Logística reversa de pneus inservíveis: modelo de otimização para decisões estratégicas e táticas

Abstract

The accumulation of waste tires is a problem of great concern worldwide, as these are non-biodegradable and low compressibility solid waste, which cause different damage to the environment and public health. In Brazil, there are laws that establish targets for the environmentally appropriate disposal of waste tires, and make tire producers and importers responsible for this activity. The tire reverse network in Brazil consists of disposal points, called ecopoints, intermediate centers, where take place tire sorting and grinding activities, and processing plants, where take place different recycling or value recovery activities. Currently, the tire producers and importers carry out the collection of waste tires, at the request of the ecopoints when they reach an established quantity. For collection planning are not considered critical information as: the ecopoints inventory capacity; the capacity and availability of intermediate centers and processing plants; and forecasts of the amount of waste tires. Reverse logistics of tires involves high transport costs, since the shape of the tires makes it difficult to compaction of the load and there is a great geographical dispersion of the ecopoints. In this sense, the opening of sorting and grinding centers can bring economic benefits by reducing transport costs and generating revenues from the sale of the steel removed from the crushed tires and the sale of reusable tires. In this research, a mathematical programming model is proposed to support the planning decisions of reverse tire logistics in Brazil, considering both strategic-level decisions (location and dimensioning of intermediate centers); and tactical-level decisions (vehicle fleet sizing, material flow between points, quantities of tires to be sorted, crushed, stocked, marketed on the secondary market and disposed in each period of time). For the computational experiments, 40 instances were considered based on real data from the State of São Paulo (Brazil). The results show that the proposed model adequately represents the reverse network and provides good solutions for the strategic and tactical decisions involved in network planning. Several heuristic methods have been developed as solution alternatives for cases where the exact methods are not efficient. The heuristics are able to find good quality solutions with low computational time. Finally, a sensitivity analysis is performed in order to show the impact of the variation of parameters such as the sale price of reusable tires in the secondary market, transportation costs and disposal costs.