An upgradable product is a product in which the valuable life is extended by exchanging or adding components. An upgradable product is both environmentally and economically advantageous compared with products requiring replacement because its functions can be improved by adding only a few components. Therefore, the design and sale of upgradable products represent effective methods for attaining a sustainable society. Previous studies of upgradable product design methods have assumed that products have a modular architecture, in which all components are functionally independent. However, actual products have both integral architectures and modular architectures. Achieving high-performance products through component optimization is easier with an integral architecture than with a modular architecture. However, the integral architecture makes it difficult to disassemble and replace individual components. It is difficult to achieve high levels of performance in products with modular architecture, but it is easy to disassemble and replace components. Therefore, upgradable product design must determine the most appropriate product architecture. Hence, this paper focuses on the product architecture of upgradable products and proposes a decision support method that yields the appropriate combination of product architecture and upgrade cycle. In addition, the authors propose evaluation models for the environmental load, cost, and customer dissatisfaction, as well as a comprehensive evaluation index based on these models. The overall model, which gives the evaluation index, considers the differences in the evaluated values resulting from differences in the product architecture and the number of upgrades. The proposed method was applied to a motherboard module design problem for a laptop computer. The results of this case study confirm that the proposed method successfully supports the designer during upgradable product design by deriving the most suitable combination from a set of product architectures and upgrade cycle candidates.
Service for parallel applications based on JINR cloud and HybriLIT resources