Abstract. Aim. Enable prediction and planning for large-scale unprecedented power outages of importance for emergency planning and national response actions. Predict outage probability, duration and restoration using a theoretical framework that is applicable globally. Methods. Data have been collected for power losses and outage duration for a wide range of events in Belgium, Canada, Eire, France, Japan, Sweden, New Zealand and USA. A new theory and correlation is given for the probability of large regional power losses of up to nearly 50,000 MW(e) without additional infrastructure or grid damage. For severe and rare events with damage (major floods, fire, ice storms, hurricanes etc.) the outages are longer and the restoration probability depends on the degree of difficulty that limits access and restoration. The dynamic reliability requirements for emergency back-up power and pumping systems are derived, and demonstrated using the flooding of New Orleans by Hurricane Katrina and of the Fukushima nuclear reactors by a tsunami. Conclusions. Explicit expressions have been given and validated for the probability and duration for the full range from “normal” large power losses to extended outages due to rare and more severe events with access and repair difficulty.

Abstract. The quantitative indicators of the dependability characteristics of process equipment, machines and entities include the reliability, maintainability, durability, storability. The Aim of the research is to develop a methodological framework of experimental determination of dependability indicators when performing static bending tests of ball valves. The aim is to be achieved by solving individual scientific problems, namely the development of the method of static bending testing of ball valves, experimental determination of the destructive test coefficient, analytical calculation of the coefficients of failure, efficiency preservation and product quality. A test stand was developed for experimental determination of the destructive test coefficient. The employed scientific methods of static testing are suggested for the first time ever. The minimal bending test load is the defining value, therefore the experimental work requires a large sample. Additionally, for the ball valve with the lowest endured bending load out of the tested items of the same diameter, it is recommended to use a specially developed methodology to identify such valve as solid-state monolithic units. The conclusions of the conducted experimental and theoretic research are in compliance with the solved research and development objectives. By conducting the experimental research, the authors evaluated the “enhanced” capabilities of ball valves by various manufacturers through comparative analysis of the conducted static bending tests. It is recommended to consider a solid-state monolithic unit along with the operation as part of a pipeline made of a certain material. The destructive test coefficients for the examined ball valves were experimentally determined. Efficiency preservation coefficients were identified that are the basic coefficients of the dependability indicator, that, in turn, is one of the primary dependability characteristics of process equipment, machines and entities.

Abstract. Aim. This paper contains a brief historical overview of the evolution of the technology dependability theory. The evolution of the concept of dependability reflects the unsolved problem of presentation of the scope and content of concepts in the dependability theory of technical objects. It is proposed to logically elaborate and deduce the terms based on the pseudophysical propositional logic. The paper presents an approach to solving the problem of introduction of the concept of an object’s intended use and deduction of alternative basic definitions of dependability. The aim of the paper is to examine the feasibility of applying the modern technology dependability theory to subsequent theoretical developments and practical application of the concept of reliability of organizations, social groups and individuals. Methods. The problem of the terminology and years-long search for the definitions of dependability consists in the deficiency of the academic development of the subject matter in philological, philosophical and logical terms. Certainly, such research is to be conducted by experts in the appropriate fields of knowledge. Let us make our own contribution as regards the subject matter of this paper. The author suggests a structural approach to terminological research. Essentially, it consists in the following. If identifying the signs of the concept content is complicated, structuring the concept scope may be an option. The structuring is done using universal observation bases: time, space, groups and their combinations: time-space, time-group, spacegroup. For that purpose, a special terminology is required. The category of “intended use” as an object’s property is introduced. The concept of intended use is large in scope, is more abstract than the concept of dependability. Let us note that quality standards were developed under the assumption that the intended use is the compliance of an object’s characteristic with the requirements. Russian standards prioritized the dependability concept, where the regulatory descriptions, definitions, such as “the ability to perform the required (specified) functions, (an object’s) ability to function”, “to function as and when required”, “functional dependability”, “parametric dependability”, “requirements specified in the documentation” are simply generalized by the category of intended use Such descriptions are none other than an indication of the property of an object’s intended use. For instance, an object’s ability to move in space is a property of the intended use, not dependability. Thus, all the terminological searches in terms of dependability standardization demonstrate an unjustified reduction of the concept of intended use to the concept of dependability. The introduction of the category of intended use solves the problems of terminology in the dependability theory. The author suggests the following definition of intended use. Intended use is the property of an object defined by the natural origin or designed application. Dependability is a set of states as the measure of concordance with the intended use of an object. Conclusion. The evolution of the concept of technology dependability reflects the unsolved terminological problem in the dependability theory of technical objects. The problem of terminology largely consists in the ambiguous use and confusion of ontological terms. Deduction of such terms based on pseudophysical logic and introduction of the category of object’s intended use is the main result of this paper in terms of the introduction of an alternative noncontroversial structure and content of dependability-related terms. The suggested approach is recommended to be used for revision of the existing standards.

Abstract. The Aim of the paper is to analyze and examine the experience of the railway companies of the European Union in designing technical asset management systems based on advanced digital technologies. Railway companies are interested in the development of efficient maintenance and repair strategies that allow increasing the volume of traffic with a high level of safety and reliability using the Big Data produced by diagnostic systems. Methods. A comparative analysis of the best practices by European railway companies was performed, the employed digital technologies were compared, and the best known and commercially available software solutions for constructing an asset management system were reviewed. Findings/ Conclusions. Railway companies will have to make a lot of effort in order to not let the vast current expenditures associated with the digitization go to waste, since the deployment of new technology will meet the resistance of the existing system of management and allocation of responsibilities between levels of management within a company. The generic architecture of the European asset management information framework is a highly diverse range of IT solutions, which is a great challenge, as any modification to the operation of the software system requires significant time, managerial and financial resources. In this context, most successful are the companies that have invested in the development of own digital asset management frameworks.

Abstract. Aim. To show a method of overcoming the uncertainty in the requirements for the quality of data in non-standard situations and ways of formalizing the decision-making process aimed at ensuring safe operation of structurally complex systems. The paper proposes a method of axiomatic construction of integrated indicators that describe the properties of a system and its operational environment through the synthesis of the risk function. Methods. Methods of system analysis of the objective, Russman’s methods of the difficulty in achieving the objectives and the Shewhart charts theory. Results. The author proposed methods of qualitative estimation of two types of safety state, i.e. “better than” (for the purpose of defining a certain target level that characterizes the safety state that is to be ideally achieved) or “not worse than” (for the purpose of defining a certain maximum allowable level that characterizes the safety state, below which it is not allowed to go), that imply certain ranges of deviation from the specified target or, respectively, the minimal allowable levels, within which the safety state evaluated with an integrated index is deemed to be acceptable. Conclusions. It is shown that, in respect to problems of safety and risk assessment of structurally complex systems, one should not try to work with specific safety-related events only. All such events are characterized by a set of properties and contributing factors with associated characteristics. One should try to identify each property and each characteristic of such property, which would later allow defining proactive and reactive control actions in response to changes in such characteristics and properties. Having worked out a property of a situation or an event, we work out a property of a risk, and it is of no significance in which specific risk this property manifests itself. Combinations of risk properties can be extremely numerous, therefore it is very difficult to predict specific situations. That causes the requirement for a proactive decision support system that ensures high-quality managerial decisions short before a critical event.

Abstract. Aim. Obtaining initial data regarding the attitude of the Saint Petersburg residents to the emergence of unmanned vehicles in public transportation, identifying possible barriers and advantages, people’s readiness to start using innovative vehicles. Methods. The research uses the methods of statistical analysis, polling, mathematical analysis. The paper presents the method of the research and makes hypotheses that it is to confirm or refute. Results. The paper highlights the correlation between the attitude of the surveyed to unmanned passenger transportation and how often they use public transportation. The authors identify the advantages and disadvantages of deployment of urban unmanned transportation based on the poll results. Conclusion. The paper highlights the city residents’ attitude to the introduction of unmanned transportation. The results of a survey of Saint Petersburg residents aimed at revealing their fears and readiness to use unmanned urban public transportation. Those surveyed prefer not to be the first who try the innovative transport, and intend to wait for the practical experience and other people’s opinion. Most importantly, people are worried about faults in the system and in its communication both with passengers and other road users. Legal issues and software vulnerability to potential cyberattacks are mentioned as well. In terms of advantages the surveyed noted that the presence of unmanned vehicles will improve the observance of traffic regulations, reduce congestion and the risk of traffic accidents.

Abstract. Aim. Out of the currently used methods of ensuring cyber security the most productive ones are traffic analysis, malware detection, denial of unauthorized access to internal networks, incident analysis and other methods of corporate perimeter protection. The efficiency of such methods however depends on the timeliness and quality of threat data. The Aim of the paper is to study the ways of improving the cyber threat awareness and capabilities to analyze texts in open sources for the purpose of cyberattack prediction, identification and monitoring of new threats, detection of zero-day vulnerabilities before they are made public and leaks are discovered. Methods. Publicly available knowledge on cyber security is acquired through continuous collection of data from the Internet (including fragments of its non-indexed part and specialized sources) and other public data networks (including a large number of specialized resources and sites in the TOR network). The collected texts in various languages are analyzed using methods of natural language processing for the purpose of extracting entities and events that are then grouped into canonical entities and events, and all of that information is used for continuous updating of a subject-matter event-entity ontology. It includes general forms of entities and events required for the context and specialized forms of events and entities for purposes of cyber security (technical identifiers, attack vectors, attack surfaces, hashes, identifiers, etc.) Such ontology can function as a knowledge base and be used for structured queries by cyber security analysts. Results. The proposed method and the system based upon it can be used for analyzing computer security information, monitoring, detection of zero-day vulnerabilities before they are made public and leaks are discovered. The information retrieved by the system can be used as highly informative features in statistical models. The latter served as the basis for a classifier that defines the risk of exploits for a specific vulnerability, as well as an IP address scoring system that can be used for automatic blocking. Additionally, a method was developed for risk-based ranking of events and entities associated with cyber threats that allows identifying – within the abundance of available information – the entities and events that require special attention, as well as taking timely and appropriate preventive measures. Conclusion. The proposed method is of direct practical value as regards the problems of analytics, risk-based ranking and monitoring of cyber threats, and can be used for the analysis of large volumes of text-based information and creation of informative features for improving the quality of machine learning models used in computer security.

Abstract. Currently, ensuring the industrial safety of hazardous industrial facilities involves – along with conventional oversight – the risk-oriented approach that is significantly more flexible. The procedure of quantitative estimation of an accidental risk for hazardous industrial facilities is essentially one of the procedures of conformity assessment, as it includes the comparison of the risk indicators obtained by means of calculation (or expert assessment) with their standard values. The Aim of the paper is to define the problem of uncertainty that is associated with all the stages of quantitative estimation of an accidental risk, make a brief historical account, analyze its types and sources, describe the approaches employed as part of quantitative estimation of this uncertainty. Currently, it is accepted to identify the terminological, parametric and model types of uncertainty, whose examples are provided in the paper. Analysis shows that a fourth – computational – type should be added, whose contribution in many cases may be considerable. It is shown that, due to a number of circumstances, scalar numbers that are normally used for defining parameter values of the physical-mathematical models of failure processes are in reality mere indicators of the ranges of their value variation. Currently, uncertainties in the values of accidental risk parameters are accounted for using probabilistic and deterministic approaches, as well as fuzzy numbers. Methods. For the purpose of quantitative estimation of uncertainty, the paper employs the method of interval analysis. In the most general case, without using the hypothesis on the behaviour of a parameter value within the range of its possible variation, the parametric uncertainty can be defined with an interval number. In that case, all the required calculations are performed using interval methods. The natural (naive) version of interval analysis has a serious drawback that consists in an unjustified increase of the width of the interval number deduced by means of interval calculations, if one or more input parameters of the model enter into the calculation formula more than once, or the input parameters are functionally interdependent. Modern interval analysis employs methods allowing to alleviate this effect. They are briefly described in this paper. It is shown that if statistical information is available on the behaviour of parameter values within their variation intervals, the results of interval calculations of the accidental risk indicators can be significantly improved. The suggested method of reducing the computational uncertainty of quantitative estimation of the accidental risk in the interval setting is illustrated with a numerical example of risk indicator calculation for the “fireball” accident scenario. The paper sets forth the results of interval calculation of an individual accidental risk for an explosion and fire hazardous facility “reservoir with a flammable liquid” in three ways: a) naive; b) accounting for the effect of parameter correlation; c) additionally, accounting for available statistical information. Conclusions. Interval methods allow not only taking into consideration the presence of uncertainty in the accidental risk parameters, but evaluating it quantitatively. There are efficient methods of alleviating the negative.