Transportation and Information Technologies in Russia

Background. The implementation of the Mobility as a Service (MaaS) concept is a strategic direction for the development of transport systems, especially relevant for dynamically developing regions with high traffic loads, such as the Krasnodar Territory. The success of the MaaS implementation is determined not only by technological aspects, but also by the willingness of the population to accept the new mobility model, the key indicator of which is user loyalty.

The purpose is to identify and analyze the key factors of forming user loyalty to the MaaS system in the process of its implementation in the Krasnodar Territory and to develop mechanisms to overcome the main barriers, primarily distrust.

Methodology. The study used system analysis, SWOT analysis to assess the potential and risks of implementing MaaS in the region, as well as modeling based on classical loyalty theories (such as the NPC model and the Oliver model).

Results. The high but unrealized potential of the Krasnodar Territory for the implementation of MaaS has been identified, due to the transport problems of cities, a diversified transport ecosystem and a high level of digitalization of the population. The key barrier to building loyalty is the distrust of users. A three-level trust management model is proposed as a solution, including technological (encryption, transparent billing), organizational and legal (ombudsman, security deposit) and communication (open dialogue, ambassadors) mechanisms.

Practical implications. The results of the study can be used by regional and municipal authorities, potential operators of MaaS platforms and transport operators in the planning, development and implementation of integrated transport solutions in the Krasnodar Territory and other regions with similar conditions.

Background. The study provides a rationale for using hybrid approaches that combine multi-criteria analysis methods (AHP, TOPSIS, PROMETHEE) with modern data processing technologies to design adaptive interactive decision support systems (DSS). These approaches allow for automatic weighting of criteria and efficient processing of large amounts of information under uncertainty. The goal of this type of task is to find the optimal ranking of alternatives in multi-criteria problems, where the system dynamically adapts to changing user preferences and external conditions, ensuring a balance between accuracy, speed, and interactivity. The paper presents the architecture of a hybrid DSS model, the functions for evaluating the closeness to the ideal solution (in TOPSIS) and the matrices of pairwise comparisons (in AHP), and the results of a comparative evaluation of the effectiveness of the hybrid approach compared to traditional static MCDA methods in terms of accuracy and computation time when processing large data sets (with a volume of > 106 records). It has been shown that the proposed approach reduces decision-making time by 25–35% and increases the accuracy of ranking by 15–20% compared to the isolated use of multi-criteria analysis methods.

Purpose. Improving the efficiency of decision-making in complex organizational systems by using hybrid methods of multi-criteria analysis and integrating modern data processing technologies for strategic and operational planning tasks.

Materials and methods. The main research method is economic-mathematical and system analysis. The paper uses a hybrid approach that combines multi-criteria analysis methods (AHP, TOPSIS, PROMETHEE) with big data processing technologies to solve problems of ranking alternatives in interactive decision support systems. The article is based on a range of sources, including scientific literature on decision-making systems, publications on multi-criteria analysis, conference materials, statistical data on the application of DSS in logistics, finance, and healthcare, as well as documentation on software tools (Python, Scikit-learn, and Tableau).

Results. The article discusses in detail the principles and architecture of adaptive interactive decision support systems that integrate multi-criteria analysis methods with modern data processing technologies. It is shown that the hybrid approach provides dynamic adaptation of criterion weights and efficient processing of large amounts of information in real time. The obtained data, including a comparative analysis of MCDA methods, the model architecture, and the test results based on examples from logistics, finance, and healthcare, can be effectively used by organizations when designing and implementing DSS to improve the accuracy, speed, and transparency of decision-making processes in uncertain environments.

Background. Modern machine vision systems largely rely on convolutional neural networks, which demonstrate high performance in image analysis due to their ability to extract hierarchical feature representations. Improvement in recognition quality is commonly achieved by increasing network depth and the number of parameters; however, this approach is associated with higher computational costs, increased sensitivity to noise, and reduced robustness to variations in scale and illumination. This circumstance highlights the need for architectural solutions that enable more efficient feature extraction without a significant increase in computational complexity. A promising direction involves the use of multiscale convolutional operations that capture both local and global image context, as well as the integration of attention mechanisms that provide adaptive concentration of computational resources on the most informative regions and features.

Purpose. To formulate an architectural approach to the design of convolutional neural networks based on the combination of multiscale convolutional processing and attention mechanisms, aimed at improving the efficiency of machine vision algorithms while maintaining an acceptable level of computational cost.

Materials and methods. The article examines three convolutional neural network architectures, including a baseline CNN, a multiscale CNN, and an extended model incorporating an attention mechanism. Operations of two dimensional convolution, batch normalization, and nonlinear activation are analyzed, along with methods for aggregating features obtained at different spatial scales. An analytical assessment of computational complexity is performed with consideration of the number of parameters, network depth, and the volume of operations during the forward pass.

Results. The application of multiscale convolutional processing expands the receptive field of the network and improves robustness to variations in object size through simultaneous analysis of features at different levels of detail. The integration of attention mechanisms provides adaptive redistribution of feature weights, reduces the influence of noise and irrelevant components, and enhances the selectivity of image analysis.

Background. Modern Russian cities with millions of residents face the need for digital transformation of passenger transport. In the context of population growth and increasing traffic load, digitalization is becoming a key tool for creating an intelligent transport environment. A comparative analysis of the two largest cities in the Southern Federal District - Krasnodar and Rostov-on-Don - reveals different approaches to the implementation of digital technologies in the transport sector.

The purpose is to conduct a comparative analysis of the level, mechanisms and practical effects of digitalization of passenger transport in Krasnodar and Rostov-on-Don.

Methodology. The study used comparative and system analysis. A multi-level criteria system was used for the assessment, including electronic fare payment, passenger information systems, trip planning tools, and integration with traffic management systems.

Results. Two different models of digitalization have been identified. Krasnodar implements a departmental model focused on creating a convenient payment ecosystem (ETC, bank cards, NFC). An integration model is used in Rostov-on-Don, where the digitalization of transport is part of the citywide management system and is aimed at optimizing traffic flows through deep integration with automated control systems. The integration model demonstrates a higher strategic potential for the development of smart urban mobility.

Practical implications. The results of the study can be used by city governments, transport operators and developers of digital solutions in the planning and implementation of digital transformation programs for passenger transport.

This paper compares the effectiveness of support vector machine (SVM), convolutional neural network-long short-term memory (CNN-LSTM), and support vector machine-long short-term memory (SVM-LSTM) models for traffic flow prediction in intelligent transportation systems (ITS). This research aims to explore the application scenarios of different machine learning and neural network models in traffic flow forecasting, focusing on verifying the effectiveness of the CNN-LSTM and the SVM-LSTM models designed in this paper in integrating spatial feature extraction with time series modeling. Experimental validation is conducted on real-world long- and short-term traffic flow datasets, and the performance of each model is systematically evaluated in terms of the number of prediction errors, computational efficiency, and robustness. Through a comprehensive analysis of metrics such as the coefficient of determination (R2) and root mean square error (RMSE), this research provides a basis for the appropriate selection of prediction models in ITS and offers theoretical support for future research in multimodal traffic data fusion modeling.

Purpose. Through systematic comparative studies, a more efficient and reliable model is screened out for the traffic flow prediction subsystem in ITS, and the effectiveness of the hybrid model in integrating multi-dimensional features is explored, thus providing an empirical basis for further optimization of model accuracy in the future.

Materials and methods. This research used a long-term traffic flow dataset from France and a short-term traffic flow dataset from Italy. Prediction experiments were conducted in the MATLAB environment using support vector machines (SVMs), CNN-LSTM models, and an SVM-LSTM model with a loss function. The method for determining model effectiveness is based on linear regression theory, focusing on calculating the number of error data and evaluating the data fit using metrics. The method for determining model effectiveness is based on linear regression theory, focusing on calculating the number of error data and evaluating the data fit using metrics.

Results. Experimental results based on real-world traffic flow datasets show that the SVM-LSTM model exhibits the best overall performance in long-term traffic flow prediction. The CNN-LSTM model demonstrates excellent time series modeling capabilities in short-term traffic flow prediction. In terms of computational efficiency, the SVM-LSTM model improves prediction accuracy by 10.2% compared to the CNN-LSTM model. Therefore, the fusion of SVM and LSTM combines the advantages of spatial feature extraction and time series modeling, and its deployment in ITS can improve traffic flow prediction efficiency.

Background. The study substantiates the need to introduce a conceptual and model-based framework capable of capturing the evolutionary dynamics of the transport security system. The ongoing transformation of the transport sector is accompanied by changes affecting infrastructure, technical means and the distribution of functions between human and technological components, which leads to the emergence of new principles of security governance. In this context, there is a growing need for a maturity-level analytical model (C0-C5) that makes it possible to trace the trajectory of this evolution and identify the structural features of the transition from regulatory procedures to reflexive forms of management. The study also introduces the concept of the “state of security singularity” as the limit state of maturity of the transport security system, achieved through the shift toward self-regulation and the integration of technical means, analytical modules and human participation into a unified reflexive-integrated loop.

Purpose. The study aims to conceptualize the state of security singularity and to develop a maturity-level model (C0-C5) that reflects the transition from traditional organizational and regulatory procedures to intellectually adaptive mechanisms of threat pre-emption.

Materials and methods. The principal research method is modeling, which was used to develop the maturity-level model (C0-C5) and to construct the reflexive-integrated loop governing the functioning of the transport security system.

The methodological foundation is based on a systemic and interdisciplinary approach, drawing upon theories of complex and sociotechnical systems, the cybernetic logic of security management, and a risk-oriented approach. The theoretical and regulatory basis of the study includes legal acts of the Russian Federation, ICAO and EASA documents, as well as strategic materials of the Ministry of Transport of the Russian Federation and the International Transport Forum. The research methods additionally include system-structural and normative-legal analysis, content analysis of strategic documents, and logic-prognostic (foresight) methods.

Results. The study proposes and theoretically grounds the concept of the “state of security singularity” as the limit state of maturity of the transport security system. It characterizes the stage at which the system reaches the level of self-reflection and self-regulation, where the mechanisms of response and threat prevention are based on a self-renewing risk management model, and the technical means, analytical modules and human participation are integrated into a unified reflexive-integrated loop. The maturity-level model (C0-C5) developed in the course of the research serves as the analytical framework for identifying the trajectory of this evolution and for assessing the degree of system readiness for further transitions. The obtained results can be used for diagnosing the level of system maturity, determining strategic directions of its development and defining competency requirements for security specialists under conditions of digital transformation.

Background. The study substantiates the use of evolutionary methods, primarily genetic algorithms, for strategic and operational transportation planning in organizational and technical systems. Genetic algorithms (GAs) perform well in multivariate problems with conflicting criteria and strict constraints. The paper proposes a GA-ML hybrid: predictive models of time, emissions, and risk generate uncertainty distributions and a “surrogate” suitability assessment, accelerating the search for Pareto tradeoffs while meeting SLAs and environmental requirements.

Purpose. Improving the efficiency of transportation management by implementing a hybrid GA-ML for multi-criteria optimization with probabilistic constraints on deadlines and CO2.

Materials and methods. Economic-mathematical and statistical modeling; genetic algorithms with adaptive operators; machine learning for time/risk/emission forecasts; criteria normalization, 1-α, 1-β probability of execution; comparison with LP, simulated annealing, and domain heuristics.

Results. A consistent improvement in the aggregate objective J(π), a reduction in the proportion of window violations and normalized outliers with fewer “expensive” recalculations, was demonstrated using the surrogate . The hybrid generates a wider and more uniform Pareto frontier, exhibits predictable sensitivity to priority weights, and resilience in stress scenarios. The resulting principles are applicable to the design and operation of route schemes, schedules, and service policies.

Background. The article provides an overview of the current state of the railway container transportation market. Statistical data on the container transportation market is presented and analyzed. The article provides an overview of the work aimed at improving the organization of container transportation by rail. A generalized scheme of container delivery using rail transport is developed. The key objectives aimed at ensuring the reliability of container delivery are identified, and the main factors affecting the speed of container delivery are systematized. A mathematical expression of the delivery time of a container for the conditions of a direct option of transshipment from one mode of transport to another and taking into account the processing and storage of cargo at the terminal. Formulated the problem of optimizing the delivery scheme of containers in multimodal transportation in the conditions of the presence of several options of container terminals, their location. At the same time, according to the conditions of the formulated task, one processing of containers at the terminal is provided. A decision support system for changing the method of transportation of container products in multimodal transportation is proposed. Conclusions are drawn about the prospects of using the proposed system by transport and logistics companies.

The purpose is to identify options for ensuring the flow and uninterrupted movement of container flows during multimodal transportation.

Methodology. The article used an analysis of the current situation in the field of container transportation, mathematical modeling of container delivery time, optimization of the container delivery scheme based on the decision to overload the container at each terminal, and formulated a decision support system for delivery management.

Results. The state of container transportation involving rail transport is determined, the analysis of studies in the field of container transportation using rail transport is performed, the directions of ensuring the reliability of container delivery are identified. A decision support system for multimodal container transportation is proposed.

Practical implications. The results obtained can be used in the railway container transportation system, as well as by freight forwarding, operator, and transport and logistics companies.