Analysis of failure data for estimating availability and reliability indicators using statistical and stochastic methods

Due to the great influence of both global climate change and human activities, climatic and hydrological processes in many basins and regions worldwide show obviously nonstationary variability, which has obvious impact on diverse practical water activities such as the hydrological simulation and forecasting, hydrological and hydraulic designs, hydrological risk evaluation and adaption, and water resources planning and management, and also affect the socioeconomic developments as well as people’s lives. Over the last decade, nonstationarity in hydrological process has been becoming a hot and frontier topic in the studies of hydrology and global climate change. However, those basic and major issues on the scientific topic have not been systemically studied, and some understandings about it are limited. In this article, we mainly focused on several major issues about the studies of hydrological nonstationarity. First, we discussed and clarified the difference between the stability of hydrological system and the hydrological nonstationarity. It is pointed out that the former is mainly to study whether the physical hydrological system in a basin keeps its stability under the impacts of both climate change and human activities, which is usually quantified by the deterministic characteristics of hydrological processes, and then simulated by proper physical hydrological models. However, the hydrological nonstationarity is mainly to study whether the statistical characteristics of stochastic hydrological process is changed, including not only the nonstationarity in stochastic components, but also the nonstationarity in deterministic components of hydrological process. Then, we further discussed some key issues about the topic of hydrological nonstationarity, mainly including the research objects, expressing ways, determining criteria, and identification and description methods. We finally pointed out that the studies of hydrological nonstationarity should focus on the stochastic components of hydrological process and the hydrological extremes (droughts, rainstorms, floods, waterlogs and others), but not the deterministic characteristics of hydrological process. All the first, second and third statistical moments (i.e., the mathematical mean, variance, and coefficient of skewness) of hydrological process should at least be carefully evaluated to judge its stationary characters. In practice, both the statistical methods (unit root test, etc.) and stochastic methods can be used for the study of hydrological nonstationarity, that is, we can first identify and remove those periodic and trend components of hydrological time series using proper stochastic methods, and then judge the nonstationary characteristics of stochastic components in series using statistical test methods. Those methods used for the identification of periodicities in hydrological time series usually include the Fourier transform, variance spectral density analysis, and the maximum entropy spectral analysis. Those methods for the trend analysis of hydrological time series mainly include the Mann-Kendall test, Spearman rank correlation test, linear regress test, moving-average test, and the wavelet-based test. The results can be a basis of stochastic simulation and hydrological forecasting. However, during the analysis process the uncertainty in these results should be carefully taken into consideration. As a result, we gave several suggestions and opinions on the future researches of hydrological nonstationarity. It is emphasized that more studies should be conducted to improve the techniques and methods for hydrological modelling, as a reliable basis of revealing the changes in physical hydrological process; further, more studies should also be done to develop the stochastic and statistical methods, mainly for more accurately describing and evaluating the nonstationary characters of hydrological process and their changes. The practice can also be helpful for studying the difficult hydrological and water resources problems under the changing environmental systems.

The role of NHPP models in the practical analysis of maintenance failure data

PurposeThe purpose of this paper is to carry out a reliability analysis of a mechanical system considering the degraded states to get a proper understanding of system behavior and its propagation towards complete failure.Design/methodology/approachThe reliability analysis of computerized numerical control machine tools (CNCMTs) using a multi-state system (MSS) approach that considers various degraded states rather than a binary approach is carried out. The failures of the CNCMT are classified into five states: one fully operational state, three degraded states and one failed state.FindingsThe analysis of failure data collected from the field and tests conducted in the laboratory provided detailed understandings about the quality of the material and its failure behavior used in designing and the capability of the manufacturing system. The present work identified that Class II (major failure) is critical from a maintainability perspective whereas Class III (moderate failure) and Class IV (minor failure) are critical from a reliability perspective.Research limitations/implicationsThis research applies to reliability data analysis of systems that consider various degraded states.Practical implicationsMSS reliability analysis approach will help to identify various degraded states of the system that affect the performance and productivity and also to improve system reliability, availability and performance.Social implicationsIndustrial system designers recognized that reliability and maintainability is a critical design attribute. Reliability studies using the binary state approach are insufficient and incorrect for the systems with degraded failures states, and such analysis can give incorrect results, and increase the cost. The proposed MSS approach is more suitable for complex systems such as CNCMT rather than the binary-state system approach.Originality/valueThis paper presents a generalized framework MSS's failure and repair data analysis has been developed and applied to a CNCMT.

This paper describes the structure of a concept of the land brigade command information system that is a product of the research work of numerous experts throughout the years done in the then-Yugoslav Army, Army of Serbia and Montenegro and today's Army of the Republic of Serbia. The institution formed for this purpose is the Centre for Command Information Systems and Information Support where command information systems are studied continuously and systematically. Today the KIS represents a primary source needed to obtain an efficient military command management system and an instrument for multiplying combat power without increasing the number of units and combat resources. The KIS is a complex multidisciplinary system requiring the implementation of prototype evolution methodology. Due to its complexity, importance end sensitivity, the KIS has to be realized in one's own military R&D institutions belonging to the Ministry of defense, because only these institutions have adequate knowledge and facilities. Fast and ad-hoc solutions in this area are unacceptable from the military, specialized and economic point of view. The primary KIS br KoV structure concept relies on six types of KIS, five technical subsystems and a number of adequate information systems. The key segment are technical systems that represent one open technical and technological (hardware and software) platform in the realization of any KIS type. The KIS is a dynamic permanent system following the users' needs and technical technology changes. It is not, therefore, a closed and finally defined system, it is the system in continuous evolution, under construction and implementations. Introduction The concept described in this paper is the result of work of numerous people, authors included, engaged in the last twenty years of KIS research. The real balance of forces in combat does not depend so much on the potentials of warring parties but rather on the efficiency of command of combat units as well as of the speed of collecting, analyzing and using information with the main goal to make the optimal decision and make the most of one's own potentials, terrain features and enemy's weaknesses. Command and information systems A general definition of command and information systems is that they are the collection of hardware and software solutions used for real-time integration of all organizational structures, military doctrines, technical and technology systems, information flows and processes aiming at the realization of efficient and rational functioning of the military (units, headquarters, etc.). Basic elements in analyzing command management and command information systems A command information system is an element of the land brigade command system intended for the support of command management in securing and leading complex combat operations. The basic principles of the KIS architecture evolution The key factor is the fact that the KIS is a very complex multidisciplinary system requiring the application of the methodology of prototype evolution together with encompassing human and material resources for better effectiveness. Concept of the land brigade command information system On the basis of long-term research and the obtained experiences with practical solutions, the concept of the KIS br KoV is based on the decomposition model of the military command management system which is decomposed onto functional parts (macro-functions) and command unit operations in real time during combat. Conclusion Command information systems (KIS) today represent the basic instrument for efficient military command management systems rather than an instrument for the multiplication of combat power without increasing the number of units and combat resources. The KIS is a complex multidisciplinary system requiring the application of the prototype evolution methodology.

Technical systems are important systems frequently used by applied sciences. Proper operation of technical systems is very important. Therefore, the statistically calculated reliability of a technical system is an important indicator for the system. Technical systems occur in different structures depending on the connection types of the components that constitute the system. The connection diagrams of components can be encountered in a highly complex situation. In such cases, the reliability of the system is difficult to calculate. There is no single method in the literature to calculate the reliability of a technical system. The methods in the literature differ according to the connection types of the systems. In this study, a method and a MATLAB program have been proposed for calculating the reliability of k-out-of-n F systems and consecutive k-out-of-n F systems. The proposed method can also be used for different connections.

Background. Taking into account external threats and impacts when assessing the reliability of technical and information systems is critical to ensuring their sustainability and security. External threats can be varied: from natural disasters (earthquakes, fires, floods) to actions of intruders (cyberattacks, sabotage, vandalism). Taking these threats into account allows developing systems with increased resistance to external factors, minimizing the risk of failures and losses. The relevance of assessing the reliability of technical systems in a safe environment has increased significantly in recent years due to the increasing complexity of systems and increased requirements for their resistance to external threats and malfunctions. Purpose of the work. Select, develop and justify the methods and information technologies for assessing the safety of complex systems, taking into account their operational safety. Materials and methods. The paper studies and presents the main directions of research in this area. These are: reliability modeling, integration of safety and reliability, risk analysis methods, as well as statistical methods and machine learning. The main approaches to solving this problem are considered. It is shown that the assessment of the reliability of technical systems, taking into account the risks of external influences, requires a comprehensive approach that includes not only the analysis of the system's own properties, but also an assessment of the influence of external factors that disrupt its performance. Based on the study and analysis of mathematical methods for solving this problem, the mathematical method of naive Bayes and the machine learning method of a complex Bayesian network for assessing the probability of failure-free operation of complex systems taking into account their safety are developed and presented. Results and conclusions. Taking into account external threats and impacts when assessing the reliability of technical and information systems is critical to ensuring their sustainability and safety. The integration of machine learning in Bayesian networks significantly increases the accuracy and efficiency of assessing their reliability taking into account safety.

The article proposes methods for adjusting the maintenance frequency of complex technical systems with a long service life. Methods for adjusting the maintenance frequency take into account the decrease in the reliability of complex systems and, as a result, the increase in the duration of their forced downtime due to maintenance and the need to eliminate failures and malfunctions. Mathematical models of downtime of complex technical systems due to maintenance are proposed depending on the density of the failure flow and the speed of eliminating failures and malfunctions, economic indicators of operation. Expressions are obtained for calculating the maintenance frequency and optimal downtime due to maintenance of complex technical systems, the implementation of which will ensure their suitability for use for their intended purpose. The purpose of the article is to develop methods for adjusting the frequency of technical maintenance and, as a result, preserving (maintaining) the efficiency and quality of the functioning of the STS, taking into account their aging. To solve the formulated problem, classical mathematical probability theory, the theory of differential equations and renewal theory were used. Before the main results, three models can be summarized that describe the problems of simple folding technical systems through their technical maintenance and elimination of faults and problems. Two of the installed models allow you to restore the reliability of the STS for the last hour of their downtime. Based on data on the reliability of systems and knowing the distribution of the average hour of work on the device, it is possible to calculate the average frequency of maintenance of older systems to maintain their availability up to using for recognition. The third model allows us to calculate the optimal frequency of maintenance of folding technical systems in order to achieve the maximum profit from their operation. The novelty of the research is that many complex technical systems of long-term use require work to maintain their operability by reducing the frequency of maintenance and carrying out work to assess the condition of such systems. Maintaining a given level of reliability and usability of aging complex technical systems, in which planned and preventive maintenance and repair strategies are implemented, requires studying the problem of adjusting maintenance parameters the number and duration of maintenance. To solve such problems, it is necessary to develop models of downtime of STS that would take into account the impact of their reliability on the total downtime. Models of downtime of aging complex technical systems have been developed and calculation formulas have been obtained to determine the optimal number and duration of their maintenance. The novelty of the research is that for many complex technical systems of long-term use, it is necessary to carry out work to maintain their usability by reducing the frequency of maintenance, and to carry out work to assess the condition of such systems.

In the last years, new technical standards for the assessment of the energy efficiency of technical building systems were developed by the European Committee for Standardization (CEN). These procedures were conceived as to combine the easiness of the calculation methodologies and their related assumptions with a sufficient level of accuracy. While the former objective is often achieved, the latter is a challenging task as the procedures sometimes fail to simulate in a proper way the actual performance of the technical systems. On the other hand, the detailed procedures applied by detailed dynamic energy simulation tools are more precise and complex; however, for their application, they need a wide range of input data that are often hard to collect. For this reason, simplified procedures are now commonly applied, above all in the case of existing buildings. Nevertheless, these procedures need to be deeply analysed and validated. In this paper, the main standard calculation procedures addressed to chillers and specified in EN 16798-13:2017 were analysed, with a focus on the required input data and the calculation procedures. The same approach was then applied to the more detailed calculation methods used in the dynamic tools EnergyPlus and TRNSYS, and the results were compared. The theoretical analysis was then followed by a case study approach. A reference office building, representative of the Italian building stock, was selected and analysed in two different Italian climatic zones. The determination of the thermal energy need for cooling was performed by means of EnergyPlus, while the assessment of the energy demand related to technical building systems was performed both with EnergyPlus and with the standard procedures. This paper is part of a wider research activity finalised to the analysis of the technical building systems calculation procedures, taking into account different generation systems and thermal coupling modes between the technical systems and the building itself.

The aim of the research is to analyze the emotional state of a person using stochastic methods and to study the dynamic characteristics of this state. The study uses statistical characteristics that are suitable for creating artificial intelligent systems, as well as for further development and design of organizational and technical systems. It is assumed that the emotional state of a person is determined by the amount of stress he experiences over a certain period of time. To account for the emotional responses of the cognitive system corresponding to the reactions of the human personality, the known results are used. In particular, to characterize the relative values of stress, the Holmes and Rahe scale was used, which contains 43 typical life situations with corresponding relative values of stress. It should be emphasized that all known results relate to time-independent stress models, and only the results of statistical processing of measurements of the impact of stress on the individual at certain time intervals are published.Materials and methods of research are the application of the Poisson model of stress occurrence in the process of system functioning. It is assumed that stresses are those impacts that are then processed, perceived, processed, and used by the individual and the organizational and technical system that models it. The occurrence of stresses over time is modeled by point Poisson processes, and the stress of each type according to Holmes and Rahe is described by a process with the corresponding individual function of the intensity of the occurrence of points (stresses). Dynamic responses of the system are described by well-known response functions in the theory of control systems. The key parameter of the response function is the time constant, which characterizes the typical response time of the system to individual stress.New results of the study are estimates of the average frequency of occurrence of various types of stress, which allowed us to determine the mentioned functions of the intensity of occurrence of stress in the Poisson model. This, in turn, made it possible to develop analytical relations for the relative amount of stress processed by the system for the current time. Thus, a model of the emotional state of the cognitive system (the development of the process of experiencing stress over time) is proposed in the form of a decreasing function of time with a certain time constant that characterizes the inertial properties of experiencing stress. Specific results and corresponding curves are obtained for the exponential response function of the system, which depends on the current time, times of stress occurrence, and relative stress values. In general, they correspond to the predictions of similar reactions for the individual obtained in published studies. In this regard, the model can be applied in the engineering design of organizational and technical systems that require accounting or modeling of emotional reactions.In conclusion, the directions of further development of the theory are indicated. In particular, the possibility of studying the reaction of the system to events, the occurrence of which is described by functions of the intensity of their occurrence, depending on time, which, of course, will bring the research results closer to real situations. Another important area may be the introduction of indicators and criteria for stress tolerance of systems to the consideration and study of behavior.

Aim. Fulfillment of the requirements for the reliability indices of complex technical products and systems is one of the priority tasks to be solved along the stages of development and testing. It is advisable to define the parameter values of the elements of the complex system diagram at the design stage, optimally, in terms of the minimum of an efficiency/cost criterion, ensuring the fulfillment of the requirements for the system reliability. Methods. The main problem that impedes to solve the task of parameter optimization of a model of the reliability diagram is a significant instability of estimation of probability of reliable operation using a Monte-Carlo method (a significant dependence of the rate of estimation error of time). In such conditions an optimization search task could be solved on provision of a stepwise determination of the number of model experiments, which ensures the required accuracy of the estimation of probability of the system reliable operation, necessary for stable operation of the parameter optimization algorithm. The studies of characteristics of estimation of the system reliable operation allowed determining the interrelation of the estimation of reliable operation and the rate of estimation error, offering its approximation in form of a simple formula. The number of model experiments that ensure the required estimation accuracy, is defined using the developed formula determining the interrelation of the estimation of reliable operation and the rate of estimation error, and the known formulas determining the rate of error of the sum N of equally distributed independent random values. Use of the obtained formulas makes it possible to organize the work of the parameter optimization algorithm of the system reliability model by determining its parameters with the required accuracy using minimum computer resources in the context of instability of estimation of probability of the optimizable system reliable operation. Results. Efficiency of the offered approach to realize parameter optimization of a statistical model of the reliability diagram is shown on the sample of estimation of optimal parameters of the system reliability diagram variant, for which there is an analytical solution for the estimation of reliable operation probability. And the results of parameter optimization with the use of analytical value of the probability of reliable operation are the basis for estimation of the accuracy of the algorithm of parameter optimization of the system reliability model operating with the use of a Monte-Carlo method. It has been shown that the offered approaches ensure the convergence of the search algorithm and the required accuracy in estimation of the parameters of the system reliability diagram that optimally ensure the fulfillment of the requirements for the system reliability. Conclusions. The results described in the article confirm technical feasibility and economic viability of determination of optimal values of the system reliability parameters at the design stage. Obtained estimations are the basis for the system integration with required elements, or for the requirements to be set to their reliability, if the development of new elements is necessary. In case there are no elements with design characteristics of reliability, the required reliability of the system can be ensured by special technical redundancy measures and (or) by the creation of the system of technical maintenance and repair.

The existing models of technical systems maintenance management provide required reliability and availability of systems, preferably with as little cost as possible. New concepts, increasingly applied in all areas of life, are based on risk. Such an approach is possible and desirable in the field of theory and engineering of maintenance of technical systems and military technical systems. In this paper, the conceptual definitions of technical systems maintenance, maintenance management models and strategies are given, followed by an overview of strategies and models of maintenance of technical systems, their shematic views and mutual comparisons and trends in the management of maintenance of technical systems, the application of which could lead to savings in the maintenance of both technical and military technical systems.

This chapter describes an assessment methodology for various sustainability indicators of technical systems, such as reliability, availability, fault tolerance, and reliability associated cost of technical safety-critical systems, based on Multi-Level Hierarchical Reliability Model (MLHRM). As an application case of the proposed methodology, the various sustainability indicators of electric vehicle propulsion systems are considered and evaluated on the different levels of the hierarchical model. Taking into account that vehicle traction drive systems are safety-critical systems, the strict requirements on reliability indices are imposed to each of their components. The practical application of the proposed technique for reliability oriented development of electric propulsion system for the search-and-rescue helicopter and icebreaker LNG tanker and the results of computation are presented. The opportunities of improvement regarding reliability and fault tolerance of such technical systems are investigated. The results of the study, allowing creating highly reliable technical systems for the specified operating conditions and choosing the most appropriate system design, are discussed in detail.

Macroscopic transport properties of porous media essentially rely on the geometry and topology of their pore space. The premise of predicting these transport properties is to construct an accurate 3D pore space. To date the methods of modeling porous media are divided into two main groups, direct measurements by some equipment and stochastic statistical methods. Direct measurements of pore structure can be acquired with current equipment such as X-ray computed tomography and laser scanning confocal microscopy, but the unavailability of the equipment and the high cost of the measurement make their widespread application impossible. Many stochastic statistical methods, such as truncated Gaussian random field and simulated annealing methods, reconstruct 3D porous media based on some 2D thin sections by means of lower-order statistical functions. However these functions cannot reproduce the long-range connectivity of pore structure. Therefore, this paper will present a stochastic technique of reconstructing 3D pore space using multiple-point statistics with the purpose of solving the proposed problems. The single normal equation simulation algorithm (SNESIM), one of the most common methods for discrete variable simulation in multiple-point statistics, is the main tool to reproduce the long-range feature of pore space. To test the method, Berea sandstone was used as a sample. In the simulation process, a 2D thin section was taken as the training image for providing patterns of pore structure and some pixels were extracted from it as the conditioning data. The models were reconstructed using the SNESIM algorithm that serves as the simulation engine. In order to test the accuracy of these reconstructed models, pore geometry and topology and transport properties of the reconstructed models were compared with those of the real model obtained by X-ray computed tomography scanning. The comparison result shows that the reconstructed models are good agreement with the real model obtained by X-ray computed tomography scanning in the two-point correlation function, the pore space features and single- and two-phase flow permeabilities, which verifies that the long-range connectivity of pore space can be reproduced by this method. Comparing with other stochastic methods, a more accurate stochastic technique of reconstructing 3D porous media is put forward when only some 2D thin sections are available.

The novel coronavirus pneumonia is an acute respiratory disease. In December 2019, this disease emerged in Wuhan, China. The Chinese government called it SARS-CoV-2 which was subsequently named COVID-19 by the World Health Organization (WHO) [1]. In January 2020, WHO confirmed it as a sustained human to human disease [2]. By March 2020, COVID-19 had been transmitted round the world rapidly and every day large number of new cases were registered. COVID-19 is a leaped type of coronavirus family such as Severe Acute Respiratory Syndrome (SARS) and the Middle East Respiratory Syndrome (MERS) that has been transmitted from wild animals to human [3].

The wind industry has been growing significantly over the past decades, resulting in a remarkable increase in installed wind power capacity. Turbine technologies are rapidly evolving in terms of complexity and size, and there is an urgent need for cost effective operation and maintenance (O&M) strategies. Especially unplanned downtime represents one of the main cost drivers of a modern wind farm. Here, reliability and failure prediction models can enable operators to apply preventive O&M strategies rather than corrective actions. In order to develop these models, the failure rates and downtimes of wind turbine (WT) components have to be understood profoundly. This paper is focused on tackling three of the main issues related to WT failure analyses. These are, the non-uniform data treatment, the scarcity of available failure analyses, and the lack of investigation on alternative data sources. For this, a modernised form of an existing WT taxonomy is introduced. Additionally, an extensive analysis of historical failure and downtime data of more than 4300 turbines is presented. Finally, the possibilities to encounter the lack of available failure data by complementing historical databases with Supervisory Control and Data Acquisition (SCADA) alarms are evaluated.