Software & Systems

Articles of journal № 3 at 2023 year.

Order result by: Public date | Title | Authors

11. Software system for production planning and inventory management [№3 за 2023 год]
Authors: Solomakha, G.M., Tulueva, V.A., Khizhnyak, S.V.
Visitors: 10476
The paper presents the process of interaction between information systems during production planning and inventory management, taking into account the dynamic production environment and the limited capabilities of the enterprise. The paper considers the features of industrial enterprises that do not allow using standard implementations of software products and systems offered on the market. The main module for optimizing the process of planning and inventory management is an ERP system. The Forecast NOW software product of the Russian company “Ingenius Team” LLC helps in sales forecasting and safety stock level estimation; production planning is carried out in the SAP APO SCM system. The result of these systems work and interaction is a production schedule, which is then passed to the MES system for production, as well as accurate data on the needs for purchasing products, raw materials, materials, and semi-finished products. The research methodology uses a system approach, as well as methods of analysis and synthesis of domestic scientists’ studies in the field of planning and inventory management. The paper lists the requirements for software modules developed in process of system integration. It is shown that the proposed software system meets these requirements. The paper presents an example of possible integration of the developed software complex with a system for monitoring production processes, and shows the corresponding integration scheme. It is shown that the application scope of this approach is not limited to large enterprises and can be adjusted for use with appropriate software systems of smaller scale enterprises. The flexibility of the resulting software solution opens up opportunities for further expansion of the application and implementation scope of more large-scale independent production software systems.

12. Developing a GH-graph proportional separation algorithm to form of object influence zones in complex technical systems [№3 за 2023 год]
Author: Muntyan, Е.R.
Visitors: 9533
The article suggests one of the possible solutions to the problem the object influence zones in complex technical systems (СTS). It considers the extended perimeter security system as a СTS example, investigates the interaction of its objects (elements). The objects of such a system are mobile or stationary security objects, quadrocopters, decision makers, possible potential perimeter violators; provided that quadrocopters have different technical characteristics, in particular, the their video camera surveillance radii differ. To simulate the process of interaction of security system objects, the authors use a model based on a fuzzy graph with different types of vertices and multiple and different types of connections (GH-graph). Multiple connections in the GH-graph are vector connections that combine several different types of connections into one. Such model allows setting all necessary relationships between system elements and at the same time, it has an advantage in the time of calculating distances compared to graphs using only same type and different types of connections. To solve this problem, the authors propose using GH-graph algorithmic modeling means, including the graph proportional separation algorithm and the means of calculating graph metrics. The paper defines the GH-graph splitting operation, defines the separation criteria: the proportionality of subgraphs by a given parameter and the possibility of subgraph intersection. The authors have carried out the synthesis of the GH-graph proportional separation algorithm in accordance with the criteria and shown the algorithm results using the example of the graph model considered. It is shown that using the proposed graph separation algorithm into proportional subsets and the means of calculating the metric characteristics of the obtained subgraphs make it possible to determine the system object influence zones in accordance with their technical parameters. The paper considers the possibilities of software implementation of the proposed algorithm.

13. Developing a data transfer protocol based on a combined data encryption algorithm [№3 за 2023 год]
Authors: Bakaeva, O.A., Baraboshkin, D.A.
Visitors: 5771
From a technical point of view, information transfer is impossible without using data transfer protocols. One of the main requirements for such protocols is data protection. The most reliable method that ensures the protection of information transmitted over various communication channels is data encryption. The article analyzes standard encryption algorithms: AES, RSA, Diffie-Hellman protocol and SHA256 data hash function. It identifies some of their features that do not allow ensuring maximum data protection during their transfer completely. Therefore, the development of a combined data encryption algorithm, the essence of which is the use of existing algorithms at different stages of encryption, will help to avoid the problems that arise when using a single protocol. The research subject is functioning of standard encryption algorithms: AES, RSA, Diffie-Hellman protocol and SHA256 data hash function. The main result of the work is creating a data transfer protocol based on a combined data encryption algorithm. The protocol includes the development of the packet structure, the implementation of the Client Resolving and Handshake processes, as well as various types of Payload structures. At the end, the parameters of the DH (Diffie-Hellman) and AES encryption algorithms are selected. This sequence of development made it possible to make this data transfer protocol universal and efficient. The article demonstrates the protocol operation that consists of two stages: establishing a connection and data transfer. The practical significance of the work is in the fact that the developed data transfer protocol will help to ensure the completeness, confidentiality and security of any type data transfer (text, graphics, audio file).

14. Developing a chatbot to monitor a university's academic timetable [№3 за 2023 год]
Authors: Sazanov, V.A. , Khlobystova, A.O. , Abramov, M.V.
Visitors: 10979
Higher education institutions often face problems in providing digital support for the educational process, in particular, the problems associated with timely informing about timetable changes. One of the solutions to this problem can be a Telegram bot. This article describes a Telegram bot designed to improve the efficiency of notifying participants about the class schedule and its changes using the example of St. Petersburg State University (SPbGU). The research methods are based on identifying the problems of the current schedule information tool, designing a structure, and then developing a client-server application to solve them. The research subject is the tools for the interaction of participants in the educational process with an academic timetable. The main result is the implementation of the ability to quickly display the timetable with three different querying options (by a group name, by searching for a teacher and by navigating through all programs), subscribing to notifications with the current timetable of a specific group or teacher with setting the time for receiving notifications and the ability to view the requested schedule in text format or a generated image. In addition, the bot notifies a user when the timetable changes. The practical significance is in the simplification of organizational processes by expanding the possibilities of timely information and reducing the time to obtain the necessary information. The students, teachers and employees of the St. Petersburg State University, who provide the educational process, already use the developed system in their activities.

15. The system for prescribing personalized treatment by case-based reasoning using a hybrid precedent extraction method [№3 за 2023 год]
Authors: Gribova, V.V., Kovalev, R.I., Okun, D.B.
Visitors: 9748
The paper describes a system for prescribing personalized treatment based on precedents. A unique feature of the system is the extraction of precedents based on a hybrid method combining the extraction of a precedent based on knowledge with a classical method of K-nearest neighbors. The novelty of the proposed approach is in ensuring maximum flexibility and correctness in assessing the convergence of precedents. The paper describes the information and software components of the system. The knowledge base used, as well as all information resources, is built according to its ontologies, which clearly define its structure and semantics. This allows making changes to them quickly without involving programmers and processing the entire system. The system is based on a multi-agent approach. The first stage involves a preliminary calculation and bringing all features to a single metric using the knowledge base. The second stage includes actual calculating the convergence by the K–nearest neighbors method. The convergence of case histories is determined collectively for each attribute. Thus, in practice, the system allows the most flexible and accurate assessment of the similarity of medical histories containing heterogeneous signs. The proposed solution is particularly effective under conditions where there is a shortage of medical knowledge and data, when other types of systems often cannot offer a correct solution, in particular those based on knowledge.

◄ ← Preview | 1 | 2