Co-authors​ :

 Shravan TORVI (ITALY), Marco DIANI (ITALY), Domenico ROTONDI (ITALY), Pigliaru PIETRO (ITALY), Marcello COLLEDANI  

Composites, and especially Fiber Reinforced Plastics (FRPs) are being increasingly used in several applications in recent years due to their excellent properties like low density, high tensile strengths, and resistance to corrosion. They find use in strategically important sectors like wind energy, automotive and aerospace, while they pose a problem at the end of their service life. Extending the life cycle of these End-of-Life (EoL) components by reuse, repair and recycling strategies has been a topic of discussion over the last 15-20 years. One of the hurdles in adopting a circular economy approach in the sector of composites is the information asymmetry and relatively lesser interactive players across the value chain. To that end, this work will present a data-driven and demand-driven Decision Support System (DSS) that aims to bridge the information gap between waste owners, recyclers, and end-users. The DSS is based on a cross-sectorial approach which facilitates an open-recycling approach to composites, in which EoL components from a sector such as wind energy can be used in a less demanding, but high-added value, application such as automotive, sporting goods or construction. The architecture of the DSS envisages a 2-level decision making. In the first level the product is evaluated in terms of EoL parameters like its service life and present defects. These parameters can be descriptive in nature such as the chemical composition of the EoL but can also be functional in nature, in which case their values determine certain decision-making criteria such as the dimension of a defect. The output of the DSS at this stage is the choice of the appropriate strategy for the EoL between Reuse/Repair/Recycle. The second level decision involves the choice of the appropriate recycling technology, strongly based on the demand-driven approach, since the it is chosen based on the end-user requirements which include not only mechanical requirements, but also capture economic feasibility. Due to the variability in the data that need to be treated, a Multi-Criteria-Decision-Making (MCDM) logic has been adopted. For each decision level, specific evaluation criteria have been identified. Usually, it is unlikely that all criteria have the same relevance. Therefore, criteria have assigned weights to take into account their relevance. To this end the Best-Worst Method (BWM) is adopted, which is a pairwise comparison technique that works on the principle of the choice of the most relevant and least relevant criteria for a specific decision. To rank the alternatives at each step, a Technical Order of Preference by Similarity to Ideal Solution (TOPSIS) is adopted, leading to a client-server web application.