Development of a Computational Intelligence Framework for the Strategic Design and Implementation of Large-scale Biomass Supply Chains

Biomass utilisation has witnessed growing attention as a promising alternative to fossil fuels. However, the biomass supply chain is generally unstable due to the fluctuation of the supply and the difficulty to accommodate all the available biomass feedstock and utilisation pathways. This has made the optimisation of the supply chain a challenging task due to the complexities of modelling the different interconnected systems consisting of various production sources, intermediate processing methods, and utilisation sinks. The objective of this study is to develop a geospatial information systems (GIS)-based decision framework integrated with statistical learning and heuristic optimisation capabilities in order to establish the design solutions for biomass supply chains on a national scale and subsequent strategic decision-making. The artificial neural network (ANN)-based surrogate models for the biomass processing system were developed and validated to predict the synthesis gas flowrate and composition linked to the input parameters of biomass attributes in terms of their proximate and ultimate analysis, as simulated in previously developed Aspen Plus models. The surrogate models were then coupled with GIS, including information about biomass sources, candidate plant locations, and utilisation sinks. The optimal biomass utilisation routes were subsequently solved using the genetic algorithm (GA). The framework implementation is illustrated through a biomass gasification case study in the state of Qatar. The results demonstrate the importance of domestic biomass resources to support the local economy through the production of methanol as a final product, and utilising primarily manure-based biomass feedstock in an optimal blend with sludge and date pits. The economic optimisation of the biomass processing-plant locations reveals production routes supporting the generation of methanol and the utilisation of manure feedstock in Qatar.