Open AccessSEARCH FOR THE OPTIMAL SOLUTION FOR ASSIGNING SUBURBAN-URBAN TRAINS ON BRANCHED LINES ON EACH OF THE POSSIBLE ROUTES
Author(s) -
Vadim N. Shmal,
Liliya R. Aysina
Publication year - 2020
Publication title -
t-comm
Language(s) - English
Resource type - Journals
eISSN - 2072-8743
pISSN - 2072-8735
DOI - 10.36724/2072-8735-2020-14-11-39-45
Subject(s) - ranking (information retrieval) , service (business) , relevance (law) , train , computer science , genetic algorithm , selection (genetic algorithm) , operations research , novelty , function (biology) , preference , transport engineering , engineering , business , artificial intelligence , economics , marketing , machine learning , philosophy , cartography , theology , evolutionary biology , political science , law , biology , microeconomics , geography
The article describes the problem of organizing train traffic on line with branches, analyzes possible separate routes on the "trunk" and "branch" sections, and brunched routes. The development of a mathematical model which consider the interests of passengers in a service provided and economic interest of passenger companies are elements of scientific novelty of presented researching. The insufficient disclosure in the existing theoretical base of the issues of train traffic organization on lines with branches confirms the relevance of the research. The deficit of clear requirements for the organization of such lines leads to empirical decisions when operating such facilities. Adhering to the goal of providing a comfortable service for passengers and taking into account economic aspects in the interests of suburban companies, the objective function is minimize the difference between the mileage of available seats (as a condition that reflects the interests of a commuter company) and the overcrowding of the train (as a criterion that ensures a comfortable ride for the passenger). The authors suggest using a genetic algorithm for solve this problem. Genetic algorithm is a tool that allows you to perform a directed search of competitive options. The article describes the method of applying the genetic algorithm, the formation of "parent" variants and "offspring" variants, compliance with a number of restrictions and ranking of offspring variants for the selection of the best "individuals". Before testing the proposed method on a real example, the authors make a preliminary analysis to identify possible shortcomings that require improvement. The possibility of using a mathematical model and a genetic algorithm to find the optimal (or close to optimal) option for organizing traffic on a line with branches is considered on an arbitrary example. The presented calculations can be performed in any analytical software package. In this research, the search for the optimal method to organize traffic on a line with branches was performed in the MS Excel product. Successful calculations on an arbitrary option using the proposed mathematical model and genetic algorithms (as a tool to determine the optimal option of movement), allows us to assume that calculations can be made based on real source data for suburban directions of railway junctions that have lines with branches.