A column generation based algorithm for solving the log transportation problem

BORDÓN, MAXIMILIANO; MONTAGNA, JORGE MARCELO; CORSANO, GABRIELA

Production Research. Communications in Computer and Information Science

Springer

Lugar: Cham; Año: 2021; p. 202 - 216

In this work, a solution approach based on column generation for the log transportation problem in theArgentine forest industry is presented. This problem involves making decisions that covers aspects related to the size of the truck fleet, different types of delivered raw materials, different regional bases where trucks begin and end the routes, legal and contractual restrictions, among others, generating a problem very difficult to solve. This work is focused on the determination of the routes to be performed by the truck fleet to supply the plants with the available raw materials in the forest sites, at minimum cost. A representative case study of the Argentine forestry sector is presented and the corresponding results are detailed. The proposed approach allows solving problems with more than 500 full truck loads to transport per day (approximately 13,500 tons of raw materials) in less than 15 min of execution and with a negligible optimality gap.