Mathematical Optimization of the Production of Fuel Pellets from Residual Biomass

One of the main concerns of mankind in the last years is the availability of energy sources, which should have the lower environmental impact possible. In the state of Guanajuato, Mexico, clay industry burns each year about 15.000 m(3) of fuel oil and residual oils, and 96.000 tons of non-sustainable wood derivatives. In particular, wood derivatives are required in high quantities due to their low density and high humidity, with costs of about 3,4 USD/GJ. Two start-up companies, GEMAR & TODO PELLET, produce solid fuel pellets from vegetable residual material. This raw material is obtained from agribusiness and from the cities nearby. The solid biofuel pellet has high density, low content of humidity and a homogeneous shape. This pellet has an energy density of about 12 GJ/m(3), which is 2.3 times the content of commonly used wood. To make the production of solid biofuels economically feasible, special care must be taken about the location of the production plants and hubs, to avoid excessive distances between the sources of the biomass and the facilities. Furthermore, to have a truly environmentally friendly fuel, the supply chain and the production process must minimize the global environmental impact. Thus, in this work a mathematical programming model to determinate the optimal location of the production plants, the collection centers, together with the optimal distribution logistics, is proposed. The multi-objective optimization problem involves the minimization of total annual costs and emissions of equivalent carbon dioxide for the whole supply chain. The problem is modeled following a general disjunctive programming approach, and then relaxed into a MINLP problem using the convex hull strategy. The problem is then solved using the GAMS software. The results show that, for the different scenarios analyzed, the optimal solution was always the same. The location of the main plant and the secondary plants has been determined, and it has been found that most of the raw material (40%) must consist on stover and straw, while the rest of the material requirement is satisfied by residues from pruning and from the agave industry. With this solution, reductions on about 18,600 ton of CO2 are expected for the main plant.