2026-07-22T13:39:09Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:4244 2024-03-06T10:29:20Z ddc:004 open_access

Solving a Freight Railcar Flow Problem Arising in Russia Sadykov, Ruslan Lazarev, Alexander A. Shiryaev, Vitaliy Stratonnikov, Alexey Freight routing multi-commodity flow column generation We consider a variant of the freight railcar flow problem. In this problem, we need 1) to choose a set of transportation demands between stations in a railroad network, and 2) to fulfill these demands by appropriately routing the set of available railcars, while maximizing the total profit. We formulate this problem as a multi-commodity flow problem in a large space-time graph. Three approaches are proposed to solve the Linear Programming relaxation of this formulation: direct solution by an LP solver, a column generation approach based on the path reformulation, and a ``column generation for extended formulations'' approach. In the latter, the multi-commodity flow formulation is solved iteratively by dynamic generation of arc flow variables. Three approaches have been tested on a set of real-life instances provided by one of the largest freight rail transportation companies in Russia. Instances with up to 10 millions of arc flow variables were solved within minutes of computational time. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Ruslan Sadykov and Alexander A. Lazarev and Vitaliy Shiryaev and Alexey Stratonnikov 2013 Is Part Of OASIcs, Volume 33, 13th Workshop on Algorithmic Approaches for Transportation Modelling, Optimization, and Systems (2013) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/OASIcs.ATMOS.2013.55 urn:nbn:de:0030-drops-42448 https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.ATMOS.2013.55 eng https://creativecommons.org/licenses/by/3.0/legalcode