Pricing Strategy of Remanufacturing Supply Chain Based on Stackelberg Game and Intelligent Optimization

This study addresses the dynamic pricing problem in remanufacturing supply chains by proposing an innovative approach that integrates Stackelberg game theory with intelligent optimization. A bi-level programming model incorporating reference price effects is developed, and a hybrid optimization algorithm is designed: the upper level employs an improved genetic algorithm to optimize manufacturer decisions, while the lower level combines KKT conditions to analytically solve the remanufacturer's response. Additionally, an LSTM network is introduced to predict reference prices. Experimental results demonstrate that this method improves profits by 12.7% compared to traditional models and maintains stability under price fluctuations. The study identifies an optimal interval for the cross-elasticity coefficient and reveals that the memory strength of reference prices significantly impacts prediction accuracy, providing quantitative guidance for enterprise pricing decisions. The research recommends establishing dynamic pricing mechanisms, optimizing cost structures, and building intelligent decision-making systems to enhance supply chain coordination efficiency. This study offers effective decision-making support for remanufacturing enterprises to cope with market uncertainties.