Improvements in the Brazilian agricultural products logistic network using a railway design method and a GIS-based optimization model

Abstract

Improvements in the Brazilian agricultural products logistic network using a railway design method and a GIS-based optimization modelBrazil is a country of continental dimensions, so it is very frequent the need to transport products through distances of over 2,000 km. Historically, freight transportation is heavily based on road mode, although it is not the most economical and environmentally efficient mode for long distances. In the past few years, the Brazilian government has focused on a more balanced and rational transportation matrix, with several planned projects aimed at linking high production areas to exporting ports and not many projects directed to the domestic market. As a result, Brazil's transportation matrix for export markets is more balanced, while freight transportation to the domestic market is done almost exclusively by road. This is an adverse scenario for Brazilian industries that depend on raw materials produced in other regions of the country, such as the poultry and pork sectors in the states of Santa Catarina and Rio Grande do Sul. These Southern states must procure maize especially from the Center-West states, traversing by road distances that can easily reach more than 1,000 km, as currently there is no railway linking these states. The higher demand occurs in months with higher freight values and lower truck availability, which could be alleviated by using a storage strategy; however, these states face a deficit of static storage capacity. Hence, this thesis aims to design alternatives for the improvement of the Brazilian logistics network related to domestic market of agricultural products in a more sustainable way, by proposing a two-level optimization model for transportation and warehouse networks, integrated with a Geographic Information System (GIS). The proposed method designs a new railway to link the Center-West and the South regions; selects the location of warehouses that need to increase the static storage capacity and determines the necessary volume for them; assesses the most used transportation infrastructures; and finds the most promising entry points to import maize from Paraguay, Argentina, and other countries. The objective function that was developed took into consideration the minimization of the total cost and the CO2 emissions. In addition, this thesis proposed a railway network design model to project a new railway that includes the criteria to define its start, intermediate, and end points and to determine the area of influence and the potential cargo. The proposed new railway generated positive results, attracting substantial cargo volume and generating a more balanced and rational transportation matrix. The positive effects increased when the new railway was combined with the optimal increase of the static storage capacity of the bulk warehouses. In addition, it was possible to compare the improved Brazilian transportation matrix with the ones from other countries of similar size. The obtained results can subside policy makers and railway designers and rein...