Data processing methods are now used in all areas of human activity. This is explained due to the development of information technology, the increase in the number of sensors, and the computing capabilities. Efective data processing is a key to sustainable development and provides an opportunity for improvement based on reliable decision-making. In the transport industry, the development of data processing algorithms is a major priority when improving the logistics support of the equipment life cycle. This paper discusses the process of optimizing logistics routes in order to minimize costs, increase the eficiency of the use of vehicles, their accuracy and timeliness of delivery. The paper gives three basic results. Firstly, the main stages of modeling the system of cost-efective logistics transportation of equipment are determined in order to develop efective strategies for operation. Secondly, the main requirements for the development of modern data processing methods for cost-efective logistics transportation of equipment are formed, which include indicators of accuracy, timeliness, forecasting, automation, safety and eficiency. Thirdly, a comprehensive approach to the development of a new innovative data processing method for cost-efective logistics transportation of equipment is developed, which allows obtaining the following results: reducing logistics costs; reducing equipment delivery time; increasing transportation reliability; automating and adapting logistics processes. The obtained results can be used for data processing algorithmic support improvement while optimizing maintenance and repair processes of transport equipment.
Data processing methods are now used in all areas of human activity. Data analytics becomes basic tool
in digital era. This is explained due to the development of information technology, the increase in the
number of sensors, and the computing capabilities [
Data analytics methods allow solving key problems in the production and operation of equipment
[
The transport industry plays an important role in both passenger and cargo services [
Transport logistics consists not only of optimizing routes, but also of controlling cargo flows using
modern digital technologies, minimizing costs, planning transportation, and ensuring security [
The formation of data processing methods for cost-efective logistics transportation of equipment is
an extremely urgent task for modern enterprises. Timely and accurate information processing allows to
optimize delivery routes, reduce transportation costs, and increase the productivity of logistics processes
[
Information processing techniques give possibility integrating data from various sources and
performing their analytical evaluation to make optimal decisions [
In scientific articles by various authors, the following key features of logistics transportation optimization
are highlighted, which go beyond simple cost reduction and include strategic and technological aspects.
In article [
Digital technologies penetrate almost all processes and areas of economic activity. The application of artificial intelligence (AI) in logistics is no exception. Using the capabilities of AI, stakeholders in logistics can improve decision-making processes, optimize resource use and minimize environmental impact [19].
Optimization of transport routes for the purpose of cost-efective transportation of equipment should ensure transparency and accuracy of processes, which leads to better customer satisfaction due to timely and complete deliveries, as well as more eficient service [ 20]. The development of an optimal route allows to reduce the time of cargo on the road and the number of intermediate stops/transshipments, reducing the likelihood of damage. In the article [21], the construction of the shortest cyclic route is presented, which ensures the delivery of homogeneous cargo from manufacturers to consumers. The use of modern digital technologies of machine learning (ML), Internet of Things (IoT) can help optimize product supply chains, improve customer service, reduce risks and create new business models. Further optimization is carried out using process analysis to increase their core competitiveness [22].
The cost-efectiveness of logistics transportation of equipment depends on the cargo capacity. Delivery by small-capacity vehicles is profitable only if the vehicles are loaded close enough to their customers [23]. It is measured by the ratio of achieved results (timely delivery, quality maintenance) to the costs incurred.
The article [24] is devoted to the optimization of transport routes with a focus on the first and last mile of transportation. The problem considered by the authors can be attributed to the problems of micromobility. At the same time, the authors identified key areas that are relevant in the field of application of data analytics in this area. Among them are: • problems of optimization and modeling of transportation systems based on fuzzy logic, reinforcement learning, and dynamic programming; • problems of infrastructure optimization in order to reduce resource consumption and increase eficiency; • problems of sustainable development in terms of reducing emissions of hazardous gases; • problems of economic and social importance; • problems of regulatory policy and documentation support.
As an indicator of eficiency, the authors of [ 24] chose the total transportation costs that need to be minimized. The optimization problem was solved in three stages by analyzing the asymmetric transportation problem, constructing the Euler circuit, and calculating the route based on the evolutionary algorithm. Overall, the developed approach allowed for a 19% improvement compared to known methods.
The article [25] considers a solution to the route optimization problem based on the use of a genetic algorithm and the Lin-Kernighan method. The proposed method was tested on real transportation data. As a result, the authors proved that the use of this method allows reducing emissions of hazardous substances by 54%.
The article [26] analyzes the problem of planning logistics routes without the use of geospatial services and maps. The author has performed a comparative analysis for five models, including a statistical regression approach and the use of a neural network, for estimating distance matrices based on geographical coordinates. It is proven that traditional statistical models systematically overestimate distances, while the neural network-based model shows significantly higher accuracy and better matches map-based routes. Overall, the article shows that practical route planning can be implemented without geospatial services and complete maps and with a cost-efective alternative to classical solutions.
The article [27] contains the approach to route optimization in road transport systems using machine learning methods, tested on the example of the transport corridor between Morocco and France. This approach takes into account key eficiency criteria, in particular safety, cost, and duration of transportation. The authors carried out a comparative analysis of various machine learning algorithms, in particular artificial neural networks, naive Bayesian classifier, support vector machine, and others, in order to assess their suitability for solving transportation problems. The obtained results allowed to conclude that the use of neural networks provides the highest prediction accuracy and contributes to the formation of safe routes with minimization of transportation costs and time.
Logistics planning tasks are also important in the case of improving the processes of technical maintenance and repair of various types of equipment. These tasks can be solved in conjunction with the optimization of the infrastructure of repair and service hubs [28], as well as the logistical support of the equipment life cycle based on the use of advanced information technologies [29].
The conducted analysis of the literature allows to make the following conclusions. The main indicator of eficiency when solving logistics problems is transportation costs. The criterion of eficiency is the minimum of this indicator. Logistics problems are currently solved using data processing methods, in particular, based on statistical approaches, machine learning methods, and artificial intelligence tools.
The purpose of this paper is to develop a new innovative data processing method for cost-efective logistics transportation of equipment.
The data processing for cost-efective logistics transportation of equipment is a systematic set of operations for collecting, verifying, processing and analyzing information related to routes, cargo, vehicles, and resources of the enterprise. It includes the initial collection of data from various sources, such as GPS trackers, transport documents, warehouse systems, and internal databases. Then they are structured and verified to ensure the accuracy and reliability of the information. The next stage is data processing using algorithms for optimizing routes, distributing cargo, and planning the use of vehicles. In parallel, an economic analysis is carried out to assess costs and determine the most efective solutions. Based on the results of the analysis, a logistics transportation plan is formed, which allows minimizing costs and equipment delivery time. An important component of the process is monitoring the implementation of the plan and making adjustments in real time in case of changing conditions. The process also involves automation and integration of data to increase the speed of decision-making. The main and auxiliary logistics processes during transportation require modeling on basic digital technologies that ensure a high level of operational eficiency and the ways that the enterprise should implement to improve it [30, 31].
Due to its comprehensiveness and analytical approach, this process ensures increased economic eficiency of logistics processes. In general, it is a key element of strategic management of equipment transportation to the customer.
The main stages of modeling the cost-efective logistics transportation system for equipment are follows: 1. Data collection and integration, which is provided by the formation of a single database of information on routes, vehicles, equipment, cargo, warehouse stocks, and resources of the enterprise for further analytics. 2. Verification and checking of data, their reliability and completeness of information, elimination of errors and duplications, which ensures the reliability of models. 3. Analysis of transportation flows, identification of bottlenecks, optimization of routes, and determination of priorities for equipment delivery to reduce time and costs. 4. Forecasting of costs and resource needs, assessment of financial and material costs for equipment transportation, and forecasting of the load on vehicles and personnel. 5. Optimization of logistics processes, development of models that allow minimizing costs, increasing the eficiency of vehicle use, and reducing downtime. 6. Scenario modeling, assessment of various equipment delivery options, and response to unforeseen situations (trafic jams, weather conditions, and technical malfunctions). 7. Automation of the decision-making process, creation of algorithms that allow for rapid processing of large amounts of data and efective logistics decisions. 8. Monitoring and control of task performance, ensuring the ability to track route implementation and adjust plans in real time. 9. Increasing economic eficiency, maximizing the profitability of logistics operations by reducing equipment transportation costs and increasing productivity. 10. Development of strategic management, providing enterprise management with sound analytical data for planning the logistics system.
Logistics systems must develop efective operating strategies that will integrate all aspects of its activities, taking into account the impact of external and internal factors, as well as potential risks [32, 33].
Requirements for the development of modern data processing methods for cost-efective logistics transportation of equipment are presented in Table 1.
Modern data processing methods for the logistics transportation of equipment must ensure accuracy, timeliness, and full integration of information from various sources [34]. They must efectively analyze routes, forecast costs, and optimize the use of transport resources. Automation and analytical algorithms
in transportation conditions
and increasing delivery reliability
increase the speed of decision-making and reduce the risk of errors. The flexibility and adaptability of the methods allow to respond to changes in delivery conditions and cargo volumes. In general, the implementation of such approaches contributes to increasing the economic eficiency and reliability of the logistics processes of the enterprise. Digital technologies allow to increase the potential of logistics enterprises, in particular in the context of safety, minimizing the negative impact on the environment, and implementing modern technologies of Industry 4.0 [35].
To ensure cost-efective logistics transportation of equipment, it is necessary to optimize the organizational structure and operational processes using information technologies for data processing, improve personnel skills, and implement logistics process automation systems [36].
A comprehensive approach to developing a new innovative data processing method for cost-efective logistics transportation of equipment is shown in Figure 1.
Expected results from the implementation of the proposed data processing method for cost-efective logistics transportation of equipment: • reduction of logistics costs by 15–25% due to optimization of routes and resources; • reduction of equipment delivery time and increase in the accuracy of order fulfillment; • increase in transportation reliability due to risk forecasting and scenario modeling; • automation and adaptability of processes that ensure the stability of the logistics system to changes in market and external conditions.
This section considers mathematical models of data processing for solving the logistical problem of transporting spare parts of equipment. Let’s specify the research problem.
It should be noted that the object of the study is the process of technical maintenance and repair of aviation equipment. Let’s assume that these processes take place in the corresponding hubs, the structure and location of which within a given country or region are already known. The corresponding infrastructure could be built based on the use of approaches described in publications [28, 29].
In general, we propose that the generalized indicator of the efectiveness of the considered problem takes into account 4 main factors:
The use of such an eficiency indicator refers this problem to the case of multi-criteria optimization. The partial indicators must be normalized to one in value and participates in the convolution to the integral indicator.
In this case, the minimum value must be used as the eficiency criterion.
The partial indicators of the cost and duration of delivery depend on the distance from the departure point and the destination, as well as the selected type of transport (road, rail, sea, or air transport). The risk of non-delivery is the probability that the required spare part will not be delivered due to problems with transport, the human factor, the occurrence of unforeseen or catastrophic events, and others. The fourth partial parameter is associated with the occurrence of aviation hazardous events, for example, equipment failures and their backup units, increased risks of air navigation services, and others. We denote the partial indices as 1, 2, 3, and 4.
Normalization of partial eficiency indicators is performed according to the formula: =
− − , (1) where is are minimum and maximum value of -th partial eficiency indicators .
As a result of normalization, we will get that the values of the parameters will vary in the range from zero to one.
For each partial indicator of eficiency, a separate weighting coeficients are set for the convolution. In this case, the sum of these coeficients is equal to one ∑︀4=1 = 1.
The values of the coeficients can be selected by a priori or a posteriori analysis. In the first case, it is necessary to conduct a theoretical analysis of the influence of each of the parameters on the integral indicator. In the second case, it is necessary to collect statistical data, and based on the results of their processing, establish a mutual correlation. In our case, we can take, for example, the following values of coeficients: 1 = 0.35, 2 = 0.35, 3 = 0.05, and 4 = 0.25.
The integral eficiency indicator is determined using the formula
The advantage of this approach is the simplicity of implementation and the ability to adjust the priority by changing the weight coeficients. In addition, the proposed approach allows to formalize the decision-making process when providing spare parts for equipment during maintenance and repair.
Let’s specify the problem and make a number of limitations. We will assume that transportation is carried out by road transport. In this case, the risks are zero. That is, the third and fourth partial parameters will be excluded from the calculations.
The initial parameters of the model are the cost of transportation, the cost of the spare part, the geographical coordinates of the starting point and the final destination of the route, and a road map.
The cost and duration of the route can be considered as correlated values. Then the eficiency indicator (2) can be transformed into the length of the path traveled by road transport.
The problem of determining the length was reduced to approximating the route (points and in geographical coordinates) by an implicit function (, ) = 0. This function in the general case has the form
(, ) = ∑︁ Φ (, ), =1 (3) (4) where Φ (, ) is predetermined basis function, is a number of basis functions, and are unknown coeficients need to be estimated. Polynomial functions, radial basis functions, trigonometric functions, and others can be used as basis functions.
The problem of choosing an approximating function transforms to the following form (︃ ˆ(, ) = arg min ∑︁ ∑︁ Φ (, ) ‖‖=1 =1 =1 )︃2
. = ∫︁ 1 √︁2 (, ) + 2(, ) 0 |(, )| ,
After determining the approximating function, the route length can be calculated using the formula where = and = are partial derevatives, 0 and 1 are ordinate of the starting point and the ifnal destination of the route.
The developed method of data processing is shown in Figure 2.
The developed method is based on the use of three constituent elements of artificial intelligence systems: DBSCAN clustering method, Dijkstra’s algorithm, implicit function approximation.
The input dataset is the map with marked starting point and destination of the route. The image is read in RGB format, due to which the transition to a six-dimensional feature space is carried out.
The DBSCAN clustering technique was used to form possible routes. In this case, for each pixel, a decision is made as to whether it belongs to a core, non-core, or noise point. As a result, a queue of pixel points is formed, starting from the initial point of the route. According to the results of clustering, two clusters are formed, which is ensured by choosing the number of points in the neighborhood of the nuclear point and the radius of this neighborhood.
After that, for the formed queue in the main cluster, a calculation is performed according to the Dijkstra algorithm, which is used in both forward and reverse directions. As a result, a route is formed. The resulting route is approximated by an implicit function in order to calculate its length. To confirm the reliability, the calculation algorithm is repeated 10 times. Based on the results of all calculations, the shortest route is selected.
An example of the algorithm implementation is shown in Figure 3. (5) (6)
The proposed innovative data processing method for cost-efective logistics transportation of equipment allows integrating various sources of information and ensures its accuracy and timeliness. The use of analytical algorithms and machine learning contributes to route optimization, cost forecasting, and increasing the eficiency of transport resources. Adaptive scenario modeling and decision-making automation provide a quick response to changes in delivery conditions.
The paper considers mathematical methods for solving the problem of calculating the optimal route. In this case, the authors have given an integral eficiency indicator, which is calculated as the convolution of partial indicators. For a simplified model, a method for calculating the optimal route has been proposed. This method includes DBSCAN clustering algorithm, Dijkstra’s algorithm, and implicit function approximation. A computer program was developed to implement the proposed method. The paper provides an example of calculations performed in this program. The results of calculations for diferent route maps confirmed the efectiveness of the proposed method.
The considered method allows reducing logistics costs and equipment transportation time, as well as increasing the overall reliability of processes. Systematic monitoring and feedback contribute to the continuous improvement of the logistics system. Thus, the implementation of this method ensures increased economic eficiency, productivity, and competitiveness of the enterprise.
The authors have not employed any Generative AI tools. [19] M. M. Mia, S. Rizwan, N. M. Zayed, V. Nitsenko, O. Miroshnyk, H. Kryshtal, R. Ostapenko, The impact of green entrepreneurship on social change and factors influencing AMO theory, Systems 10 (2022). doi:10.3390/systems10050132. [20] N.-A.-T. Nguyen, C.-N. Wang, T.-T. Dang, Advanced process optimization in logistics and supply chain management, Processes 13 (2025). doi:10.3390/pr13061864. [21] E. Bronshtein, R. Gindullin, Exact solutions of some optimization problems of transport logistics, Mathematical Models and Computer Simulations 6 (2014) 332–336. doi:10.1134/ S2070048214030053. [22] J. Xie, C. Chen, Supply chain and logistics optimization management for international trading enterprises using iot-based economic logistics model, Operations Management Research 15(3-4) (2022) 714–722. doi:10.1007/s12063-022-00254-y. [23] R. Masson, A. Trentini, F. Lehuede, N. Malhéné, O. Peton, H. Tlahig, Optimization of a city logistics transportation system with mixed passengers and goods, EURO Journal on Transportation and Logistics 6 (2017) 81–109. doi:10.1007/s13676-015-0085-5. [24] A. Banyai, I. Kaczmar, T. Banyai, Route optimization and scheduling for asymmetric micromobilitybased logistics, Symmetry 17 (2025). doi:10.3390/sym17040547. [25] H. Naganawa, E. Hirata, N. Firdausiyah, R. G. Thompson, Logistics hub and route optimization in the physical internet paradigm, Logistics 8 (2024). doi:10.3390/logistics8020037. [26] P. Veres, ML and statistics-driven route planning: Efective solutions without maps, Logistics 9 (2025). doi:10.3390/logistics9030124. [27] N. El Karkouri, L. Hassine, Y. Ledmaoui, H. Chaibi, R. Saadane, N. Enneya, M. El Aroussi, Enhancing route optimization in road transport systems through Machine Learning: A case study of the Dakhla-Paris corridor, Future Transportation 5 (2025). doi:10.3390/futuretransp5020060. [28] M. Zaliskyi, et al., Methodology for substantiating the infrastructure of aviation radio equipment repair centers, in: CEUR Workshop Proceedings, volume 3732, 2024, pp. 136–148. URL: https: //ceur-ws.org/Vol-3732/paper11.pdf. [29] M. Zaliskyi, V. Ivannikova, O. Solomentsev, I. Ostroumov, N. Kuzmenko, The approach to optimization of the structure of the repair process of aviation radio equipment, Transport 40 (2025) 50–63. doi:10.3846/transport.2025.24012. [30] O. Arefieva, Z. Poberezhna, S. Petrovska, S. Arefiev, Y. Kopcha, Devising approaches to modeling enterprise business processes under conditions of modern digital technologies, Eastern-European Journal of Enterprise Technologies 1 (2024) 69–79. doi:10.15587/1729-4061.2024.298143. [31] I. Popovych, O. Semenov, A. Hrys, M. Aleksieieva, M. Pavliuk, N. Semenova, Research on mental states of weightlifters’ self-regulation readiness for competitions, Journal of Physical Education and Sport 22 (2022) 1134–1144. doi:10.7752/jpes.2022.05143. [32] S. Bondarenko, O. Makeieva, O. Usachenko, V. Veklych, T. Arifkhodzhaieva, S. Lernyk, The legal mechanisms for information security in the context of digitalization, Journal of Information Technology Management 14 (2022) 25–58. doi:10.22059/jitm.2022.88868. [33] A. Bieliatynskyi, S. Yang, V. Pershakov, M. Shao, M. Ta, The use of fiber made from fly ash from power plants in China in road and airfield construction, Construction and Building Materials 323 (2022) 126537. doi:10.1016/j.conbuildmat.2022.126537. [34] O. Bazaluk, O. Anisimov, P. Saik, V. Lozynskyi, O. Akimov, L. Hrytsenko, Determining the safe distance for mining equipment operation when forming an internal dump in a deep open pit, Sustainability 15 (2023). doi:10.3390/su15075912. [35] S. Smerichevskyi, Z. Poberezhna, O. Mykhalchenko, Y. Petrova, V. Smilyanets, Ensuring the sustainable development of the aviation enterprise in the context of forming innovative potential using digital technologies, in: CEUR Workshop Proceedings, volume 3732, 2024, pp. 174–185. URL: https://ceur-ws.org/Vol-3732/paper14.pdf. [36] M. Zaliskyi, O. Solomentsev, Z. Poberezhna, O. C. Okoro, B. Chumachenko, S. Chumachenko, Information technologies of data processing for linear deterioration process during aviation equipment operation, in: CEUR Workshop Proceedings, volume 3732, 2024, pp. 32–44. URL: https://ceur-ws.org/Vol-3732/paper03.pdf.