Optimization of transportation in a large-scale transport network taking into account non-determined data

Abstract

Purpose: the authors develop a multi-criteria discrete model of the planning problem and organizing passenger or freight transportation routes in large-scale transport networks, which allows optimization based on non- deterministic data. Discussion: the authors present a methodology for solving the problem of planning optimal routes in a large-scale transport network, taking into account a number of economic requirements specified in the form of criteria that allow one to evaluate the found route system with non-deterministic data. Authors consider interval estimates as non- deterministic data. Consideration in the task of optimizing non-deterministic data will allow a more adequate assessment of real transport systems. The model of the presented problem is the multicriteria problem of isolating simple chains on a prefractal graph. In this model in the form of a pre- fractal graph a large-scale transport network appears, where the nodes of the transport system correspond to the vertices of the graph, and the edges represent segments of roads connecting the corresponding nodes of the transport system. The optimization criteria are the main economic and social requirements for the transport system. It is required to select such a route system that is optimal according to the given criteria so that each node of the transport system is included in at least one route. Results: the authors proposed an effective algorithm that allows automate business processes related to the optimization of planning and routes in a large- scale transport network. In addiction the authors constructed a computer implementation of the proposed algorithm on the example of the motor transport network of Russia, including 15360 transport nodes. Testing the constructed algorithm shows a decrease in computational complexity when finding optimal routes in comparison with classical methods.