A GENERIC TACTICAL PLANNING MODEL TO SUPPLY A BIOREFINERY WITH BIOMASS

ABSTRACT The supply chains which bring biomass to biorefineries play a critical role in biofuel production. Optimization models can help decision makers to design more efficient chains and minimize the cost of biomass delivered to the refineries. This article based on a French national research project on biomass logistics considers one refinery, able to process several crops and several parts of the same crop, over a one-year horizon divided into days or weeks. A network model and a data model are first developed to let the decision maker describe the supply chain structure and its data, without affecting the underlying mathematical model. The latter is a mixed integer linear program which combines for the first time various features, either original or tackled separately in the literature. Knowing the refinery demands, it determines the activity levels in the network (amounts harvested, baled, transported, stored, etc.) and the required equipment, in order to minimize a total cost including harvesting costs, transport costs and storage costs. Numerical evaluations based on real data show that the proposed model can optimize large supply chains in reasonable running times.

摘要
将生物质（biomass）输送至生物炼制厂（biorefinery）的供应链在生物燃料（biofuel）生产中发挥着至关重要的作用。优化模型可助力决策者设计更高效的供应链，并最小化输送至炼制厂的生物质成本。
本文基于一项针对生物质物流的法国国家级研究项目，针对一家可加工多种作物及同一作物不同组分的炼制厂，以按日或周划分的一年期为研究时间跨度展开研究。首先构建了网络模型与数据模型，使决策者能够在不影响底层数学模型的前提下，描述供应链结构及其相关数据。
底层模型为混合整数线性规划模型（mixed integer linear program），首次整合了现有文献中单独提出或分别处理的各类原创特性与既有特性。在已知炼制厂需求的前提下，该模型可确定供应链各环节的作业水平（收获、打捆、运输、存储等作业量）及所需设备，以最小化包含收获成本、运输成本与存储成本在内的总成本。
基于真实数据开展的数值评估结果表明，所提模型可在合理的运行时长内优化大规模供应链。