Hierarchical joint optimization of modular product family and supply chain architectures considering sustainability

Abstract

Since sustainability is a growing concern, businesses aim to integrate sustainability principles and practices into product and supply chain architecture (SCA) design. Modular product architecture (MPA) is essential for meeting sustainability demands, as it defines shareable and detachable modules to facilitate assembly/disassembly and recovery. However, Traditional architectural modularity overlooks interface complexity and fails to balance commonality, complexity, and cost for effective facility sharing and ensuring ease of disassembly and upgradeability. Therefore, it is inefficient for economic design, modular recycling, and assembly/disassembly operations. Furthermore, the existing modular clustering methods cannot characterize components that may generate complex modular structures and decrease compatibility for remanufacturing and upgrading. Therefore, such methods cannot offer guidance on modular granularity to reduce premature product discarding through economic upgrading. This paper outlines a modular PFA and SCA by bi-level programming prioritizing sustainability. It emphasizes reducing product discarding and waste generation by focusing on lower remanufacturing costs and higher upgradeability in modular products and SCA. Here, the similarity of operation, ease of disassembly, and upgradability have been used for SCA with architectural and interface modularity. A fuzzy inference system (FIS) differentiates each product component and provides a basis for remanufacturing and upgrading compatibility. The developed leader-follower decision structure allows the selection of suitable and sustainable module candidates to share among product families, considering product architecture, processing activities, and upgrading opportunities. The designed case study on refrigerator PFA shows that the proposed approach saves 23.93 %, 7.21 %, and 17.2 % of SCA costs compared to non-modular, random clustering, and sequential approaches. Further numerical analysis evaluates the proposed model's economic, social, and environmental implications. Accordingly, the model is suitable for guiding modularization for financial gains while promoting environmental and social sustainability.