Dynamic Optimization of Railcar Traffic Volumes at Railway Nodes
Aleksandr Rakhmangulov,Aleksander Sładkowski,Nikita Osintsev,Pavel Mishkurov,Dmitri Muravev
DOI: https://doi.org/10.1007/978-3-319-51502-1_10
2017-01-01
Abstract:A major direction in the development of modern world transport systems involves the concentration of freight and traffic flows within international transport corridors and transport nodes in terminals and hubs. The changing role of rail transport is taking place under these conditions. Increased structural complexity and irregularities in cargo and railcar traffic volumes have been observed, despite the higher levels of transport equipment and technology standardization, the increased container transport volumes and consequent reduction in the cost of intermodal operations, and the interaction between different modes in transport nodes. This is largely due to the increasing need for cargo owners to lower logistics and warehouse costs, which is achieved by reducing the size of freight shipments and ensuring their uniform delivery. Moreover, privatization of the railway industry in certain countries and the sale of rolling stock to operating companies have made the coordinated management of rail fleets more difficult. The demand for improved efficiency of railcar traffic volume management in the case of complex structures is especially relevant for large railway nodes, particularly the transport systems of industrial enterprises. Here, the application of traditional approaches to the management of transportation processes involving individual elements of traffic flow (trains, railcars, locomotives) and transport infrastructure (railway stations, loading areas, rail hauls) leads to additional transport costs as a result of the increased length of time that railcars are located at the railway node. The aim of this study, therefore, is to provide improved methods for the management of RTVs at complex railway nodes based on a systematic review of RTVs in conjunction with transport infrastructure and traffic control systems. The authors review the case of a systematic approach to the organization and management of railcar traffic volumes, both for mainline rail transport and at railway nodes and industrial rail transport. This study investigates the impact of irregular railcar traffic volumes on railway node functioning by applying a dynamic simulation model of the transport system of a large metallurgical enterprise. The application of the original methodology for assessing the amount of information in the operational management system for rail transportation allowed us to estimate the effect of structural complexity and railcar traffic volume irregularities on the efficiency of the dispatch control system. The authors propose a new system of parameters and indicators for assessing railcar traffic volumes, taking into account factors of complexity and irregularity, and provide a comprehensive assessment of its effectiveness for railcar traffic volume management. For this purpose, a method of dynamic optimization of these parameters (dynamic programming) is selected. The main hypothesis of the study is that improved accuracy in parameter optimization for irregular railcar traffic volumes is achieved by adjusting the duration of the base periods which constitute the optimization period in the dynamic problem. In this study we have formulated a mathematical model for dynamic parameter optimization of railcar traffic volume on the basis of base periods of variable duration, with an algorithm for model implementation as well. Experimental verification of the effectiveness of the developed model and algorithm is conducted on the constructed process-centric railway node simulation model. Three series of experiments are conducted: without railcar traffic volume optimization, and with dynamic optimization of railcar traffic volume with the application of base periods of constant and variable duration. The experimental results demonstrate increased optimization accuracy with the use of the proposed model, reducing transport costs (time of railcar traffic volume handling) at the railway node by 11% on average. For the realization of the model and algorithm, a method is proposed for their integration in information management and intellectual transport systems.