenance policies are ranked using the benefit-cost ratio, TOPSIS and VIKOR. The method provides a basis for consideration of different priority factors governing decisions, which may include the rate of return, total profit, or lowest investment. When the preference is the rate of return, the benefit-cost ratio is used, and for the total profit TOPSIS is applied. In cases where the decision maker has specific preferences, such as the lowest investment, VIKOR is adopted. The proposed method has been tested through a case study within the aviation context for an aircraft system. It has been found that using the methodology presented in the paper, the relative advantage and disadvantage of each maintenance strategy can be identified in consideration of different aspects, which contributes to the consistent and rationalized justification of the maintenance task selection. The study shows that application of the combined AHP, TOPSIS, and VIKOR methodologies is an applicable and effective way to implement a rigorous approach for identifying the most effective maintenance alternative.
In this paper, software reliability models based on a nonhomogeneous Poisson process (NHPP) are summarized. A new model based on NHPP is presented. All models are applied to two widely used data sets. It can be shown that for the failure data used here, the new model fits and predicts much better than the existing models. A software program is written, using Excel & Visual Basic, which can be used to facilitate the task of obtaining the estimators of model parameters.
This paper presents a comprehensive survey of discrete probability distributions used in reliability for modeling discrete lifetimes of nonrepairable systems. The basic properties of each model are given. A classification into two families is proposed, highlighting the interest of using a Pólya urn scheme. The quality of the estimation of models parameters is numerically assessed. Some criteria are given in order to select among the presented distributions the most useful for applications.
This paper reviews selected published literature on common-cause failures. These publications have been collected from conference proceedings, engineering reliability and safety journals, textbooks etc. The collected literature is grouped into several categories and a comprehensive list of references is presented.
Control charts based on geometric distribution have shown to be very useful in the monitoring of high yield manufacturing processes and other applications. It is well known that the traditional 3-sigma limits will give too many false alarms and the probability limits should be used. This paper shows that the average time to alarm may even increase at the beginning when the process is deteriorated. A new procedure is established for the setting of control limits so that the average run length is maximized when the process is at the normal level. Hence the chart sensitivity can be improved. For the derivation of the control limits in this new procedure, a simple adjustment factor is suggested so that the probability limits can be used after the adjustment.
This paper presents new sequential imperfect preventive maintenance (PM) models incorporating adjustment/improvement factors in hazard rate and effective age. The models are hybrid in the sense that they are combinations of the age reduction PM model and the hazard rate adjustment PM model. It is assumed that PM is imperfect: It not only reduces the effective age but also changes the hazard rate, while the hazard rate increases with the number of PMs. PM is performed in a sequence of intervals. The objective is to determine the optimal PM schedule to minimize the mean cost rate. Numerical examples for a Weibull distribution are given.
Suppose that shocks arrive and act on a system according to a Poisson distribution with mean rate of arrival equal to λ shock(s) per unit time. A δ-shock failure model is proposed in this paper, which assumes that when a system is acted on by a shock, it will recover fully in time δ(>0), and after that it will function as if no shock had occurred before. If the time lag between two successive shocks is less than δ, the second shock will cause failure of the system. Theoretical expressions related to the distribution of the failure time of the system are derived. These results can be used to optimize the design of a system from a costing point of view.
