|
|
 Выпуск 81
- Kazakov A., Lempert A., Le Q.M. On multiple coverings and packings problems in a two-dimensional non-euclidean space
The article is devoted to the study of two significant problems of computational geometry. The first one is the multiple circle covering problem for a closed bounded set in a two-dimensional metric space, and the second one is the multiple circle packing problem. In the first case, the objective is to minimize the radius of the circles, in the second one is to maximize it. In both cases, the number of circles k is given. The considered metric is generally non-Euclidean. The source of such a statement is tasks from transport logistics, where the distance between objects is necessary to be replaced with a minimum time to travel between them. And optimum is not always achieved with straight line moving due to the terrain or urban features. We propose computational algorithms to solve these problems. They include the joint use of an optical-geometric approach based on the principles of Fermat and Huygens and the K-means method. The key step is to construct a generalized k-order Voronoi diagram. Each Voronoi cell with a fixed set of n centroids includes points, which are closer to some k centroids than to the remaining n-k . The cells can intersect each other. Computational experiments are carried out.
- Sorokin A. Development of rules base for a fuzzy inference system in the conditions of unharmonious experts opinion
The article proposes a method of forming a rule base for fuzzy inference systems when an expert group forms incompatible rules. The method is to include incompatible rules in the rule base, taking into account the value of the confidence coefficient. The value of confidence coefficient is calculated depending on the qualifications and number of experts who have proposed it. The proposed method was researched with numerical experiments. For one series of experiments, a fuzzy inference system has been developed, the rule base for which was obtained by interviewing a group of 10 experts. Using the proposed method, the confidence factors for each rule are calculated. During the experiments, it was shown that fuzzy inference systems have the greatest ability to classify various combinations of input parameters, which based on the rule base, where experts agree on the output values of the rules containing only the maximum and minimum values, and on the other combinations of the input variables proposed inconsistent rules. The proposed method should be used for the initial assessment of the element states of socio-technical or socio-economic systems. Additionally, the research results can be used for the analysis of the expert group opinions consistency.
- Belenky A., Kornhauser A., Fedin G. Modeling the interaction of parties of a public-private partnership for the design/development of regional freight transportation infrastructure
An approach to the interaction between a regional administration and private investors based on estimating the volume of private investment needed to design/develop a regional freight transportation infrastructure is proposed. The estimates of the required investment volume and those of the amount of the revenue expected to be generated by the infrastructure functioning are key ones in negotiations of the regional administration with the private investors on forming a potential public-private partnership to finance the development/modernization of the infrastructure. A decision-support tool for estimating the required volume of private investment to this end is proposed. The tool includes a) a mathematical model underlying three problems formulated on its basis depending on the information available to the decision makers–two mixed programming problems and a minimax problem, which is proven to be reducible to a mixed programming one with all integer variables being Boolean, b) a standard software package for solving mixed programming problems, and c) a software package for processing data. The use of the tool in structuring negotiations between a regional administration and private investors is discussed.
- Farkhadov M., Petukhova N., Vaskovskii S., Farkhadova M. Improving the efficiency of speech interface with the use of cognitive and linguistic knowledge
In the modern information society a huge amount of voice data is accumulated daily, and the automation of processing can provide fundamentally new opportunities for the analysis of management decision-making automation systems, therefore the development of methods and means of automating speech information recognition and analysis systems is of great scientific and state significance. Information technology plays an enormous role in modern society. Specifically, the most natural form of communication between a man and an information environment is a speech interface. Unfortunately, not all of the results of novell theoretical and applied research are up taken to design speech interfaces. This happens because there are no easily identifiable criterias, the solutions have strong context dependence, the problems itself are hard to formalize, the problems are very labor-intensive. At the same time, this research is of high demand since speech interfaces become increasingly widespread and the developers need practical guides. In this paper we investigate the problems of speech interfaces. Specifically, we demonstrate new methods to solve some problems that arise while designing man-machine speech technology-based interfaces. Finally, we formulate the rules, that are based on our research and experience in building speech-based applications, to design effective interfaces.
- Loginovskiy O., Shestakov A., Shinkarev A. Building of modern enterprise information systems
The paper presents an analysis of modern approaches and practices that are used in the building of enterprise information systems. The problems of the isolation of practice from theoretical research are revealed. The main characteristics of big data, solutions for storing enterprise information, the applicability of machine learning, its tools and relevance in modern conditions are considered. Among other things, such necessary aspects of development of information systems as reliability, scalability, maintainability and safety are considered. As a rule, existing systems store only the current state. The event flow is proposed as a method of storing, processing and presenting data for management decision-making in conditions of instability, which is relevant to the current business needs. The reasons behind the complexity of the transition to the flow of events as a storage system are revealed. In light of the changing set of used technologies and the new focus on data analysis, the basic labor market requirements for specialists at the intersection of programming and data science are being laid down. The main skills and market requirements include work at the intersection of several areas, expertise in deploying applications in cloud services, writing clean code, knowledge of mathematical apparatus at the level necessary for rational use of machine learning models, as well as competent ability to build parallel processing in several streams. The main competitive advantages of business are flexibility and the ability to get the most out of information.
- Kourganov D. Calculation of additional oil production after well’s conversion to inlector in oil reservoir management
Machine learning, namely supervised learning models is widely used for decision making in oil field development. An essential condition for method’s application is the availability of digital databases with representative results which allows adequate model training. In this paper SVM-rank model is applied for injectivity prediction of infill wells for giant Western Siberian oilfield. Ranking algorithm also uses Voronoi diagram, proven as an approximation to the well drainage area. Complex method allows combine different reservoir and production parameters: productivity of surrounding wells, area pressure, frac parameters etc without common reservoir dynamics model, which in this particular case is not able to clarify and confirm the parameters of the reservoir system. There is double model used: the first model utilizes productivity and capacity reservoir parameters, the second one uses correlation analysis between infill candidate and surrounding production wells. The method can be particularly useful in complicated reservoirs, e.g. in dual porosity ones, where the relationship between formation parameters (permeability, porosity, saturation) and production rates is unclear and cannot be set by traditional development analysis, particularly in frac environment.
|
|
