(689f) Dimensionality Reduction Techniques Coupled with Multi-Objective Optimization Applied to the Design of Bioethanol Supply Chains In Argentina

Energy security and environmental concerns have triggered the interest on bio-based sources of energy. Bioethanol, in particular, is the world's leading transportation biofuel with a worldwide production in 2009 of almost 74 billion liters ¹. This scenario has created a clear need for systematic tools to assist in the design and operation of biofuels facilities.

The overwhelming majority of the optimization studies for designing bioethanol infrastructures have concentrated on improving the economic performance. Few studies have accounted for the environmental impact of these systems. Zamboni et al. ⁴ formulated a multi-objective optimization model to reduce GHG emissions associated with the future corn-based Italian bioethanol network. Later, Giarola et al. ⁵ extended this model by considering second generation bioethanol production technologies. As pointed out in the literature^6,7,approaches focusing on minimizing GHG emissions as unique criterion are somehow limited, as they can lead to solution where this impact is mitigated at the expense of increasing others. Hence, the environmental assessment of bioethanol infrastructures requires the application of holistic methods capable of accounting for several impacts occurring throughout the entire production chain.

The aim of this work is to provide a decision-support tool for the strategic planning of integrated bioethanol-sugar supply chains (SCs) that accounts for the simultaneous optimization of several environmental impacts. The design task aims to determine the number, location and capacities of the SC facilities to be set up in each sub-region of a given country, their expansion policy for a given forecast of prices and demands over the planning horizon, the transportation links and number of trucks that need to be established in the network, and the production rates and flows of the involved feedstocks, wastes and final products.

The problem is mathematically formulated as a mixed-integer linear program (MILP). To encompass all possible conversion pathways, the proposed model includes production facilities of two types: sugar mills and distilleries. Depending on the utilized technology, the sugar mills can give two main by-products: molasses or honey, both of which can be fermented to obtain bioethanol. The liquid and solid materials require different storage conditions, so the model considers two options of warehouses depending on the physical state of the stored material. The region of interest (i.e., Argentina) is divided into a number of sub-regions, where the SC facilities can be installed. Different trucks are also considered for transporting materials between sub-regions of the country.

The environmental impact is quantified via a set of life cycle assessment (LCA) metrics⁸ that measure the environmental performance of the integrated sugar-bioethanol SC in eleven damage categories. To tackle this high dimensional multi-objective problem, we follow a combined approach that relies on the use of a MILP-based dimensionality reduction technique recently developed by the authors. This strategy allows identification of redundant environmental objectives that can be left out of the analysis, shedding light on the interactions between the environmental damages caused by the bioethanol infrastructure.

Numerical results show that several environmental effects of the sugar-bioethanol network are highly correlated, which makes it possible to focus our attention on a reduced set of damages. Our tool allows uncovering relationships between environmental impacts, providing valuable insight into the design problem and suggesting process alternatives for holistic environmental improvements.

References

1. ENERS Energy Concept Ltd.,  (2010). Production of biofuels in the world
2. Dias Leite A., (2009). Energy in Brazil: towards a renewable energy dominated system. Earthscan: London
3. Renewable Fuels Association, (2009). Ethanol industry overview
4. Zamboni A., Bezzo F., Shah N., (2009). Spatially explicit static model for the strategic design of future bioethanol production systems. 2. Multi-objective environmental optimization. Energy & Fuels 23, 5134–5143
5. Giarola S., Zamboni A. , Bezzo  F., (2011). Spatially explicit multi-objective optimisation for design and planning of hybrid first and second generation biorefineries. Computers & Chemical Engineering, DOI: 10.1016/j.compchemeng.2011.01.020
6. Niven  R., (2005). Ethanol in gasoline: environmental impacts and sustainability review article. Renewable & Sustainable Energy Reviews 9 (6), 535–555.
7. Scharlemann J.,  Laurance W., (2008). How green are biofuels? Science 319 (5859), 43–44
8. PRe´-Consultants, The Eco-indicator 99, A damage oriented method for life cycle impact assessment.Methodology Report and Manual for Designers. Technical Report, PRe´ Consultants, Amersfoort, The Netherlands, 2000.