Minerals Engineering, Vol.20, No.9, 875-890, 2007
The use of fuzzy rule models to link automotive design to recycling rate calculation
Product design selects materials from primary metal and material cycles and combines them into multi-material designs, in which the various materials (metals and non-metals) are complexly integrated (e.g. welded, glued, alloyed, layered). Since the quality of recyclates and the recycling rate of a product is largely affected by the design of the product it is required that tools are available that link computer aided design (CAD) software and fundamental recycling models as developed previously by the authors in order to predict recyclability of the car during the design phase. However, complex system models for the prediction of recycling rates of new cars are far too complex to be linked directly to CAD. Therefore fuzzy rule models have been developed to create the interface between the complex system models and the design tools for automobiles. The quality of recyclates and hence the recycling rate of the product is largely affected by the liberation of materials during shredding, which have been connected in the design. Therefore the prediction of liberation behaviour is of crucial importance to the calculation of recycling/recovery rates. A fuzzy rule based approach is developed to describe the liberation behaviour of materials from automobiles and hence the recycling rate, linking material combinations and connections in product design with the particle composition and degree of liberation after shredding. This approach permits a description of the liberation mechanism of materials during shredding as a function of design. The heuristic rules for this were developed during extensive sampling and gathering of data from scrap samples from various shredders. This realizes a model basis for Design for Recycling. The elegant simplicity of the developed fuzzy set models makes them easy to integrate into LCA tools. This ensures that such environmental models are provided with fundamental information on the end-of-life behaviour of products, which include (i) physics and thermodynamics of separation processes, (ii) the quality of recyclates as a function of physical design choices and (iii) recycling rate calculations on a statistical basis. This methodology is currently been applied by the automobile industry to ensure among others that recycling rates have the fundamental basis that challenges but also provides a solid legal basis for the calculated recycling rate of new automobiles. (c) 2007 Elsevier Ltd. All rights reserved.