Aim. To consider matters of dependability of highly critical non-recoverable space products with short operation life, whose failures are primarily caused by design and process engineering errors, manufacturing defects in the course of single-unit or small-scale production, as well as to define the methodological approach to ensuring the required reliability.
Methods. Options were analysed for improving the dependability of entities with short operation life using the case study of single-use mechanical devices and the statistical approaches of the modern dependability theory, special methods of dependability of actuated mechanical assemblies, FMEA, Stage-Gate and ground experiments on single workout equivalents for each type of effect.
Results. It was concluded that additional procedures need to be conducted for the purpose of predicting, mitigation and (or) eliminating possible failures as part of the design process using exactly the same approaches that cause failures, i.e., those of design and process engineering. The engineering approaches to dependability are based on early identification of possible causes of failures, which requires a qualified and systemic analysis aimed at identifying the functionality, performance and dependability of an entity, taking into account critical output parameters and probabilistic indicators that affect the performance of the required functions with the allowable probability of failure. The solution is found using a generalized parametric model of operation and design engineering analysis of dependability.
Conclusion. For highly critical non-recoverable space entities with short operation life, the reliability requirements should be considered primarily in terms financial, economic, safetyrelated and reputational risks associated with the loss of spacecraft. From a design engineer’s standpoint, the number of nines after the decimal point (rounded to a smaller number of nines for increased confidence) should be seen as the indicator for the application of the appropriate approaches to ensuring the required reliability at the stage of product design. In case of two nines after the decimal point it is quite acceptable to use analytical and experimental verification techniques common to the aerospace industry, i.e., dependability calculations using the statistical methods of the modern dependability theory and performance indicators, FMEA and Stage-Gate, ground experiments on single workout equivalents for each type of effect. As the required number of nines grows, it is advisable to also use early failure prevention methods, one of which is the design engineering analysis of dependability that enables designers to adopt substantiated design solutions on the basis of engineering disciplines and design and process engineering methods of ensuring quality and dependability. The choice of either of the above dependability strategies is determined solely by the developer’s awareness and understanding of potential hazards, which allows managing the risk of potential rare failures or reasonably refusing to do so.

Aim. To modify the classical method [1, 4] that causes incorrect estimation of the required size of SPTA in cases when the replacement rate of failed parts is comparable to the SPTA replenishment rate. The modification is based on the model of SPTA target level replenishment. The model considers two situations: with and without the capability to correct requests in case of required increase of the size of replenishment. The paper also aims to compare the conventional and adjusted solution and to develop recommendations for the practical application of the method of SPTA target level replenishment.
Methods. Markovian models [2, 3, 5] are used for describing the system. The flows of events are simple. The final probabilities were obtained using the Kolmogorov equation. The Kolmogorov system of equations has a stationary solution. Classical methods of the probability theory and mathematical theory of dependability [6] were used.
 Conclusions. The paper improves upon the known method of estimating the required size of the SPTA with a safety stock. The paper theoretically substantiates the dependence of the rate of backward transitions on the graph state index. It is shown that in situations when the application is not adjusted, the rates of backward transitions from states in which the SPTA safety stock has been reached and exceeded should gradually increase as the stock continues to decrease. The multiplier will have a power-law dependence on the transition rate index. It was theoretically and experimentally proven that the classical method causes SPTA overestimation. Constraint (3) was theoretically derived, under which the problem is solved sufficiently simply using the classical methods. It was shown that if constraint (3) is not observed, mathematically, the value of the backward transition rate becomes uncertain. In this case, correct problem definition results in graphs with a linearly increasing number of states, thus, by default, the problem falls into the category of labour-intensive. If the limits are not observed, a simplifying assumption is made, under which a stationary solution of the problem has been obtained. It is shown that, under that assumption, the solution of the problem is conservative. It was shown that, if the application is adjusted, the rate of backward transition from the same states should gradually decrease as the stock diminishes. The multiplier will have a hyperbolic dependence on the transition rate index. This dependence results in a conservative solution of the problem of replenishment of SPTA with application adjustment. The paper defines the ratio that regulates the degree of conservatism. It is theoretically and experimentally proven that in such case the classical method causes SPTA underestimation. A stationary solution of the problem of SPTA replenishment with application adjustment has been obtained. In both cases of application adjustment reporting, a criterion has been formulated for SPTA replenishment to a specified level. A comparative analysis of the methods was carried out.

An estimation of the failure flows is a prerequisite for the operation of industrial products. It is based on statistical data about failures that occur within technical items in the process of their operation. In the technical product documentation, this indicator shall be featured in the “Dependability parameter estimation” section. The dependability analysis of rolling stock is still affected by the difficulty of defining the methodology for evaluating this parameter at various system levels. For the purpose of analysing a multicomponent system, a reliability block diagram should be developed, and the possible replacement (redundant) elements should be taken into consideration. Multicomponent systems are often represented through various block diagrams, where, among others, the “m-out-of-n” structure may be used referring to a system with a parallel arrangement of elements that is operable when at least m elements operate. An example of such system is a set of passenger car doors. The manufacturers and customers may have different approaches to calculating technical system dependability. First, the required dependability indicator for the entire train is defined that, in turn, defines the dependability requirements for a car. At the same time, the dependability indicator for a car is determined by the respective values of its components (subsystems, units and parts). However, the nature of the relationship between a car and its components is not always taken into account. At the same time, car manufacturers can and should define in the regulatory documentation (and later supervise in operation) the dependability indicators for a set of doors (components of a car in our case) as a single system. However, the failure criteria of a set of doors are not always defined. This paper examines the method of calculating the failure flow for a set of passenger car doors based on operational data and the failure flow of a single door.
Aim. To propose a method for calculating the failure flow of a set of 6 car doors by analysing the possible reliability block diagrams with subsequent transition to transition and state graphs.
Conclusions. A number of block diagrams were developed for the purpose of dependability calculation of sets of passenger car doors based on the system failure criterion. The failure flow of a set of car doors was calculated according to the developed block diagrams. It is concluded that the Markovian method of calculating the failure flow is of higher priority than the logic-and-probability approach, since it takes into account the recovery factor. A Markovian method was proposed for calculating the failure flow and recovery time of a set of car doors for the “3-out-of-4” reliability block diagram.

Aim. The paper primarily aims to simulate the operation of railway transportation systems using the queueing theory with the case study of marshalling yards. The goals also include the development of the methods and tools of mathematical simulation and queueing theory.
Methods. One of the pressing matters of modern science is the development of methods of mathematical simulation of transportation systems for the purpose of analyzing the efficiency, stability and dependability of their operation while taking into account random factors. Research has shown that the use of the most mature class of such models, the singlephase Markovian queueing systems, does not enable an adequate description of transportation facilities and systems, particularly in railway transportation. For that reason, this paper suggests more complex mathematical models in the form of queueing networks, i.e., multiple interconnected queueing systems, where arrivals are serviced. The graph of a queueing network does not have to be connected and circuit-free (a tree), which allows simulating transportation systems with random structures that are specified in table form as a so-called “routing matrix”. We suggest using the BMAP model for the purpose of describing incoming traffic flows. The Branch Markovian Arrival Process is a Poisson process with batch arrivals. It allows combining several different arrivals into a single structure, which, in turn, significantly increases the simulation adequacy. The complex structure of the designed model does not allow studying it analytically. Therefore, based on the mathematical description, a simulation model was developed and implemented in the form of software.
Results. The developed models and algorithms were evaluated using the case study of the largest Russian marshalling yard. A computational experiment was performed and produced substantial recommendations. Another important result of the research is that significant progress was made in the development of a single method of mathematical and computer simulation of transportation hubs based on the queueing theory. That is the strategic goal of the conducted research that aims to improve the accuracy and adequacy of simulation compared to the known methods, as well as should allow extending the capabilities and applicability of the model-based approach.
Conclusions. The proposed model-based approach proved to be a rather efficient tool that allows studying the operation of railway marshalling yards under various parameters of arrivals and different capacity of the yards. It is unlikely to completely replace the conventional methods of researching the operation of railway stations based on detailed descriptions. However, the study shows that it is quite usable as a primary analysis tool that does not require significant efforts and detailed statistics.

The paper aims to improve the terminology used in dependability-related state standards. Examples are given of the use of the terms “estimate” and “definition” in the “Risk management” and “Dependability in technics” series of state standards. The meanings of those terms were clarified based on the existing regulatory documents. Requirements for the integrity of the used terms were defined. Wordings were proposed for the term definitions that feature the words “estimate” and “definition”.
Aim. To examine and discuss the common, but not sufficiently substantiated terms “estimate” and “definition” used in state standards, i.e., to consider the legitimacy of their application as part of the above series of state standards. Proposals as to the improvement of such terms’ application were also set forth.
Methods. Examples are given of the use of the terms “estimate” and “definition” in state standards. Based on the existing state standards, the actual meanings of the considered terms were clarified: “definition” refers to the way a term is defined, while “estimate” and “estimation” are closely associated with mathematical statistics. The requirements for the integrity of the used terminology are defined and come down to it being unambiguous, consistent within itself and across the relevant state standards. In this context, the shortcomings of the examined terms are shown that are associated with the above requirements, i.e., the meaning, content, essence and key features of such terms are clearly defined. Any comments or references to other regulatory documents are missing as well.
Results. In most standards, in the “Terms and definitions” section, the concept of “definition” is used correctly, i.e., terms are defined. However, in other cases, the concept of “definition” is used in a different sense, as nothing is actually being defined. Based on the term integrity requirements and in light of the above shortcomings, proposed replacements for the terms in question were defined. In most cases, instead of the terms “estimate” and “definition”, it is proposed to use the terms “calculation” and “computation”, as well as their cognates, “calculate”, “compute”. It should be noted that along the state standards, these
terms are used in technical documentation, science papers, monographs and textbooks.
Conclusions. The use of the examined terms in some standards lacks integrity. The requirements of the standardization recommendations are not observed, the terms are not unambiguous and consistent with other standards. Based on these requirements, the paper proposes improved ways of using the terms “estimate” and “definition”. The suggested terms should be considered as a tentative proposal. Final definitions and/or replacements of these terms are to be developed through extensive discussion and compromise.

Aim. Today, there is a pressing matter of protection against steganography-based attacks against information systems. These attacks present a danger as they use the most common data files – especially graphics files – as containers that deliver malicious code to a system or cause a leak of sensitive information. Developing methods of detecting such hidden information is the responsibility of a special subsection of steganography, the steganalysis. Such methods should be extensively used in computer forensics as part of security incident investigation, as well as in automated security systems with integrated modules for analyzing data files for malicious or dangerous information. An important feature of such activities is the need to examine a wide variety of elements and containing files. In particular, it is required to verify not only the colour values of the pixels in images, but their frequency characteristics as well. This raises a number of important questions associated with the best practices of applying steganalysis algorithms and making correct conclusions based on the outputs. The paper aims to briefly analyse the most important and relevant methods of steganalysis, both spatial and frequency, as well as to make conclusions regarding their performance and ways to analyse the outputs based on the test results of the software that implements such methods.
Methods. The steganalysis of concealment within the least significant bits of an image’s pixels uses Pearson’s Chi-square statistical analysis, as well as the Regular-Singular method that involves signature analysis of pixel groups and analytical geometry tools for estimating the relative volume of the hidden message. The Koch-Zhao method of steganalysis is used for the purpose of detecting information embedded in the frequency-domain image representation. It also allows identifying the parameters required for extracting the hidden message.
Results. A software suite was created that includes the software implementations of the analysed methods. The suite was submitted to a number of tests in order to evaluate the outputs of the examined methods. For the purpose of testing, a sample of images of various formats was compiled, in which information was embedded using a number of methods. Based on the results of the sample file analysis, conclusions were made regarding the efficiency of the analysed methods and interpretation of the outputs.
Conclusion. Based on the test results, conclusions were made on the accuracy of the steganalysis methods in cases of varied size of the embedded message and methods of its concealment. The patterns identified with the help of the analysis outputs allowed defining a number of rules for translating the outputs into conclusions on the identification of the fact of detection of hidden information and estimation of its size.

Aim. In this paper we discuss the risk model of the German Corona Warning App in two versions. Both are based on a general semi-quantitative risk approach that is not state of the art anymore and for some application domains even deprecated. The main problem is that parameter estimates are often only ordinal scale or rank numbers for which such operations as multiplication or division are not clearly specified. Therefore, it may results in underestimation or overestimation of the associated risk.
Methods. The risk models that are used in the apps are analyzed. Comparison of the nomenclature of model parameters, their influence on the result, approaches to the generation of a combined risk assessment is carried out. The effectiveness of the models is analyzed.
Results. It is shown that most of the parameters in the model are used only as binary indicator variable. It has been found that the Corona Warning App uses a much more limited model that does not even assess risk, but relies on one parameter which is weighted exposure time. It has been shown that the application underestimates this parameter and therefore may erroneously reassure users. Thus, it may be concluded that the basic risk model implemented before version 1.7.1., is rather a dosimetric model that depends on the calculated virus concentration and does not depend on exposure and other parameters (excluding some threshold values). It is not even a risk model as defined by many standards. Changes of the risk model in the later version are not fundamental. In particular the later model also assesses not individual risk, but individual exposure according to the results. In addition, the model greatly underestimates the duration of exposure. Although it is reported that about 60% of the app’s users have shared positive test results, the absolute number of published results is less than 10% of all positive test results. Therefore, from an individual point of view, the application is effective only in 10% of cases, or even less.
Conclusions. As the Corona Warning App also has other systematic limitations and shortcomings it is advised not to rely on its results but rather on Covid testing or vaccination. In addition, if there are enough virus tests available in the near future, the application will even become outdated. It will be better to develop an application that can assess risks a priori, as a kind of decision support for its users based on their individual risk profile.

The paper Aims to examine various approaches to the ways of improving the quality of predictions and classification of unbalanced data that allow improving the accuracy of rare event classification. When predicting the onset of rare events using machine learning techniques, researchers face the problem of inconsistency between the quality of trained models and their actual ability to correctly predict the occurrence of a rare event. The paper examines model training under unbalanced initial data. The subject of research is the information on incidents and hazardous events at railway power supply facilities. The problem of unbalanced data is expressed in the noticeable imbalance between the types of observed events, i.e., the numbers of instances.
Methods. While handling unbalanced data, depending on the nature of the problem at hand, the quality and size of the initial data, various Data Science-based techniques of improving the quality of classification models and prediction are used. Some of those methods are focused on attributes and parameters of classification models. Those include FAST, CFS, fuzzy classifiers, GridSearchCV, etc. Another group of methods is oriented towards generating representative subsets out of initial datasets, i.e., samples. Data sampling techniques allow examining the effect of class proportions on the quality of machine learning. In particular, in this paper, the NearMiss method is considered in detail.
Results. The problem of class imbalance in respect to the analysis of the number of incidents at railway facilities has existed since 2015. Despite the decreasing share of hazardous events at railway power supply facilities in the three years since 2018, an increase in the number of such events cannot be ruled out. Monthly statistics of hazardous event distribution exhibit no trend for declines and peaks. In this context, the optimal period of observation of the number of incidents and hazardous events is a month. A visualization of the class ratio has shown the absence of a clear boundary between the members of the majority class (incidents) and those of the minority class (hazardous events). The class ratio was studied in two and three dimensions, in actual values and using the method of main components. Such “proximity” of classes is one of the causes of wrong predictions. In this paper, the authors analysed past research of the ways of improving the quality of machine learning based on unbalanced data. The terms that describe the degree of class imbalances have been defined and clarified. The strengths and weaknesses of 50 various methods of handling such data were studied and set forth. Out of the set of methods of handling the numbers of class members as part of the classification (prediction of the occurrence) of rare hazardous events in railway transportation, the NearMiss method was chosen. It allows experimenting with the ratios and methods of selecting class members. As the results of a series of experiments, the accuracy of rare hazardous event classification was improved from 0 to 70-90%.