Abstract: Addressing the optimal design of sustainable and personalized supply chain networks for new energy vehicle(NEV)parts under the context of carbon neutrality, a robust optimization model based on a sustainable and personalized closed-loop supply chain network is established to tackle uncertainties within the supply chain network. Firstly, uncertain parameters are described using uncertainty sets, and a mixed-integer programming model is formulated for the NEV parts supply chain network structure, aiming to minimize transportation costs, processing costs, operational costs, and carbon emission costs. Additionally, a personalized evaluation mechanism for supply chain products is proposed based on the hypergeometric distribution method. Secondly, the execution strategy parameters of the original wild dog algorithm(WDA)are dynamically adjusted, and the population group attack strategy is integrated with the Lévy flight strategy to enhance the algorithm's global optimization accuracy and convergence speed. Finally, numerical and simulation analysis conducted in MATLAB demonstrates that the proposed robust optimization model, combined with the improved WDA, exhibits significant advantages in solving the design problem of closed-loop supply chain networks for NEV parts.

Key words: carbon neutral, personalization, improved wild dog algorithm, robust optimization