1. Intelligent software-based decision support systems
One of the priority areas is the development of intelligent software systems capable of analyzing large amounts of data, identifying hidden patterns, assessing risks and making recommendations for management decisions. Within this area, methods of machine learning, statistical analysis, data mining, classification, forecasting and automated selection of informative features are studied.Such approaches can be applied in ERP systems, CRM platforms, B2B environments, digital analytics, production planning and enterprise resource management systems. Promising tasks are the creation of software modules for forecasting the success of orders, evaluating customer behavior, supporting commercial decisions, optimizing business processes and increasing the efficiency of automated management systems.
2. Machine learning and automated feature selection in software systems
An important area of research is the use of machine learning algorithms to build predictive models in software systems for various purposes. Particular attention is paid to the preparation of data, feature selection, optimization of hyperparameters, evaluation of the quality of models and explainability of results.Gradient boosting algorithms are studied, in particular XGBoost, feature selection methods, statistical criteria, comparative analysis of models, as well as approaches to the integration of predictive algorithms into computer-integrated information systems. This area is relevant for creating software that not only accumulates data, but also turns it into practically useful solutions for business, industry, education, the agricultural sector and other areas.
3. Computer-integrated monitoring and control software systems
Much of the research is related to the development of software for computer-integrated systems for monitoring, measuring, processing data and managing technical objects.In this area, the architecture of software and hardware complexes, algorithms for collecting and processing measuring information, software components of embedded systems, interfaces with sensors, data visualization tools and mechanisms for transmitting information to local or cloud services are considered.The practical value of the direction is to create software systems that provide continuous monitoring of theobject parameters, prompt response to changing conditions, increase the accuracy of measurements and support automated control.
4. IoT systems, edge-computing and embedded software
A separate research area is related to the development of software for IoT systems, wireless sensor networks, edge-computing architectures and embedded devices. Within this area, methods of data collection, filtering, aggregation and intelligent processing directly at the device or edge computing node level are investigated.Particular attention is paid to the creation of software components that can operate in conditions of limited computing resources, unstable communication, distributed infrastructure and the need for autonomous decision-making. Such studies are important for agromonitoring, industrial automation, environmental control, logistics, energy, smart cities and cyber-physical systems.
5. Software technologies for measuring information processing
An important area is the development of algorithms and software components to improve the accuracy, reliability and speed of measuring information processing. Methods of error compensation, digital filtration, statistical processing of measurement results, analysis of signal instability, correction of temperature drift, adaptive calibration and software support for experimental research are investigated.This area combines software engineering with metrology, sensor technologies and automated measurement systems. Its results can be used to create reliable software solutions for industry, security, environmental monitoring, energy and technical diagnostics.
6. Modeling, simulation and verification of software and hardware systems
One of the key areas of research is the modeling of complex software and hardware systems before the stage of their practical implementation. This approach allows you to assess the behavior of the system, check algorithms, identify critical modes of operation and reduce the risks of errors in real conditions.Methods of computer simulation, simulation experiment, software simulation, analysis of operation scenarios, testing of control algorithms and quality control of software components are investigated within the framework of the direction. Simulations are particularly important for systems in which errors can lead to significant technical, economic or safety consequences.Such systems include industrial facilities, sensor networks, hazardous environment monitoring systems, cyber-physical complexes and automated control systems.
7. Reliability, quality and maintenance of software
Research in the field of software quality is aimed at creating methods, models and tools that ensure stable, predictable and safe operation of software systems throughout their life cycle.Within this area, the issues of testing, validation, verification, requirements analysis, assessment of the quality of architectural solutions, code support, monitoring the performance of software components and software support after implementation are considered.Of particular relevance are software quality studies for systems that work with real-time data, sensor flows, automated decisions, and critical processes.
8. Software Engineering for Cyber-Physical and Industrial Systems
Modern cyber-physical systems combine software, sensors, actuators, network infrastructure, analytical modules and automated control tools. Their development requires an integrated approach to architecture, software implementation, testing, security, and scaling. Within the framework of this direction, methods for creating software components for industrial, agricultural, energy, transport and environmental systems are investigated.Particular attention is paid to the interaction of the software with the physical environment, real-time data processing, adaptive control and ensuring the stability of the system to external influences. This direction is promising for the development of Industry 4.0, smart agriculture, smart city, digital infrastructure and automated production.
9. Cloud services and software platforms for distributed monitoring
The actual direction is the creation of software platforms that provide data collection, storage, transmission, processing and visualization in distributed monitoring systems.Architectural approaches to building cloud and hybrid systems, data exchange mechanisms between sensor nodes and server components, APIs for integrating software modules, time series databases, monitoring panels, mobile interfaces and remote control tools are studied. Such software platforms can be used to monitor agrotechnical, industrial, environmental, transport and technical facilities.
10. Application software for data analysis and forecasting
A separate area of research is related to the creation of application software for data analysis, event forecasting, risk assessment and management decision support. Within this area, software systems using statistical methods, machine learning, fuzzy logic, neural network approaches, regression analysis, classification, clustering and methods of explainable artificial intelligence are considered.The result of such research may be software modules for predicting the state of technical systems, the success of business processes, the likelihood of undesirable events, the effectiveness of resource management or the behavior of complex objects in variable conditions.
Potential research topics for higher education students.
Applicants for higher education can join the research of the department through coursework, qualification, master's and dissertation works. The following topics can be promising:
- development of a software system for predicting the success of B2B orders using machine learning methods;
- automated feature selection in forecasting tasks for ERP and CRM systems;
- software architecture of the IoT system for monitoring the parameters of the technical or natural environment;
- development of an edge-computing module for pre-processing of sensory data;
- software system for visualization and analysis of streaming data from wireless sensor networks;
- using XGBoost to build an intelligent decision-making support module;
- development of software to compensate for errors in digital measuring systems;
- simulation and testing of the software component of the cyber-physical system;
- creation of a cloud service for collecting, storing and analyzing IoT monitoring data;
- study of the quality, reliability and performance of real-time software systems;
- use of fuzzy logic and neural network models in software control systems;
- development of a software platform for intelligent monitoring of industrial or agricultural facilities.