This article describes an algorithm developed by the authors for constructing a commuter train turnaround schedule. The goal is to minimize the total downtime of locomotive crews while waiting for trains to depart. This optimization results in a reduction in the required crew complement and operating costs for organizing service on sections with high commuter train traffic. The description of the proposed algorithm is formalized in the form of a flowchart that defines the logic for assigning locomotive crews. The model incorporates the following constraints: minimum allowable turnaround time, as well as established work and rest standards for personnel.

This paper begins a series of publications, making a case and opportunity for creation of advanced diagnostics of technical condition of the railway rolling stock and provides essentially new artificial intelligence (AI) solution based on deep learning techniques. Here, the abstraction for anomalous data is formed, statements of industrial equipment failure detection and forecasting problems are set, core components of software implementation of the solution algorithm using the recurrent network (LSTM autoencoder) are described as well as the quality assessments and the reasoning for choosing deep learning over the rest AI methods.

The article is devoted to the study of the possibilities of using intelligent data analysis methods and bioinspired optimization algorithms in the management of returns in e-commerce. The paper proposes the architecture of a decision support system (DSS) that combines an earthworm algorithm for multi-criteria optimization of return routes and a fuzzy random forest for classifying the causes of returns. It is shown that the use of bioheuristics makes it possible to form stable and balanced routes taking into account the distance, time and complexity of the logistics network, and the use of a fuzzy random forest provides an interpretable analysis of subjective and incomplete customer data. The proposed DSS architecture demonstrates high flexibility, scalability, and the ability to integrate with logistics platforms and WMS, providing comprehensive support for return flow analysis. The economic efficiency of the implementation of the system is considered, including reducing logistical costs, reducing the number of unjustified refunds and improving the quality of customer service.

For the successful implementation of a logistics operation, it is necessary to predict the possibility of unforeseen situations. They can cause undesirable costs. It requires making decisions that help to overcome situations. However, such decisions often have to be made in conditions of incompleteness and uncertainty of the available information. The variability of the environment requires constant updating of information about the factors affecting the implementation of logistics operations. It is almost impossible to get a complete picture of all these factors, so experience and knowledge play an important role in making decisions. However, the applicability of knowledge about solving a particular situation is limited by the factors that were observed at the time of implementation. In order to overcome this disadvantage, the article proposes the development of a model for transferring useful knowledge from precedents to a new situation using the example of the problem of determining the zones of influence of anomalies in a logistics project. The use-Case-Based Reasoning method is used to implement the model. It allows you to solve a new problem by applying or adapting a previously used solution. Logistics projects are described by geoinformation models.

Based on the study of the issues of complex optimization of the parameters of the locomotive fleet, the expediency of including indicators of labor productivity of locomotive crews (LB) in the criteria for the quality of train operation management is shown. The formulation of criteria for optimal LB labor productivity in the context of managing the operation of the locomotive fleet at the landfill is proposed. It is shown that the solution to the problem of optimizing LB labor productivity is closely related to the task of finding optimal ways of traction maintenance of landfill sites within the framework of the accepted model of locomotive operation. Special attention is paid to the relationship between the hourly output of LB and the degree of development of the functionality of the train driving system in autopilot mode. An approach to determining the rational relationship between the productive power of labor and its intensity is outlined, and its relationship to the assessment of the proportionality of the use of manual and unmanned train driving modes is highlighted.

The article is devoted to the development of methods for testing the road video surveillance subsystem as part of intelligent transport systems. The main focus is on creating an approach that allows for the confirmation of the subsystem's ability to perform its main functions, which include the generation, transmission, and display of visual information about the road situation. A methodology has been proposed based on the definition of detection, recognition, and identification zones for surveillance objects (pedestrians, vehicles, and license plates) while maintaining the minimum required technical specifications for the equipment. The results of field tests conducted in a pilot zone in the Republic of Tatarstan have allowed for the quantification of surveillance zones and the assessment of metrological error parameters. The developed method ensures the reproducibility of results and can be used in the design, commissioning, and operation of road video surveillance systems.

The article presents a comprehensive analysis of the evolution, current state and prospects for the development of control and diagnostic systems for railway rolling stock. A detailed review of global trends has been conducted, including a step-by-step transition from local sensors to integrated intelligent and robotic complexes based on machine vision, laser scanning, predictive analytics and artificial intelligence technologies. Special attention is paid to domestic developments, in particular, the Integrated Post for Automated Reception and Diagnostics of Rolling Stock and its development. The principles of the hybrid data processing model, the architecture and functionality of the PPSS, as well as its modular expansion, including the PAK-M and Element systems, are described. Approaches to the integration of diagnostic systems into information and control complexes of stations, the implementation of predictive maintenance and the creation of robotic diagnostic clusters are considered. The article is addressed to specialists in the field of railway transport, diagnostics, automation and information technology.

The article provides a comprehensive analysis of the safety of autonomous vehicles as a promising solution to the global problem of road traffic injuries. Based on up-to-date data for 2023-2025 from international sources (NHTSA, Waymo, AAA), a comparative assessment of the accident rate of unmanned and traditional vehicles is carried out. The technological potential of autonomous systems to eliminate the human factor, which is the main cause of an accident, is considered in detail. The key barriers to mass adoption are being investigated: technical limitations, cybersecurity issues, legal uncertainty, and a low level of public trust. In conclusion, strategic directions for the safe integration of technologies are formulated, including the development of a regulatory framework, increasing transparency of testing and improving artificial intelligence algorithms.