Sustainable Value Recovery of NdFeB Magnets: A Multi-Objective Network Design and Genetic Algorithm

• Published in: ACS sustainable chemistry & engineering Vol. 6; no. 4; pp. 4767 - 4775
• Main Authors: Jin, Hongyue, Song, Byung Duk, Yih, Yuehwern, Sutherland, John W
• Format: Journal Article
• Language: English
• Published: American Chemical Society 02.04.2018
• Subjects: algorithms; business enterprises; clean energy; ecosystem services; electric vehicles; infrastructure; magnetic materials; manufacturing; mathematical models; rare earth elements; recycling; risk; social support; supply chain; transportation; uncertainty; wind turbines; Rare earth magnets; Supply chain optimization; Recycling; Rare earth elements; Genetic algorithm
• ISSN: 2168-0485; 2168-0485
• DOI: 10.1021/acssuschemeng.7b03933

Neodymium–iron–boron (NdFeB) magnets are widely used in clean energy applications such as wind turbines and electric vehicles whose demand is escalating. However, rare earth elements (REEs) for manufacturing NdFeB magnets are subject to significant supply uncertainty due to Chinese near-monopolistic supply. To mitigate the risk, companies are actively pursuing value recovery from end-of-life magnets. However, the questions of how to collect used magnets and smoothly transfer them through the reverse supply chain require further investigation. To address this challenge, this paper designs an efficient NdFeB magnet recovery infrastructure by identifying the optimal processing facility locations and defining the capacities and transportation flows that maximize the economic and environmental benefits and social support for the new business. Mathematical model and a multi-objective network design genetic algorithm (MONDGA) were designed to calculate solutions. When compared with the exact solutions implemented by CPLEX (an optimization package), MONDGA provided (near) optimal solutions with significantly improved computation efficiencies. As the real world model application required a large-scale optimization, MONDGA was superior to CPLEX, which failed to provide any solution. The results confirmed that our proposed model and algorithm offer a promising strategy for the NdFeB magnet recycling industry to enhance the economic and environmental sustainability and maximize social support.