A FRAMEWORK FOR THE FLEXIBLE GROUPING OF PRODUCTS FOR DISASSEMBLY

The increased scrutiny of a product's effects on the environment has led industry to embrace Environment Conscious Design and Manufacture (ECDM) practices. Disassembly of a product at the end of its useful life is an integral part of the ECDM paradigm. Disassembly facilitates the end-of-product life options such as recycling, remanufacture, reuse and safe disposal. As industries take active steps to move to a "greener" economy, the emergence of dedicated disassembly facilities is foreseeable. The increasing variety of new products requires that these disassembly factories be able to handle a wide spectrum of products. Today, dedicated disassembly facilities exist for disassembling high value products such as computers and automobiles. However, the flexibility of a disassembly facility is particularly important to accommodate the disassembly of moderate to low value, and high variety consumer products. We address this issue by proposing a framework for the formation of product groups based on their disassembly characteristics as a step towards increasing the flexibility of disassembly factories. A two-level hierarchy of generic disassembly characteristics has been employed to evaluate the similarity of diverse products. We have used a modified version of the "growing neural gas" neural network model to classify the products. The first level or primary Disassembly Product Groups (DPGs) are obtained based on similarities in product structure. They are further subdivided into secondary DPGs based on the disassembly resource requirements of the products. An implementation for the disassembly-oriented grouping using a few example products has been carried out for demonstration purposes.